Trimble Inc. (NASDAQ: TRMB) — Cheap on the Numbers the Auditor Won’t Vouch For

Report date: 2026-07-25 · Sector: Information Technology — Application Software (Construction / Geospatial / Field Systems) · SEC CIK 0000864749 · Fiscal year is a 52/53-week year ending the Friday nearest 31 December; FY2025 ended 2026-01-02. Figures reconcile to the FY2025 Form 10-K (filed 2026-02-25), the Q1-2026 Form 10-Q (filed 2026-05-06), the 2026 DEF 14A (filed 2026-04-09), the Q4-2025 and Q1-2026 earnings calls, and the trailing 60-month SEC filing corpus unless otherwise noted.

The analysis in Sections 1–15 below carries no recommendation and no price target. The single exception is the clearly-labelled Claude's Take block immediately below, which is the author’s own subjective opinion.

⚡ Claude’s Take

This block is the author’s own subjective opinion. It is general information, not investment advice, and not a recommendation to buy or sell any security. The analysis in Sections 1–15 that follows takes no position and names no price target.

Verdict: AVOID here at ~$52.91. HOLD-if-owned; not a short. Conviction: medium-high. Directional zone: I would only become interested in the high-$30s to mid-$40s — roughly $38–44/share, which triangulates from three independent anchors (~9.5x FY26E adjusted EBITDA of $1,151M, ~12x FY26E non-GAAP EPS of $3.55, and a 7.5–8% owner-free-cash-flow yield on ~$680M). And I would want one thing that money cannot buy alongside the price: a clean internal-control opinion. Tag: “The bull case is built entirely out of the numbers the auditor just refused to sign off on.”

Trimble looks cheap and, on management’s presentation, is cheap: ~14.9x forward non-GAAP EPS, ~11.2x forward EV/EBITDA, a ~7.0% headline free-cash-flow yield, 69–71% gross margins, 79% software/services/recurring revenue, 1.1x leverage, and a buyback that has retired ~6% of the shares in eighteen months. The sell side is nearly unanimously positive and the stock sits at its 52-week low. The problem is what sits underneath. On 2026-02-25 KPMG issued an adverse opinion on the effectiveness of Trimble’s internal control over financial reporting, and management concluded that material weaknesses — specifically in “revenue and related accounts, income taxes, and other financial reporting processes” — “exist as of January 2, 2026.” That is the third consecutive year unremediated, it was still unremediated in the Q1-2026 10-Q, and it is the exact process that produces ARR and the segment margins the entire thesis rests on. Meanwhile the economics do not support the platform framing: ROIC is 6.92% and lower than it was in 2021 (9.24%), tangible common equity is negative $327.6M against $5.24B of goodwill, and the FY2025 margin expansion is — in the 10-K’s own words, three separate times — attributable to “the loss of lower margin divestiture income,” while the one pure-software segment that grew 10% organically (AECO) posted a margin of 34.2% against 34.1%: flat. Three different FY2025 growth rates are in circulation — −3% as reported, +6% organic in the 10-K, +10% on the call after excluding Ag, Mobility, a 53rd week and January-1 licence renewals.

The framing is falling knife with an unresolved company-specific cause — not an abandoned quality compounder, and that is an evidentiary claim, not a rhetorical one. Trimble fell against its own sector twice: −12.8% over 14–20 January while the IGV software ETF fell 5.1%, and −19.5% from 5–18 May, into a software tape that ROSE 5.2% (PTC +4.7%) — on a beat-and-raise. Factor data agree: −0.29 Sharpe and −7.6% annualized over five years, negative Momentum (−0.29 Base), negative Quality (−0.03), no Value loading in any model, and factor neighbours that are mid-cap index ETFs rather than software peers. And within the same AI-fear de-rating, Trimble has fallen less than its betters: own-history composite valuation percentiles are BSY 1.3rd, PTC 2.0th, ADSK 3.0rd — and TRMB 18.4th, with ADSK earning ~23% ROIC on 91% gross margins to Trimble’s 6.9% on 69%. The cheapest-ever P/B (1.3rd percentile) is a trap: book value was inflated ~$1.3B by the 2024 Ag gain and is 106% intangibles. Finally, the tell: not one insider has bought a share on the open market in eighteen months and a 40% decline; the CEO sold ~$18.0M and directly holds ~11,897 shares. Autodesk’s CEO, CFO and a director bought ~$1.8M code-P at their lows. That is the difference between fear and a problem.

Conviction: medium-high — the bear case here rests on filed documents rather than forecasts, which is why it is not merely “medium”; what holds it back from high is that ~$680M of owner-FCF, 1.1x leverage and a real #1 machine-control franchise make a genuine floor, and the Ag divestiture proved this management can sell well. Bullish flip (one thing): a clean, unqualified ICFR opinion for FY2026 together with AECO segment operating margin expanding on organic growth — proving operating leverage rather than divestiture arithmetic. Insider open-market buying would accelerate it. Bearish flip: a fourth consecutive adverse opinion or any restatement, organic ARR growth decelerating through ~10%, or the 2027 “3-4-30” targets being retired without a credible 2028 framework.

📈 Stock Price Action — Five-Year Event Map

Factual price history — no recommendation and no price target. Prices are split- and dividend-adjusted closes from the AZI five-year CSV (9,069 rows from 1990-07-20; 1,271 rows in the trailing five years). The price move is FACT; the attributed cause is INTERPRETATION.

Arc. Trimble has round-tripped a full cycle and then broken down. From a ZIRP/software-mania peak close of $95.72 (30-Aug-2021) it fell 57% to a five-year low of $40.82 (13-Nov-2023), recovered +109% to a 52-week high of $85.24 (25-Jul-2025), and has since de-rated 37.9% to $52.91 (24-Jul-2026) — printing a 52-week low close of $48.87 on 17-Jun-2026. The 52-week range is roughly $49–$85; the stock trades below its 50-day EMA ($54.26) and well below its 200-day EMA ($64.08), with a beta of 1.26 and alpha of −0.23. Over the full five years the shares are −44.7% from the high and have delivered a negative annualized return of −7.6% with a −0.29 Sharpe ratio.

Cycle narrative. (1–2) Trimble peaked with the 2021 software bubble and de-rated for two years; the November-2023 low was punctuated by a 13% one-day drop on the Q3-2023 print, and the trough coincided with peak leverage after the $2.09B Transporeon acquisition (net debt $2,836.8M, net-debt/equity 63%). (3) The recovery was a portfolio event, not an operating one: the Ag business went into the AGCO PTx Trimble JV for $1.9B cash plus a retained 15% stake, ~half of which repaid debt, and the shares rose 17.9% in a single session on 6-Nov-2024. (4) April 2025’s round trip was the market-wide tariff shock. (5) The run to $85.24 in July 2025 was the Connect & Scale narrative working — Mobility divested, gross margin through 70%, record ARR. (6–7) The January 2026 leg is the first genuinely unexplained move: TRMB fell 12.8% over five sessions while IGV fell 5.1%, ADSK 2.8% and PTC 3.3%; a 1.26 beta on the sector move accounts for roughly 6pp of it and no 8-K, guidance change or company disclosure accompanies the residual (the only 8-K in the window is the 4-Dec-2025 revolver). (8) The May 2026 leg is the decisive one and it is unambiguously company-specific: on 6-May Trimble beat Q1 revenue by ~$35M, raised full-year revenue guidance by $15M and raised EPS guidance — and the stock fell 7.2% that day and 19.5% over eight sessions into a software tape that rose 5.2%, with PTC up 4.7%. Analysts cut forecasts on “upbeat” results. What the call actually contained: a $35M beat converted into only a $15M raise, the top end of EBITDA guidance trimmed 20bps, a Field Systems full-year organic guide of “low-to-mid single-digit” against a Q1 print of +12%, and six of roughly ten analysts asking, in different words, whether large language models will compress seat-based licensing. (9) The bleed continued to a 17-June low of $48.87 before a modest recovery.

1. Executive Summary

Trimble sells the technology that positions, measures, designs, manages and bills physical work — construction sites, survey and mapping, and freight networks. FY2025 revenue was $3,587.3M, down 3% as reported (+6% organic), at a 69.1% GAAP gross margin and a 16.5% GAAP operating margin ($592.0M), generating reported free cash flow of $360.9M that normalizes to roughly $668M once the one-time $277.4M final tax payment on the 2024 Ag divestiture (plus ~$30M of related costs) is added back. The business is now 79% software, services and recurring revenue, with annualized recurring revenue of $2,392.3M at FY2025 year-end and $2,435M at Q1-2026, and it runs at 1.1x leverage against a 2.5x target. Three segments: AECO ($1,498.6M, 42% of revenue, 34.2% segment margin), Field Systems ($1,539.5M, 43%, 31.1%) and Transportation & Logistics ($549.2M, 15%, 21.9%).

The moat is real, narrow, and in the wrong place for the story being told. The strongest genuine barrier to entry is in Field Systems, and it is a Greenwald supply/cost-plus-distribution advantage rather than a software network effect: Trimble machine-control and positioning technology was displayed in the booths of 24 leading construction OEMs at CONEXPO in February 2026, sold through a global independent-dealer network, and built on a grade-control engine extended to new machine classes. Displacing that requires re-tooling an OEM, not switching a licence. AECO has genuine customer captivity but competes against Autodesk (~91% gross margin, ~23% ROIC) and Bentley (81.5%, 11.6% ROIC) from a weaker margin structure. T&L’s Transporeon network is the most defensible network-effects claim in the portfolio and the smallest and least profitable segment.

The moat does not surface in the return on capital, which is the test that matters. ROIC is 6.92% — below any defensible cost of capital for a 1.26-beta equity, and below the 9.24% earned in 2021. Tangible common equity is negative $327.6M: book value is 106% intangibles, sitting on $5,239.7M of goodwill accumulated through 13 acquisitions and 25 divestitures since 2020 and roughly $4.9B of acquisition spend since 2018. The competitive position was substantially purchased, and the price paid consumed the return.

Three findings define the investment case. First, the reported numbers carry an adverse control opinion. KPMG issued an adverse opinion on ICFR dated 2026-02-25; management states the material weaknesses in ITGCs and in “revenue and related accounts, income taxes, and other financial reporting processes” “exist as of January 2, 2026”; disclosure controls were again “not effective” in the Q1-2026 10-Q. This is the third consecutive unremediated year, following a 2024 episode in which EY — auditor since 1986, replaced by KPMG in 2025 — discovered the weakness while preparing for a PCAOB inspection, the FY2023 ICFR opinion was withdrawn, three 10-Qs were filed late in a single batch, Nasdaq issued listing-rule notices and Form S-3 eligibility was lost until April 2026. Second, the margin expansion is subtraction, not leverage: FY2025 non-GAAP operating margin rose 210bps to 27.5%, but the 10-K attributes the two expanding segments’ gains to “the loss of lower margin divestiture income,” and AECO — 42% of revenue, +10% organic — delivered a flat 34.2% margin. You can only prune a portfolio once. Third, the presentation gap is wide and systematic: non-GAAP diluted EPS of $3.13 is 78% above GAAP’s $1.76 ($0.79 vs $0.42, or 88%, in Q1-2026), on $396.1M of add-backs of which $151.5M is stock-based compensation and $53.5M is recurring “restructuring”; and FY2025 growth was reported as −3%, +6% and +10% depending on how many layers of adjustment are applied.

Valuation is fair-to-slightly-cheap, not compelling, and it is less advanced than the sector’s. At $52.91 the equity is ~$11.74B and EV ~$12.92B: 3.50x TTM revenue, 5.05x TTM gross profit, 15.1x TTM EBITDA, 19.7x TTM EBIT, 27.7x GAAP TTM earnings; on FY2026 guidance (revenue $3.875B, +8%; non-GAAP EPS $3.55; 29.7% EBITDA margin; ARR +13%) it is 11.2x EV/EBITDA and 14.9x earnings for a headline ~7.0% FCF yield that becomes a 5.8% owner-FCF yield once ~$145M of after-tax SBC is deducted. The critical relative fact: measured against their own multi-year histories, Bentley sits at the 1.3rd percentile, PTC the 2.0th and Autodesk the 3.0rd — Trimble the 18.4th. The AI-fear de-rating in design and construction software is real and sector-wide; Trimble has participated in it less than every higher-quality incumbent. The cross-sectional P/S discount (3.43x vs 4.74x/5.98x/6.66x) is approximately proportional to the quality discount rather than additive to it.

This article concludes that Trimble is a genuinely improved business in a structurally decent industry, with a real but narrow and mis-capitalised advantage, growth of adequate-to-good quality, below-average capital allocation, and a price that is defensible only if one accepts financial statements whose controls the auditor has declined to endorse. No recommendation and no price target appear below; valuation is discussed only as embedded expectations.

2. Business Overview

What Trimble sells. Trimble converts physical position into digital information and back again. Its products measure where things are to centimetre precision (GNSS receivers, robotic total stations, laser scanners), steer construction machines automatically along a digital design (machine control and grade control), author and coordinate the designs themselves (SketchUp, Tekla, civil design), run the money and paperwork of a construction business (estimating, project management, construction ERP), and match freight to trucks across Europe and North America (Transporeon, MAPS). The company’s framing for this is Connect & Scale: “connecting work in the office and the field, connecting our hardware and software, connecting the physical and digital worlds,” with Trimble Connect as “the orchestration layer” and a claimed single source of truth for the digital twin.

Three reportable segments (descriptions from the Q1-2026 10-Q, Note 7; figures FY2025 from the 10-K MD&A):

AECO serves architects, engineers, contractors and asset owners — architectural and interior design, structural and civil engineering, building and infrastructure construction, and asset operations and maintenance. Its commercial vehicle is Trimble Construction One, a bundled offering extended into Europe (ProjectSight) and, in Q1-2026, Asia-Pacific. It is the highest-margin and fastest-growing segment, with “26 consecutive quarters of mid-teens or higher organic ARR growth” and ARR of $1,475M at FY2025 year-end (+16% organic), reaching $1.51B in Q1-2026 (+14%).

Field Systems is the historical Trimble — surveying and mapping, civil construction machine control, building-construction field services, and positioning services. It is the hardware-heavy segment, though management notes it “reached an important milestone, with software and services driving over 50% of revenue” in FY2025, alongside 20% organic ARR growth. It is also the segment with the least revenue visibility: “this is the one that has — we have the least visibility with the hardware, particularly.”

T&L provides shipper/carrier/intermediary software: Transporeon (the European freight-procurement and execution network, acquired for ~$2.09B in 2023), MAPS, enterprise transportation management and forestry. Transporeon “transacts well over $100 million of revenue through tens of millions of annual transactions,” which is the company’s only genuinely consumption-based revenue at scale.

How it makes money, and the mix shift. Revenue is reported as product versus subscription and services. FY2025: product revenue fell 12% as reported and 1% organically; subscription and services rose 2% reported and 9% organically. Software, services and recurring revenue reached 79% of FY2025 revenue (78% in Q1-2026, as hardware outperformed). Deferred revenue was $894.0M current plus $104.7M non-current at FY2025 year-end.

Understanding ARR, precisely. The metric on which management, the proxy and the sell side all anchor is defined in the 10-K as follows: ARR “is calculated by taking our subscription and maintenance and support revenue for the current quarter and adding the portion of the contract value of all our term licenses attributable to the current quarter, then dividing that sum by the number of days in the quarter and then multiplying that quotient by 365.” Two consequences follow, and both matter. (a) It is a single quarter annualized, so a strong quarter mechanically inflates it and a seasonally-heavy quarter distorts it — which is precisely why AECO’s Q4-2025 segment margin of 44% “was aided by January 1 term license renewals” and why the FY2025 comparison required an explicit adjustment for those renewals. (b) It includes term-licence contract value, which under ASC 606 is recognised substantially upfront rather than ratably; ARR is therefore not a contracted-backlog measure. Total remaining performance obligations were approximately $2.0B at Q1-2026 (“we expect to recognize $1.4 billion, approximately 70%… during the next 12 months”) against reported ARR of $2,435M. Part of that gap is the ASC 606 practical expedient permitting exclusion of contracts of a year or less, so the comparison is indicative rather than dispositive — but it does establish that ARR should not be read as committed revenue.

Customers and geography. The base is diversified: management states each business pillar carries more than $230M of ARR, the mix is “centric to capturing the North American opportunity” with Europe growing faster than North America in Q1-2026 within AECO, and there is “a balanced mix between direct, distribution, and OEM customers as well as enterprise-level customer relationships.” No single-customer concentration is disclosed as material. What is not disclosed anywhere in the FY2025 10-K, the Q1-2026 10-Q, the 2026 proxy or either recent earnings call is any measure of net revenue retention, gross retention, churn or seat count. For a company that asks to be valued on ARR, and whose revenue-process controls carry an adverse opinion, that is a material gap.

Verdict. A genuinely repositioned company: from a hardware-led conglomerate spanning agriculture, telematics, construction and survey, to a three-segment, 79%-recurring, 69–71%-gross-margin software-and-systems business. The transformation is real and visible in the mix. The open question that the rest of this article pursues is whether the transformation created value for the owner or merely re-labelled the revenue.

3. Industry Dynamics

Market structure differs sharply by segment, and averaging them is the most common analytical error made on this name.

Construction and infrastructure technology (AECO + Field Systems, 85% of revenue). The end market is construction spending — cyclical, fragmented on the demand side, and structurally under-digitised. Management’s own characterisation of the customer is the key to the industry’s economics: “construction is a relatively low-margin industry, yet remains one of the most risk-exposed industries in the world. More than 80% of projects exceed budget. And when disputes arise, the average claim in North America tops $60 million.” A low-margin, high-risk, litigious customer base is a good customer for software that demonstrably reduces risk and rework — the ROI case writes itself — and a bad customer for anything discretionary. It also means pricing power is tied to demonstrable productivity, not to brand.

Two structural tailwinds are genuine. Labour scarcity: “There’s a labor shortage. We’re hundreds of thousands of workers short in North America alone.” Automation that substitutes for scarce labour sells itself in that environment, and it is the strongest single argument against the AI-seat-compression bear case (see Variant Perception). Infrastructure and data-centre construction: Field Systems’ Q1-2026 outperformance was attributed to “end market strength in infrastructure and data centers.”

Competitive intensity in AECO is high and rising, and Trimble is the weakest-capitalised of the three incumbents on margin structure. The comparison is unflattering and is drawn from public filings plus third-party computed ratios:

Autodesk’s AECO segment alone generated $3,583M of revenue, +22% in FY2026 — roughly 2.4x Trimble’s entire AECO segment, growing more than twice as fast — anchored on the Revit .rvt format and reinforced by government BIM mandates. Bentley owns the infrastructure-engineering niche at 81.5% gross margins. Trimble’s differentiation in AECO is genuine but is breadth across the workflow (design → estimating → project management → ERP → field execution) rather than depth of format lock-in at any single point.

Field Systems is the structurally attractive part of the portfolio, and it is not a software market. The competitive set is Topcon, Hexagon/Leica, and the OEM-embedded offerings of Caterpillar, Deere and Komatsu. Barriers here are physical and commercial: factory-fit relationships with construction OEMs (Trimble appeared in 24 OEM booths at CONEXPO’s 140,000-attendee show), a global independent-dealer network that services and calibrates precision instruments, and a proprietary grade-control engine extended to compact track loaders, dynamic-swing-boom excavators and ground-penetrating-radar-integrated asphalt compaction. This is a genuine oligopoly with high entry costs and slow share migration. It is also cyclical, hardware-inventory-exposed, tariff-exposed and — on management’s own admission — the part of the business with the least forward visibility.

T&L is the weakest industry position. Freight software is crowded (project44, Samsara, Motive, Platform Science — in which Trimble now holds up to 32.5% after selling it the Mobility business), the end market is cyclical and currently weak (“the macro environment remains challenged,” “a constrained freight market,” with North America “beginning to show some signs of market recovery”), and the segment carries the lowest margin in the portfolio at 21.9%. Transporeon’s European network density is a real asset; the segment as a whole is not.

Capital-cycle placement (Marathon lens). The AEC and vertical-software complex spent a decade attracting capital because reported returns were high; Trimble responded as the theory predicts, with 13 acquisitions since 2020 and roughly $4.9B of acquisition spend since 2018. The theory’s other prediction has also come true: goodwill grew from $3,540.0M (2018) to $5,239.7M (2025), +48%, while ROIC fell from 9.24% (2021) to 6.92% (2025). This is Marathon’s asset-growth anomaly in a single company. What is not yet visible is the supply-side reckoning — capital is now leaving the sector (all four incumbents are at or near decade-low own-history valuations), which historically precedes better returns for whoever survives with the strongest position. That argues for patience rather than for purchase at the current price.

Regulation. Comparatively light. BIM mandates in the UK, EU and parts of Asia are a demand tailwind that accrues disproportionately to format owners (Autodesk) rather than to workflow breadth. Trimble is pursuing FedRAMP certification — an investment noted as a drag on Q1-2026 Field Systems margin — which is a genuine incremental moat for US public-sector work if achieved. Tariff policy is a live and named uncertainty for the hardware business.

Verdict: structurally mixed, tilting good — but Trimble sits in the weakest competitive slot of the good part. Field Systems occupies a genuinely attractive OEM/dealer oligopoly. AECO is a good industry in which Trimble is the third-best-positioned incumbent by every economic measure. T&L is a poor industry. On a revenue-weighted basis this is an above-average industry mix and a below-average position within it.

4. Competitive Position

Management’s moat claim, stated in full. From the Q1-2026 call: “Integrated workflows with deep domain knowledge at scale is a differentiator and a moat. Our ecosystem of third-party connectivity and platform extensibility compounds value delivery and drives network effects where every connection point and transaction improves the next autonomous decision, making the data not peripheral to the product, but the product itself.” The supporting scale data are substantial and, taken at face value, impressive: more than 30 million projects created in Trimble Connect; over 50 million users since inception; over 130 marketplace extensions; thousands of third-party integrations; “trillions of dollars of construction run through Trimble today, tens of billions of freight… hundreds of thousands of instruments.”

Pressure-testing the network-effect claim against management’s own data. Trimble’s Q4-2025 investor slide 8 discloses the following cross-sell KPIs: “only 20% of our customers currently buy more than one product”; customers with more than three products grew 18%; “more than 70% of our ACV bookings came from cross-sell and upsell motions”; and projects in Trimble Connect grew 18%.

That first figure is the single most important disconfirming datapoint in this file. Connect & Scale has been the stated strategy for approximately five years. After five years, four out of five Trimble customers buy exactly one product. Management presents the 20% as headroom, and as a forward-looking statement about opportunity it is fair. But as evidence about a present moat it points the other way: a genuine ecosystem network effect — where each additional connection point raises the switching cost of the whole — would show up as a rising multi-product attach rate across the base, not as a 20% penetration figure disclosed as a growth opportunity. The 70%-of-ACV-from-cross-sell-and-upsell figure is not the counterweight it appears to be, because “upsell” to existing single-product customers is included in it and it measures the bookings mix, not the installed-base attach rate.

Naming the moat properly (Greenwald taxonomy), by segment.

• Field Systems — a genuine supply/cost-plus-distribution advantage. This is the real moat. Its mechanism is physical and commercial, not digital: factory-fit integration across 24 construction OEMs, a global independent-dealer network that sells, services and calibrates precision instruments, and a grade-control engine whose extensibility (new machine categories, GPR integration, dynamic swing booms) is expensive to replicate because it requires OEM engineering partnership rather than software. Painter’s own framing is instructive and, unusually, understated rather than promotional: “Trimble has been an autonomy company for decades. We happen to call it machine control and guidance. It’s between Level 2 and 3 autonomy today.” Displacing an embedded machine-control standard means re-tooling an excavator line, not cancelling a subscription. This advantage has an economies-of-scale character too: the R&D and OEM-integration cost is fixed and Trimble amortises it over the largest installed base.
• AECO — real but partial customer captivity. Construction ERP, estimating and project-management systems are genuinely sticky: they hold the historical cost data, the job-cost structures and the integrations that a contractor’s finance function depends on. Trimble Connect adds a data-gravity layer. The problem is that captivity is unquantified — no NRR, churn or seat disclosure exists — and it is shared with two incumbents whose margin structures (91% and 81.5% gross margin) indicate stronger positions at the format and authoring layer. The Greenwald market-share-stability test, the cleanest empirical moat test available, cannot be run on Trimble because the company discloses no share, seat or customer-count series.
• T&L — the most defensible network-effects claim in the portfolio, in the smallest segment. Transporeon is a genuine two-sided freight network: FY2025 added “over 10,000 carriers and over 100 shippers,” and the company reports double-digit growth in Transporeon customers exceeding €1M of ARR. Network density in European freight procurement is a real barrier. It is 15% of revenue at a 21.9% margin.
• SketchUp — brand and installed base, and the most AI-exposed asset. Management is candid that “the bulk of the user count” is consumer/maker rather than professional, and monetisation happens “at the professional grade level.”

The decisive test: does the advantage appear in the financial outcomes? This is where the analysis turns, and the answer is largely no.

*FY2024 ROE of 48.6% is an artifact of the $1,687.9M pre-tax Ag divestiture gain, not an operating result.

Gross margin has risen 1,330bps in five years — that is the mix shift, and it is real. ROIC has fallen over the same period, from 9.24% at the 2021 peak to 6.92%, with a 4.71% trough in 2024. Tangible common equity has been negative in every year of the series. The standard applied here is simple: if a moat cannot be tied to a financial outcome that would deteriorate without it, it is not a moat. Trimble’s advantages plainly do produce a superior gross margin relative to a pure hardware company. What they do not produce is a superior return on the capital assembled around them — because approximately $4.9B of acquisition spend created $5,239.7M of goodwill, and the denominator swallowed the numerator.

Direct comparison against the key competitor. Autodesk earns roughly 23% on invested capital at 91% gross margins with positive tangible equity, and trades at 5.98x sales. Trimble earns 6.9% at 69% gross margins with negative tangible equity, and trades at 3.43x sales. The 43% sales-multiple discount is, on these numbers, an approximately fair reflection of a two-thirds ROIC deficit — not a mispricing.

Verdict: a real but narrow, segment-specific advantage that has been mis-capitalised. Durable and genuinely valuable in machine control and positioning; real but contested and unquantifiable in AEC software; genuine but sub-scale in freight networking. Trimble is not a wide-moat compounder, the “network effects” framing is not supported by its own attach-rate data, and the clearest financial signature of a moat — a high and rising return on invested capital — is absent and has been deteriorating for four years.

5. Growth History and Forward Opportunities

The eight-year record is close to flat, and this is the fact the ARR narrative obscures.

FY2025 revenue is 1.9% below FY2021 and 5.6% below the FY2023 peak. The 2018→2025 revenue CAGR is +2.1%. GAAP diluted EPS in 2025 ($1.76) is below 2019 ($2.03). Over a period in which the company spent roughly $4.9B on acquisitions, the top line has essentially not grown.

The organic picture is better, and the presentation of it is the problem. Three different FY2025 growth rates circulate:

The bridge from −3% to +6% is disclosed and legitimate: acquisitions +1%, divestitures −10%. The bridge from +6% to +10% requires four further adjustment layers, stated in the Q4 call’s own preamble: “we will exclude the divested agriculture and mobility businesses as well as the 53rd week of fiscal 2024. For the fourth quarter, we will also adjust for the timing of January 1 term license renewals,” on top of “non-GAAP performance metrics on an organic basis.” The 10-K itself quantifies the calendar component: “the organic impact of the software renewals and additional week in 2024 was an approximate 2% negative impact on revenue growth for 2025.” Each individual adjustment is defensible. The cumulative effect is that the headline growth rate management delivers to investors is roughly 13 percentage points above the figure in the audited summary table.

The same pattern applies to ARR, the metric on which everything is anchored. The FY2025 10-K’s own table reads: ARR $2,392.3M vs $2,257.8M, $134.5M, +6%. The Q4 call and the proxy both say “+14% year over year on an organic basis.” For Q1-2026 the call said “ARR at $2.435 billion, up 13%” while the 10-Q says “growth of 12% year-over-year… organic ARR growth was 12%.” Both figures in each pair are defensible under their stated definitions; only one of each pair appears in a document reconciled inside a filing.

Growth composition is genuinely improving, which is the bull’s strongest ground. FY2025 organic product revenue was −1% (“lower demand in surveying and positioning products, partially offset by strong end-user demand for Civil Construction solutions”) while organic subscription and services revenue was +9% (“strong software term license and subscription growth across all segments, particularly in AECO”). AECO delivered +10% organic revenue and +16% organic ARR, capping 26 consecutive quarters of mid-teens or higher organic ARR growth — a genuinely impressive streak. Field Systems delivered +20% organic ARR growth as it converted hardware customers to recurring models, crossing 50% software-and-services revenue. This is real, high-quality mix improvement.

Q1-2026 was strong on any basis, and decelerating on the metric that matters. Revenue $939.9M, +11.8% as reported and +12% organic; AECO +16.6% reported revenue with a 420bps segment-margin expansion to 31.5%; Field Systems +13.9% reported revenue; non-GAAP EPS $0.79 against a $0.72 consensus. But T&L revenue declined 4.4% as reported ($139.6M vs $146.0M, +7% organic ex-Mobility), Field Systems’ segment margin fell 90bps to 28.8%, and organic ARR growth decelerated from 14% (Q4-2025) to 12% (Q1-2026) against a full-year guide of 13%.

Forward opportunities, ranked by evidential support.

1. Cross-sell into the 80% single-product base (best-supported). Management sizes a “$1 billion-plus cross-selling opportunity… in Construction.” With ARR at $2.4B this is a large and specific prize, and the mechanism — Trimble Construction One bundling, now extended to Europe and APAC — is demonstrably working where deployed (Europe grew faster than North America in Q1-2026). The evidential caveat is the flip side of Section 4: the attach rate is 20% after five years of trying.
2. Field Systems recurring-revenue conversion (well-supported, self-limiting). Converting hardware buyers to recurring models produced +20% organic ARR growth in FY2025 and crossed the 50%-software threshold. It also creates a reported-revenue headwind — Q1-2026 Field Systems revenue grew 12% “while absorbing headwinds due to the continued model conversions to recurring revenue.” This is high-quality growth that suppresses the headline; it is also finite.
3. AI as a TAM expander and new monetisation vector (asserted, largely unproven). Trimble launched SketchUp AI in Q4-2025 as an $11.99/month add-on bundling a fixed monthly credit allowance, and in April 2026 an integration letting Anthropic’s Claude generate SketchUp models via an MCP service, with a Trimble ID required to download. Management reports “the usage is growing and… almost all of those credits… are being consumed.” It also acquired Document Crunch (announced 2 April 2026) to create “a new AI-powered risk management category,” linking contract intelligence to estimating, project management and ERP workflows. The strategic logic is sound and specific. What is absent is any quantification — no attach rate, no incremental ARR, no revenue contribution.
4. Geographic expansion (modest, real). Trimble Construction One launched in APAC in Q1-2026; ProjectSight entered Europe; European AECO growth exceeded North America’s.
5. Consumption/hybrid monetisation (early, but not hypothetical). Transporeon’s “>$100 million of transactional revenue… tens of millions of annual transactions” plus autonomous procurement and autonomous quotation, which management says monetise “at a higher rate than the traditional non-AI capabilities.”

The forward guide implies material deceleration and the framework runs out next year. FY2026 guidance is revenue of $3.875B (+~8%), ARR growth of 13% and a 29.7% EBITDA margin; Q2-2026 is guided to $950M (+7.5%) with a 27.7% EBITDA margin (+30bps y/y — a notably thin expansion). Field Systems, which grew 12% in Q1, is guided to “low to mid-single-digit organic growth” for the full year; management attributes the implied slowdown to comps, “the Middle East, the uncertainty around the tariff policy” and hardware visibility, and denied pull-forward twice (“no pull forward in the quarter whatsoever”). More significantly, Barclays pressed the arithmetic directly: Trimble beat Q1 revenue by ~$35M and raised the full year by only $15M, while trimming the top end of EBITDA guidance by 20bps. Painter’s answer was a policy statement rather than an operating one: “by our own policy, I do not like to raise guidance after the first quarter to be 3 months into the year.”

And the multi-year framework is nearly exhausted. The Investor Day “3-4-30” targets — $3B ARR, $4B revenue, 30% EBITDA margins in 2027 — sit against a FY2026 guide of $3.875B revenue and 29.7% EBITDA margins. Revenue and margin will be within ~3% and ~30bps of the 2027 targets in 2026. Asked about 2028, Painter said: “We’re not yet ready to talk about 2028 or to set a frame on that. We haven’t yet even thought about when would be the next logical Investor Day to have.”

Verdict: growth quality is genuinely improving; growth quantity is modest and decelerating; and the presentation of it is systematically flattering. The mix shift from hardware to recurring software is authentic and is the best thing about this company. But eight-year revenue growth is 2.1% annualized, FY2025 revenue is below FY2021, organic ARR growth is decelerating from 14% to 12%, the largest segment is guided to low-to-mid single-digit organic growth, and the multi-year target framework investors have been underwriting will be substantially met next year with nothing published behind it. This is medium-quality growth described in high-quality language.

6. Financial Quality

The presentation gap is the defining feature of Trimble’s financial reporting, and it is wide, systematic and growing.

The bridge, from the 10-K’s own segment reconciliation:

Non-GAAP operating income adds back exactly the four non-corporate items: $172.0M + $151.5M + $19.1M + $53.5M = $396.1M. Two of those four are not defensible as non-economic. Stock-based compensation of $151.5M is 38% of the total add-back and is a real transfer of ownership — Trimble spent $863.4M on buybacks in FY2025, of which roughly 17% merely offset issuance. “Restructuring and other costs” of $53.5M (against $32.4M in FY2024) is not episodic — for a company that has completed 13 acquisitions and 25 divestitures since 2020, reorganisation is a permanent operating cost, and it is rising. Amortization of purchased intangibles ($172.0M) is the most defensible add-back on a cash basis, but it is also the accounting record of the acquisition spend that produced the low ROIC — excluding it makes the returns look better precisely where they are worst.

Segment margin is where the operating-leverage story falls apart. Consolidated non-GAAP operating margin rose 210bps in FY2025 (25.4% → 27.5%). The 10-K explains where it came from, in its own language, three separate times:

Field Systems: “operating income as a percentage of revenue was favorably impacted by the loss of lower margin Ag divestiture income.” T&L: “Operating income as a percentage of revenue increased primarily due to the loss of lower margin Mobility divestiture income.” Consolidated: “operating income as a percentage of revenue was favorably impacted by the loss of lower margin divestiture income.”

And the one segment that is pure software, that grew 10% organically, and that should therefore demonstrate genuine operating leverage — AECO — posted a segment margin of 34.2% against 34.1%. Flat. Field Systems (+230bps) and T&L (+220bps) supplied all of the expansion, and the filing attributes both to divestiture mix rather than to leverage. The FY2025 margin expansion is portfolio subtraction, not operating leverage. That matters enormously for the forward case, because you can only prune a portfolio once — 25 divestitures in six years have already removed most of the low-margin revenue, and the remaining mix is 79% software and services.

Q1-2026 offers a partial counterpoint the bull is entitled to: AECO’s margin expanded 420bps to 31.5% on +16.6% revenue, which is exactly the operating leverage that was missing in FY2025. But Field Systems’ margin fell 90bps to 28.8% in the same quarter (management cited CONEXPO trade-show timing and FedRAMP certification spend) against a guide of ~31% for the rest of the year — a required 220bps improvement — and AECO’s Q4-2025 margin of 44% was, on management’s own admission, “aided by January 1 term license renewals.” One quarter of AECO leverage in a seasonally distorted metric is not yet a trend.

Cash generation is genuinely good, and reported FY2025 free cash flow badly understates it. This is the most important normalization in the file and it favours the bull.

FY2025 CFO of $386.2M fell 27%, and the 10-K states the cause plainly: “The decrease in cash provided by operating activities was primarily driven by higher tax payments related to the Ag divestiture” — specifically “the remaining $277.4 million final tax payment for the Ag divestiture, which was made during the second quarter of 2025.” Income taxes payable fell from $325.0M to $17.7M. Management’s own framing on the Q4 call: “reported free cash flow remains strong at $361 million when considering the $307 million of tax payments and other costs primarily related to divestitures.” Normalized FY2025 free cash flow is therefore approximately $668M, on which the reported $360.9M is a ~46% understatement. Capex is trivial at $25.3M (0.7% of revenue) — this is a genuinely capital-light model at the operating level. A further real tailwind: OBBBA’s permanent repeal of domestic R&D capitalization delivers “cash tax reductions of approximately $53 million in 2025 and approximately $53 million in 2026.” Q1-2026 reported free cash flow was $275M.

The balance sheet is liquid, levered lightly, and hollow.

Liquidity and leverage are fine: 1.1x net-debt/EBITDA against a stated 2.5x target, no drawings on the new $1.25B five-year revolver entered 4 December 2025, $1,400.0M of principal debt outstanding. But tangible common equity is negative $327.6M at FY2025 year-end and has been negative in every year from 2018 to 2025 (ROIC.ai’s tangible-common-equity ratio: −107.9%, −54.3%, −35.5%, −20.8%, −22.2%, −71.1%, −6.9%, −10.4%). Book value per share of $23.80 is 106% intangibles. This is the direct consequence of building the business by acquisition, and it is why the “cheapest-ever price-to-book” observation (see the Valuation section) is a trap rather than a margin of safety.

One further balance-sheet item deserves attention. Equity investments of $616.8M are Level 3, marked on unobservable inputs. The Mobility business was sold to Platform Science on 8 February 2025 not for cash but “in exchange for equity ownership interests with a fair value of $253.9 million… based on unobservable inputs, including discounted cash flow projections, market comparables, and an option pricing model,” leaving Trimble owning or holding rights to acquire 32.5% of Platform Science. Trimble also retains 15% of PTx Trimble, the AGCO joint venture. Roughly 6.9% of total assets is therefore private-company equity carried at management-determined fair value — and equity-method results have already been volatile (a $48.1M loss in FY2024, $0.2M in FY2025).

Returns on capital are the indictment. ROIC: 8.54% (2020), 9.24% (2021), 7.74% (2022), 6.42% (2023), 4.71% (2024), 6.92% (2025). Return on capital 9.79%; ROE 11.87%; ROA 4.51%. Against a 1.26 beta and a mid-teens cost of equity, Trimble’s return on invested capital does not cover its cost of capital, and has not since at least 2022. The FY2024 ROE of 48.6% is an artifact of the Ag gain and should be disregarded. The economics do not improve with scale in the sense the Financial Quality verdict asks about: gross margin improves with mix, and the return on the capital base deteriorates with acquisition.

Quality-of-earnings items to carry into valuation.

• FY2024 GAAP diluted EPS of $6.09 contains a $1,687.9M pre-tax divestiture gain; the FY2024 effective tax rate of 25.0% is likewise gain-distorted (FY2025: 16.8%; Q1-2026: 24.7%).
• Cash-flow-to-net-income was 0.35x in FY2024 (gain in earnings, cash in investing) and 0.91x in FY2025 (tax paid) — neither is representative; a normalised ratio is roughly 1.5x.
• Working capital swung violently on the same tax item: accounts payable and accruals contributed +$327.8M in FY2024 and −$339.1M in FY2025.
• Accounts receivable fell from $856.0M (FY2025 year-end) to $617.5M (Q1-2026) — normal seasonality after a Q4-weighted quarter, but worth monitoring given the revenue-controls weakness.
• All of the above are produced by processes over which the auditor has issued an adverse control opinion (see Changes and Headwinds, below).

Verdict: good cash economics, poor capital returns, and an audit-quality overlay that should widen any discount rate applied to the reported numbers. The model genuinely converts revenue to cash — ~$668M normalized FY2025 free cash flow on $25.3M of capex is a real result, and the mix shift to 69–71% gross margins is authentic. But the return on the capital assembled to produce it is ~6.9% and falling, tangible equity is negative, the margin expansion is divestiture arithmetic rather than leverage, non-GAAP earnings run 78–88% above GAAP on add-backs that are 38% stock compensation and include a rising “restructuring” line, and the revenue and tax processes carry an unremediated material weakness. Economics do not improve with scale here; they improve with subtraction.

7. Capital Allocation

The record: 38 portfolio transactions in six years. Per the Q1-2026 10-Q, Trimble has completed 13 acquisitions and 25 divestitures since 2020 (the FY2025 10-K and the proxy say 23 divestitures; two more closed by Q1-2026, one explicitly “a small business in Field Systems”). Acquisition cash outflows since 2018:

Roughly $4.9B deployed on acquisitions since 2018 to build goodwill of $5,239.7M — against which FY2025 revenue is 15% higher than FY2018’s and ROIC is 6.92%. That is the capital-allocation verdict in one sentence, and it is unfavourable.

What went right. Two decisions deserve credit and should not be understated. The Ag divestiture was good capital allocation: a low-margin hardware business exchanged in Q2-2024 for $1.9B of cash plus a retained 15% stake in PTx Trimble, at a price the buyer set, with roughly half the proceeds used to repay debt (net debt fell from $2,836.8M to $651.8M) and a $122.0M tax payment. Balance-sheet discipline is genuine: leverage of 1.1x against a 2.5x target, a fresh undrawn $1.25B revolver, and no debt-funded acquisition since 2023. The Mobility exit — trading a telematics business for 32.5% of Platform Science — is defensible as a way to retain optionality in a business Trimble could not win alone, though taking illiquid Level 3 equity rather than cash transferred risk rather than removing it.

What went wrong. Transporeon is the transaction on which the ROIC verdict turns, and three years on it has not earned its price. Trimble paid ~$2.09B in 2023, funding it with debt that took net debt to $2,836.8M and net-debt-to-equity to 63%. T&L today is the smallest (15% of revenue), lowest-margin (21.9%) and slowest-organically-growing (+2% in FY2025) segment, and its reported revenue fell 30% in FY2025 and 4.4% in Q1-2026. Transporeon’s network is a genuine asset and its transactional revenue is strategically valuable as the company’s only real consumption-based model. But a $2.09B purchase that leaves the acquiring segment at a 21.9% margin and 2% organic growth is not a good use of shareholder capital, and it is the single largest contributor to the goodwill that suppresses ROIC.

Buybacks: large, accelerating, and programmatic rather than opportunistic.

The stated policy is “at least 1/3 of our free cash flow… for repurchasing shares.” The FY2025 programme was well-conceived in funding terms — it deployed the residual Ag proceeds — and poorly executed in price terms: it was spent across a year in which the shares traded roughly $62–$85, and a further $317M was spent in Q1-2026 at an average well above the current $52.91. $608M remains authorized. This is not counter-cyclical repurchasing; it is a programmatic return of a divestiture windfall, and the shares have since fallen 38%. Against roughly $150M of annual SBC, approximately 17% of the FY2025 buyback simply offset issuance. There is no dividend.

Incentives: the structural problem, stated precisely. From the 2026 DEF 14A (filed 2026-04-09):

There is no return-on-capital metric and no free-cash-flow metric anywhere in the executive incentive structure. A search of the 2026 proxy returns zero instances of free cash flow as a compensation measure. Management is paid on revenue, adjusted EBITDA and ARR — three metrics that are entirely indifferent to the price paid for acquired revenue — at a company that has spent ~$4.9B on acquisitions to accumulate $5.2B of goodwill and earns 6.9% on invested capital. The incentive system is structurally incapable of registering the one thing that has gone wrong. There is a second, sharper problem: ARR appears in both the annual bonus and the long-term plan, and ARR is generated by the revenue processes carrying an unremediated material weakness and an adverse ICFR opinion. A metric that determines pay and is produced by an inadequately-controlled process is a governance issue, not merely a design preference.

The 2025 outcome illustrates the mechanism: the corporate financial measures paid out at 138.6% of target, and the ±10% non-financial modifier “was not applied to the bonuses for any of the Named Executive Officers,” in a fiscal year in which reported revenue declined 3%, reported ARR grew 6%, ROIC was 6.9%, and the shares subsequently fell 38%. CEO Painter’s 2025 total compensation was $19,129,944 ($950,000 salary; $13,311,864 stock awards; $2,230,637 options; $2,633,443 non-equity incentive) — $21.6M of stock-award value at maximum PRSU achievement. Sawarynski (CFO) $4,453,662; Bisio $4,572,608; Schwartz $4,900,270; Keating $2,208,469. Say-on-Pay support was 93.5% in 2025. To management’s credit, PRSU TSR weighting was raised from 25% to 50% in 2024 “based largely on stockholder feedback,” and a new 300% maximum TSR payout was added in 2025 only for a top-of-index TSR outcome.

Insider behaviour: the cleanest signal in the file, and it is negative. Across the Form 4 corpus from 15 January 2025 to 17 June 2026 — a window in which the shares fell from roughly $75 to roughly $50 — the transaction-code census is option exercises (M) 76, tax withholding (F) 63, sales (S) 16, grants (A) 14, gifts (G) 1, and:

• Zero open-market purchases (code P). Not one, by anyone, over eighteen months and a 40% decline.
• Open-market sales totalled 256,358 shares for ~$20.83M, of which CEO Robert Painter sold 216,347 shares for ~$17.96M across five transactions. Others: M. Schwartz (SVP, AECO) 16,725 shares / ~$1.12M; C. Keating 13,074 / ~$1.01M; J. Allison 3,571; P. Large 3,000; J. Dalton 2,141; R. Bisio 1,500.
• All sale lines appear in filings referencing Rule 10b5-1 plans, which appropriately blunts the signal from the sales. It does nothing to blunt the signal from the absence of purchases.
• Following the 14 April 2026 sale of 7,500 shares at $66.94, press coverage reports Painter directly owned 11,897 shares — approximately $630,000 at the current price, for the CEO of an $11.7B company. Outside directors are held to an 8x-retainer standard (~$520,000) and all meet it; there is no comparable executive figure that this direct holding satisfies.

The contrast with the nearest comparable situation is stark and instructive. Autodesk’s shares fell ~40% into mid-2026 on the same AI-disruption fear, and Autodesk’s CEO, CFO and a director bought approximately $1.8M of stock on the open market in 2026 at the lows, code-P, with no 10b5-1 plan (Autodesk Forms 4, 2026). Trimble’s management told analysts on 6 May 2026 that “nothing has fundamentally changed in the market” and raised guidance. No one bought.

Verdict: below average. Two good decisions (the Ag exit; leverage discipline) sit against a decade of goodwill accumulation that cut the return on capital roughly in half, a $2.09B acquisition that produced the portfolio’s worst segment, a programmatic buyback executed materially above the current price, an incentive structure containing no cost-of-capital or cash-flow metric and paying 138.6% of target in a year of declining reported revenue, and an insider cohort that has not purchased a single share into a 40% decline. Capital allocation is the bridge between business value and shareholder value; here it is where the value has been leaking.

8. Changes and Headwinds — Last Two Years

1. The internal-control failure — the dominant development, and it is unresolved. The sequence, from primary filings:

Three points of interpretation. First, the mitigating facts are real and should be stated: KPMG’s opinion on the financial statements themselves is unqualified (“present fairly, in all material respects”), and the 10-K states “the control deficiencies did not result in a material misstatement… for the year ended January 2, 2026.” No restatement of FY2025 has occurred. Second, the weaknesses are nonetheless in exactly the wrong place — revenue and income taxes are the two processes producing every number in the bull case, and the specific failure includes “the completeness and accuracy of information produced by the Company utilized in the performance of controls, including the retention of relevant evidence.” Third, and most importantly for a forward view: this is the third consecutive year unremediated, it was found by the auditor rather than by management, it was found while preparing for a regulatory inspection, it survived a change of auditor, and the FY2025 remediation section still lists the same four workstreams (ITGCs, process-level controls, personnel capability, risk assessment) as ongoing. Management did report progress: during 2025 it “designed, implemented and operated effective ITGCs over certain of our IT systems, including our core ERP system, revenue management system and key reporting tool.” That is genuine and is the basis on which a FY2026 clean opinion becomes plausible.

2. Portfolio reconstruction completed. The Ag divestiture (Q2-2024) into the AGCO PTx Trimble JV — $1.9B cash plus a retained 15% stake, a $1,687.9M pre-tax gain, roughly half the proceeds to debt repayment, $122.0M of tax paid in 2024 and $277.4M in Q2-2025. The Mobility divestiture (8 February 2025) to Platform Science for $253.9M of Level 3 equity plus rights to 32.5% of the company. Together these removed roughly $700M of low-margin revenue, reset the segment structure to three, and drove the mix to 79% software/services/recurring — while producing the −3% reported revenue decline and the divestiture-mix margin expansion analysed in Section 6.

3. CFO transition. David G. Barnes was CFO through the onset of the control episode (he signs the May 2024 NT 10-Q); Phillip Sawarynski signs the March 2025 NT 10-K and is CFO throughout FY2025 and the 2026 proxy. The 10-K does not link the transition to the control matters, and no cause is asserted here.

4. The AI pivot, and the AI question. Trimble launched SketchUp AI in Q4-2025 (an $11.99/month credit-bundled add-on), announced the Anthropic Claude / SketchUp MCP integration in late April 2026, agreed to acquire Document Crunch (2 April 2026) to create an AI risk-management category, and states it is “approaching our target to dedicate 10% of our development resources to an applied AI organization.” Management’s position is that AI expands the addressable market: “we see a world where AI increases the size of the addressable market, and we believe we have a compelling right to win.” The market’s position, judging by the Q1 call, is the opposite — six of roughly ten analysts asked variations of whether frontier models will erode seat-based licensing, including Berenberg asking directly whether Anthropic “might be able to replicate some of the SketchUp features through the data access.” Painter’s answer on seat risk was about demand rather than pricing architecture: “our customers have significant backlog. There’s a labor shortage… in the near term, it’s hard to see any kind of fundamental worker displacement.” That may well be right, and it does not address whether the unit of monetisation survives.

5. Macro and geopolitical headwinds, named by management. Tariff-policy uncertainty; “the conflict we see in the Middle East”; “high oil prices, high memory prices”; a “constrained freight market” in T&L with North America only “beginning to show some signs of market recovery.” These are cited as the reasons for a Field Systems full-year organic guide of low-to-mid single digits against a Q1 print of +12%.

6. Financing and structure. A new $1.25B five-year unsecured revolver (4 December 2025, expandable to $1.75B, undrawn) replaced the 2022 facility; $1,400.0M of principal debt outstanding; leverage 1.1x. Director J.C. Dalton (since 2020) is not standing for re-election in 2026.

7. The market’s verdict, twice delivered against the tape. −12.8% over 14–20 January 2026 (IGV −5.1%) with no company disclosure; −19.5% over 5–18 May 2026 on a beat-and-raise, while IGV rose 5.2% and PTC rose 4.7%. Analysts cut forecasts on “upbeat” Q1 results.

Verdict: the changes strengthen the business and weaken the thesis. The portfolio is genuinely better — more recurring, higher-margin, more focused, less levered. But the same two years produced an unremediated three-year control failure culminating in an adverse audit opinion on the revenue process, a delisting-rule brush, the loss of a 39-year auditor, a lost multi-year growth framework with no successor, and two idiosyncratic price rejections. A better business that the market trusts less is a specific and uncomfortable combination, and it is the correct description of Trimble today.

9. Risk Analysis

Risk of catastrophic or total loss: low. Leverage is 1.1x, the revolver is undrawn, normalized free cash flow is ~$668M against $25.3M of capex, and the businesses are diversified across three segments and multiple geographies. The realistic downside is a multi-year de-rating and dead money, not impairment of the enterprise. The single risk that could change that assessment is #1 escalating from an adverse control opinion to a restatement — which is not the base case (KPMG’s financial-statement opinion is unqualified and no misstatement has been identified) but which cannot be ruled out while the revenue-process controls remain ineffective for a fourth year.

10. Valuation Discussion

No price target and no recommendation appear in this section. Valuation is discussed only as embedded expectations and scenarios.

Inputs (24 July 2026). Price $52.91. Market capitalisation ~$11,737M on the AZI-implied 221.8M shares. Short- plus long-term debt $1,412.8M and cash $234.1M (Q1-2026 10-Q) give net debt of $1,178.7M and EV ≈ $12,916M. Caveat: the exact current share count is not in a filed document — Q1-2026 diluted weighted-average shares were 236.9M, on which market capitalisation would be $12,534M and EV $13,713M. Every conclusion below is unchanged in direction; the sensitivity is noted where it matters.

Multiples, trailing and forward.

The owner-FCF adjustment is the crux of the valuation and it is not optional. Management guides free cash flow to “approximately 1x non-GAAP net income,” and non-GAAP net income excludes stock-based compensation running at $146–159M per year. Reported free cash flow adds SBC back as non-cash, but the dilution is real and Trimble spends real cash offsetting it — roughly 17% of the FY2025 buyback did nothing but neutralise issuance. Deducting after-tax SBC gives owner-free-cash-flow of approximately $680M and a 5.8% yield. On the alternative 236.9M share count the yield is 5.4%. That is the honest number, and it is materially less attractive than the 7.0% headline. For scale, applying the identical adjustment to PTC produces a ~5.8% owner-FCF yield at $124 — i.e. Trimble and PTC currently offer the same owner cash yield, with PTC earning roughly twice the ROIC on an 84% gross margin.

Own-history context, read correctly. AZI’s valuation-index percentiles at 24 July 2026: P/E 27.75 at the 21.0th percentile, P/B 2.22 at the 1.3rd, P/S 3.43 at the 32.8th, composite 18.4th (n = 3 components).

The 1.3rd-percentile price-to-book reading is the most-cited bullish statistic on this name and it should be discarded. Book value was mechanically inflated by the 2024 Ag divestiture gain — retained earnings rose from $2,437.4M (2023) to $3,757.6M (2024) on a $1,687.9M pre-tax, non-recurring gain — so the denominator of P/B grew ~$1.3B for reasons unrelated to earning power, while the numerator fell. Worse, that book value is 106% intangibles: tangible common equity is negative $327.6M, so there is no asset backing behind the “cheapest-ever on book” claim. Per the LEARNINGS.md caveat on distorted GAAP inputs, the informative percentiles here are P/S at the 32.8th and P/E at the 21.0th — cheap relative to Trimble’s own decade, but nowhere near the extremes the P/B figure implies.

And the relative own-history comparison is the single most important valuation fact in this article. Same source, same date:

Measured against their own multi-year histories, every incumbent peer has de-rated further than Trimble — Bentley, PTC and Autodesk all sit at or inside the 3rd percentile of their own ten-year ranges while Trimble sits at the 18th. The AI-disruption de-rating in design-and-construction software is real, sector-wide, and Trimble has participated in it less than the three companies that earn 1.7x to 3.3x its return on capital at 12 to 22 percentage points higher gross margins. This directly contradicts the popular “the market still sees the old business / deep value” framing: on its own history Trimble is the least washed-out incumbent in its complex, not the most.

Cross-sectional comparison, normalized. Price-to-sales: TRMB 3.43x, PTC 4.74x, ADSK 5.98x, BSY 6.66x, IOT 11.09x. Trimble is genuinely the cheapest on sales — and it has the lowest gross margin (69.1% vs 81.5–91%) and the lowest ROIC (6.9% vs 11.6–23%) of the group. Normalising for mix using EV/gross profit, the right lens when hardware content differs: Trimble at 5.05x against Autodesk at roughly 7.33x (EV $50.1B ÷ ~$6,831M of TTM gross profit) is a ~31% discount, for a business earning roughly one-third of Autodesk’s ROIC on a balance sheet with negative tangible equity, in a segment where Autodesk’s AECO business is 2.4x larger and growing more than twice as fast. The discount is explained by the quality gap rather than additive to it.

Embedded expectations — what the current price requires. At $52.91, a 5.8% owner-FCF yield and 8% guided revenue growth, the market is underwriting four propositions:

1. ~8% revenue and ~13% ARR growth are sustainable. Plausible on the run-rate, but Q1-2026 organic ARR growth was 12% and decelerating from 14%, and Field Systems is guided to low-to-mid single-digit organic growth.
2. A 29.7% EBITDA margin holds and progresses toward 30%. Plausible — but the FY2025 expansion was divestiture mix, AECO’s FY2025 margin was flat, Field Systems’ Q1-2026 margin fell 90bps against a required 220bps H2 improvement, and Q2-2026 is guided to just +30bps y/y.
3. ~5% annual share shrink continues. Well-supported: $608M remains authorized and the policy is at least one-third of free cash flow.
4. The reported numbers are reliable. This is the proposition the auditor has declined to endorse, and it is the one the market appears to be discounting.

Equally instructive is what the price does not embed: any re-rating toward peer software multiples; any value for the Level 3 stakes (32.5% of Platform Science, 15% of PTx Trimble, carried at $616.8M — roughly $2.78/share at carrying value, entirely unrecognised in the multiple); or any 2028+ growth framework, because none has been published.

Scenarios (assumption: no further large debt-funded acquisition).

The base case lands approximately at the current price, which is the correct reading: at $52.91 Trimble is priced roughly for delivery of its own guidance with no re-rating, and the shareholder return in that case comes almost entirely from the buyback. The bear case is not a catastrophe (~−26% from here on the equity), and the bull case is a genuine double — but the bull case’s first condition is an audit outcome rather than an operating one.

Sum-of-the-parts sanity check. Applying segment-appropriate multiples to FY2025 segment operating income before corporate costs: AECO ($512.1M, 34.2% margin, mid-teens ARR growth) at 16x = $8.2B; Field Systems ($478.1M, 31.1%, cyclical hardware-and-software) at 11x = $5.3B; T&L ($120.5M, 21.9%, +2% organic) at 9x = $1.1B. Gross $14.6B, less capitalised unallocated corporate expense ($122.6M at 11x = $1.3B) = ~$13.3B, plus $616.8M of Level 3 stakes, less $1,178.7M of net debt = ~$12.7B of equity value, or ~$57/share on 221.8M shares. That is ~8% above the current price and confirms the base case: the parts are worth roughly the whole, with no large hidden discount. It also identifies where the value sits — AECO is ~62% of the equity value on ~42% of the revenue — and therefore why AECO’s flat FY2025 margin matters far more than the consolidated figure.

Verdict: fair-to-slightly-cheap on the reported numbers, and the reported numbers carry an adverse control opinion. Trimble trades at a defensible 11.2x forward EBITDA and 14.9x forward non-GAAP earnings for a 5.8% owner cash yield and 8% growth — a fine price for a business earning its cost of capital, and a full one for a business earning 6.9% on invested capital with negative tangible equity. The cheapness is real but shallower than the headline P/B suggests, less advanced than every peer’s on an own-history basis, and approximately proportional to the quality gap on a cross-sectional basis. What is not in the price is a re-rating; what is also not in the price, arguably, is the possibility that the numbers themselves require revision.

11. Variant Perception

The consensus belief. The sell side is close to unanimous: eleven covering firms with nine buys, one strong buy and one hold as of February 2026, with a mean target implying roughly 54% upside as of 22 July 2026. The published bull thesis is captured in a July 2026 Seeking Alpha title — “Trimble: The Market Still Sees The Old Business” — arguing that Trimble is “evolving into a software-driven, recurring-revenue business, yet the stock trades at a discounted low-teens forward earnings multiple,” supported by 12% organic revenue growth, 13% ARR growth, 71% gross margins, robust free cash flow, and a valuation “attractive versus peers like Bentley and PTC.” Zacks upgraded the stock to Strong Buy in February 2026 and published a “buy the dip” note on 18 May at the bottom of the May decline. Consensus, in short, believes this is a quality software transition mispriced as legacy hardware.

The strongest bull case, stated fairly. It is better than a sceptic would like. (1) The transformation is real: 79% software/services/recurring, gross margin up 1,330bps in five years to 69.1%, adjusted EBITDA margin 29.2%, AECO at 26 consecutive quarters of mid-teens-plus organic ARR growth. (2) Field Systems is a genuine oligopoly with OEM factory-fit across 24 construction OEMs and a dealer network that cannot be replicated with software. (3) The cash economics are good and understated: reported FY2025 free cash flow of $360.9M normalizes to ~$668M once the one-time $277.4M Ag tax payment is added back, on $25.3M of capex, with a further ~$53M/year of OBBBA cash-tax relief. (4) The balance sheet is safe at 1.1x with an undrawn $1.25B revolver. (5) Capital return is substantial: $863.4M in FY2025 and $317M in Q1-2026, retiring ~6% of the shares in eighteen months with $608M authorized. (6) The setup is empirically washed out: −37.9% from the 52-week high, 21.1% idiosyncratic volatility, negative momentum, at the 21st/33rd percentile of its own P/E and P/S history. (7) There is unpriced optionality: $616.8M of Level 3 stakes (32.5% of Platform Science, 15% of PTx Trimble) worth ~$2.78/share at carrying value and worth nothing in the multiple. (8) AI genuinely could expand the market — labour is short by “hundreds of thousands of workers” in North America and automation substitutes for it. (9) Q1-2026 was a real beat with real AECO operating leverage (+420bps).

The strongest bear case. (1) The auditor has issued an adverse opinion on the controls over revenue and income taxes, for the third consecutive year, and the weakness was still unremediated in the Q1-2026 10-Q. The bull case is constructed entirely from ARR, adjusted EBITDA and segment margins — all outputs of that process. (2) ROIC is 6.92% and lower than in 2021, below any defensible cost of capital for a 1.26-beta equity; ~$4.9B of acquisition spend produced $5.24B of goodwill and negative $327.6M of tangible common equity. Growth at a sub-WACC return destroys value regardless of how fast ARR grows. (3) The margin expansion is subtraction, not leverage — the 10-K attributes it three times to “the loss of lower margin divestiture income,” AECO’s FY2025 margin was flat at 34.2%, and 25 divestitures in six years means the pruning is nearly done. (4) The presentation gap is systematic: −3% / +6% / +10% for the same year’s revenue; +6% / +14% for the same year’s ARR; non-GAAP EPS 78–88% above GAAP on add-backs that are 38% stock compensation with a rising restructuring line. (5) The 2027 framework is nearly met in 2026 with no 2028 successor — “We’re not yet ready to talk about 2028.” (6) Incentives contain no return-on-capital and no free-cash-flow metric, paid out at 138.6% of target in a year of −3% reported revenue growth, and are anchored on ARR — the metric produced by the uncontrolled process. (7) Zero insider open-market purchases in eighteen months and a 40% decline; the CEO sold ~$18.0M and directly holds ~11,897 shares. (8) Only 20% of customers buy more than one product after five years of a strategy premised on ecosystem network effects.

The three to five assumptions that actually matter.

1. Are the reported ARR and segment margins reliable? This is not a normal analytical question and it is the reason this memo reaches the conclusion it does. The mitigants are real — an unqualified financial-statement opinion, no identified misstatement, documented ITGC progress in 2025 on “our core ERP system, revenue management system and key reporting tool.” The aggravants are also real — three years, auditor-found rather than management-found, found while preparing for a PCAOB inspection, surviving an auditor change, and located precisely in revenue.
2. Is 12–13% organic ARR growth the trend or the peak? Q4-2025 14% → Q1-2026 12%, against a 13% FY guide, with Field Systems guided to low-to-mid single-digit organic growth after a +12% quarter.
3. Does AECO show genuine operating leverage, or was FY2025’s flat margin the truth? Q1-2026’s +420bps says leverage; FY2025’s +10bps says not. AECO is ~62% of the sum-of-the-parts equity value, so this single line resolves most of the thesis.
4. Does AI expand Trimble’s market or compress its unit of monetisation? Currently unfalsifiable in either direction, which is precisely why the whole complex has de-rated.
5. Will capital allocation change? The incentive structure says no: no return-on-capital metric, an explicit ongoing “tuck-in acquisitions… primarily in construction software” programme, and a formulaic rather than valuation-contingent buyback.

The factor-positioning read, as evidence on where consensus is offsides. The empirical positioning contradicts the consensus framing in three specific ways. First, there is no Value loading in any of the four nested models — Base, Base+Sector, Base+Sector+Industry, All Factors. If Trimble were being priced as a value stock, a Value beta would appear; it does not. What appears instead is Market +1.20, DividendYield +0.52, SmallSize +0.33, BetaFactor +0.31 and Momentum −0.29 in the Base model, and Cloud Computing +0.32, Software +0.20, LowVolatility −0.22 and Momentum −0.25 in the industry model. The market is treating Trimble as a high-beta, negative-momentum, small-size software expression — not as a cheap asset. Second, the Quality loading is slightly negative (−0.03), which is unusual for a 69%-gross-margin business and is consistent with the ROIC and negative-tangible-equity findings rather than with the “quality compounder on sale” thesis. Third, and most tellingly, the nearest factor neighbours are mid-cap and multifactor index ETFs — FMDE 0.870, OAKM 0.869, VFMF 0.865, SMLF 0.865, VB 0.864, PKW 0.864, IJH 0.863, MDY 0.862 — with KKR the only individual common stock in the top ten. Trimble currently has no tight single-stock factor comparable: it trades as generic mid-cap beta rather than as a member of a software cohort, which is itself evidence that the market has stopped valuing it as a software franchise. The risk-adjusted record supports that: a −0.29 Sharpe and −7.6% annualized return over five years with a −57.4% peak drawdown, and −1.29 Sharpe over one year. The regime is hostile — Market +17.9% over 252 days (z +0.46) with InterestRate +5.1% over 21 days (z +1.91), a rising-rate, rising-market environment that punishes negative-momentum long-duration software with no Value support. (Loadings are read within single models only, per the hierarchical-orthogonalization rule; these are third-party statistical estimates, and the persistence claim is interpretation, not fact.)

Where I think consensus is offsides, and it is not where the bulls think. The bull thesis and the bear thesis agree that Trimble is cheap; they disagree on why. Consensus reads the de-rating as a sector-wide AI-fear wash from which Trimble will re-rate with its peers, and points to the 52-week low and the 1.3rd-percentile P/B. Two pieces of evidence say this is wrong. First, the price action was idiosyncratic, twice. Trimble fell 12.8% over 14–20 January while IGV fell 5.1%, and 19.5% over 5–18 May while IGV rose 5.2% and PTC rose 4.7% — on a beat-and-raise, with analysts cutting forecasts on “upbeat” results. Autodesk and PTC fell with their sector; Trimble fell against it. That is not a sector wash, it is a company-specific verdict. Second, the own-history percentiles say Trimble is the least washed-out incumbent in its complex, not the most — 18.4th composite against Bentley’s 1.3rd, PTC’s 2.0th and Autodesk’s 3.0rd, while earning the lowest ROIC and the lowest gross margin of the four. If one wants to buy the AI-fear de-rating in design-and-construction software, the evidence says Trimble is the worst risk-adjusted vehicle for doing it: Autodesk offers 23% ROIC, 91% gross margins, a 3rd-percentile own-history valuation and — uniquely — a 2026 insider open-market buying cluster from its CEO, CFO and a director.

What would falsify each side. Falsifies the bull: a fourth consecutive adverse ICFR opinion or any restatement; organic ARR growth printing below 10%; AECO segment margin flat or down on positive organic growth for two consecutive quarters; the 2027 “3-4-30” targets retired without a 2028 framework; another debt-funded acquisition above 6x sales. Falsifies the bear: a clean, unqualified FY2026 ICFR opinion from KPMG; two or more quarters of AECO segment-margin expansion on organic growth (proving leverage rather than mix); organic ARR growth stabilising at or above 13%; ROIC crossing 10%; and insider open-market purchases.

12. Fact vs. Interpretation

13. Open Questions

1. Will KPMG issue a clean ICFR opinion for FY2026? The single most important question on this name. Management reported designing and operating effective ITGCs over “our core ERP system, revenue management system and key reporting tool” during 2025, which makes it plausible; three consecutive failures make it uncertain. Resolution: the FY2026 10-K, expected ~February 2027.
2. What caused the 14–20 January 2026 decline? TRMB fell 12.8% while IGV fell 5.1%, with no 8-K, guidance change or company disclosure in the window (the only 8-K is the 4-December-2025 revolver). Roughly 6pp is explained by a 1.26 beta on the sector move; the residual is unexplained and is not attributed to a cause in this article.
3. What is net revenue retention, gross retention and churn? Not disclosed in the FY2025 10-K, the Q1-2026 10-Q, the 2026 proxy or either recent earnings call. For a company valued on ARR this is a material gap, and it cannot be independently reconstructed while the revenue-process controls carry an adverse opinion.
4. What is the exact current share count? AZI implies 221.8M at 24 July 2026; Q1-2026 diluted weighted-average was 236.9M; $317M was repurchased in Q1-2026 with $608M remaining authorized. The Q2-2026 10-Q resolves it.
5. Was AECO’s Q1-2026 +420bps margin expansion structural or seasonal/timing? FY2025 AECO margin was flat; Q4-2025’s 44% was “aided by January 1 term license renewals.” Two more quarters are needed to distinguish leverage from calendar.
6. What is the reconciliation between ARR ($2,435M at Q1-2026) and total RPO (~$2.0B)? Part is the ASC 606 short-contract practical expedient, but the split between contractually committed and annualized-run-rate ARR is not disclosed.
7. Will a 2028+ growth framework be published, and at what level? Painter suggested the November 2026 Trimble Dimensions user conference as a possible venue but said “we haven’t yet even thought about when would be the next logical Investor Day.”
8. What are the Platform Science and PTx Trimble stakes actually worth? $616.8M carried on “unobservable inputs, including discounted cash flow projections, market comparables, and an option pricing model.” Platform Science competes with Trimble’s own former Mobility customers; PTx Trimble depends on the agricultural equipment cycle.
9. How much of the Field Systems H2-2026 deceleration is conservatism versus visibility? Guided low-to-mid single-digit organic against a +12% Q1; management denied pull-forward twice and cited comps, the Middle East and tariffs.
10. Was the CFO transition (Barnes → Sawarynski) connected to the control failure? No causal disclosure exists; no inference is drawn here.
11. What is Trimble’s actual seat exposure to AI substitution? SketchUp’s professional-versus-consumer seat split is not disclosed, nor is the share of AECO ARR that is per-named-user.
12. Will the buyback become valuation-contingent? The policy is formulaic (“at least 1/3 of our free cash flow”); $317M was spent in Q1-2026 well above the current price.

14. What Must Be True

The bull case requires all of the following

1. The reported financials are substantially accurate despite three years of ineffective controls over revenue and income taxes. Falsification test: any restatement of FY2024 or FY2025 revenue, or a fourth consecutive adverse ICFR opinion in the FY2026 10-K, kills the thesis outright. A clean FY2026 opinion confirms it. Binary, dated, and resolves by ~February 2027.
2. Organic ARR growth holds at 12–13% or better and does not decelerate through 10%. Falsification test: two consecutive quarters of organic ARR growth below 11% as reported in the 10-Q (not the call). Watch Q2-2026 (~early August 2026) and Q3-2026.
3. AECO delivers genuine operating leverage — segment margin expanding on organic growth, sustained beyond one quarter. Falsification test: AECO segment operating margin flat or down year-over-year on positive organic revenue growth in two of the next three quarters, per the 10-Q segment note. AECO is ~62% of the sum-of-the-parts equity value, so this line carries most of the thesis.
4. AI expands the addressable market rather than compressing the unit of monetisation. Falsification test: disclosed decline in named-user seats or in AECO ARR per customer, or a shift in guidance language from “expanding TAM” to “transitioning pricing models.” Currently unfalsifiable in either direction, which is why the whole complex is discounted.
5. Capital allocation does not repeat the Transporeon error, and the buyback continues at ≥1/3 of free cash flow. Falsification test: any acquisition above ~$500M at more than ~6x sales, or the buyback being cut to fund M&A.
6. Field Systems’ low-to-mid single-digit organic guide proves to be conservatism rather than the run-rate. Falsification test: Field Systems organic revenue growth below 5% in Q2 and Q3 2026.

The bear case requires all of the following

1. The control weaknesses matter economically — either through restatement or through a persistent trust discount that prevents re-rating. Falsification test: a clean FY2026 ICFR opinion followed by a multiple re-rating toward the peer complex within two quarters. This is the bear’s most vulnerable leg: the weaknesses may simply be remediated, in which case the discount closes.
2. ROIC stays below the cost of capital, so growth continues to destroy rather than create value. Falsification test: ROIC crossing 10% and rising for two consecutive years, driven by operating income rather than by write-downs of the goodwill denominator.
3. The margin expansion was divestiture mix and does not continue once the pruning is exhausted. Falsification test: consolidated non-GAAP operating margin expanding more than 150bps in FY2026 with AECO segment margin expanding — i.e. leverage without further disposals.
4. The de-rating is company-specific and does not simply mean-revert with the sector. Falsification test: TRMB outperforming IGV by more than 15pp over two quarters without any change in the control or growth facts, which would indicate the January and May moves were sector beta after all.
5. Insider behaviour is informative — management’s non-purchasing reflects a view. Falsification test: a code-P open-market purchase cluster from the CEO, CFO or a director. This is the fastest and cheapest signal available and it would materially weaken the bear case.
6. The absence of a 2028 framework reflects genuine uncertainty about the growth algorithm beyond 2027. Falsification test: a 2028–2030 framework published at Trimble Dimensions in November 2026 with a growth algorithm at or above the current 8% revenue / 13% ARR pace.

15. Source Appendix

The full source appendix is provided as Appendix B to this article. Primary sources relied upon: the FY2025 Form 10-K (filed 2026-02-25, period ended 2026-01-02) including Item 9A and the KPMG audit reports; the FY2024, FY2023, FY2022 and FY2021 Forms 10-K and the FY2023 10-K/A; the Q1-2026 Form 10-Q (filed 2026-05-06) and the trailing 10-Q series including the three delayed 2024 filings of 2025-01-16; the four Form 12b-25 late-filing notifications (2024-05-08, 2024-08-07, 2024-11-07, 2025-03-05); the 2026 DEF 14A (filed 2026-04-09) and the 2022–2025 proxies; the 8-K corpus (67 filings) including the 2025-12-05 credit-agreement filing; the Form 4 corpus (353 filings enumerated, 122 pulled for the 2025–2026 window); the Q1-2026 (2026-05-06) and Q4-2025 (2026-02-10) earnings-call transcripts via the ROIC.ai MCP; ROIC.ai computed statements and ratios for TRMB, ADSK and BSY; the AZI price CSVs and valuation-index percentiles for TRMB, ADSK, PTC, BSY and IOT; the FactorsToday factor model (stock-loadings, leaderboard, stock-info, stock-specific-vol, related-stocks, factor-returns); and the author’s own prior published coverage of Autodesk and PTC for peer cross-read.

APPENDIX A — Standard Diligence Questionnaire

Trimble Inc. (NASDAQ: TRMB) · Report date: 2026-07-25

A standard diligence questionnaire, answered from the primary sources listed in Appendix B and labelled FACT / INTERPRETATION / ASSUMPTION / OPEN QUESTION where the distinction matters. Where a question does not map to the business model, the correct sector analog is given. No recommendation and no price target appear in this appendix.

General

What thoughtful questions have other investors asked about this company?

The Q1-2026 earnings call is an unusually clear window into what sophisticated holders actually care about, because six of roughly ten analysts asked variations of a single question: does AI destroy the seat-based licence? The best-formed versions:

• Berenberg (Nay Soe Naing) asked the sharpest one — whether Anthropic gains access to SketchUp-created data and “might be able to replicate some of the SketchUp features through the data access.” Painter’s answer (“I don’t see a near-term concern”) did not address the medium term.
• JPMorgan (Tami Zakaria) asked why Claude users would migrate to SketchUp “instead of just staying on Claude that probably keeps getting better at giving customized designs” — and got a genuinely substantive answer about iteration, multi-user collaboration, rendering and daylight analysis being the professional workflow that an LLM output cannot complete.
• William Blair (Jonathan Ho) asked the question that actually matters commercially: “with the use of AI potentially maybe displacing some workers, do you see any threat to any of the seat-based licensing models?” Painter answered about demand (“our customers have significant backlog. There’s a labor shortage”) rather than about the unit of monetisation.
• Barclays (Guy Hardwick) asked the best financial question on the call: Trimble beat Q1 revenue by ~$35M and raised the full year by only $15M while cutting the top of the EBITDA range by 20bps — “is it really a pull forward you saw in Q1?” Painter denied pull-forward and cited a policy of not raising guidance three months into a year. Hardwick then asked the second-best question: with 2027 targets nearly met, what about 2028? Answer: “We’re not yet ready to talk about 2028.”
• Wolfe (Joshua Tilton) asked whether Field Systems’ outlook contained more conservatism than 90 days earlier. Answer: “no fundamental change in view.”
• Oppenheimer (Kristen Owen) asked about back-half scenarios and the level of conservatism in the guide; Piper Sandler (Clarke Jeffries) asked whether recurring-revenue operating leverage is genuinely beginning to show in margin.

The questions not asked on either recent call are as revealing: no analyst asked about the adverse internal-control opinion, the third consecutive year of unremediated material weaknesses in revenue and income taxes, the return on invested capital, the negative tangible common equity, or the absence of insider buying. INTERPRETATION: the sell side is underwriting the AI question and ignoring the accounting-control and capital-return questions, which is where the greater analytical edge appears to sit.

Cyclicality & Earnings Nature

Are earnings at a cyclical high or low? INTERPRETATION: mid-cycle with a mixed split by segment, and closer to a high than a low on margin. FY2025 gross margin (69.1%) and operating margin (16.5% GAAP / 27.5% non-GAAP) are all-time records, and management describes them as such. Against that, T&L is operating in “a constrained freight market” with North America “beginning to show some signs of market recovery” — a cyclical low. Field Systems saw “strong end-user demand and competitive wins for Civil Construction” partially offset by “lower demand in Surveying” — mixed. The margin records are the important part: FACT — the 10-K attributes the margin gains to “the loss of lower margin divestiture income,” which means the record margin is a structural mix outcome rather than a cyclical peak, and it will not mean-revert down. But it also will not repeat.

Driven by the external environment or internal actions? FACT/INTERPRETATION: overwhelmingly internal actions, and specifically portfolio actions. The FY2021→FY2025 transformation — gross margin 55.6% → 69.1%, recurring mix to 79%, revenue −2% — is the arithmetic of 13 acquisitions and 25 divestitures since 2020, not of end-market improvement. Reported revenue fell in 2024 and 2025. The Ag and Mobility disposals removed roughly $700M of low-margin revenue. External environment has been a net headwind (tariffs, Middle East conflict, weak freight, high memory and oil prices, all named by management).

How stable are revenues? FACT: 79% of FY2025 revenue is software, services and recurring (78% in Q1-2026); ARR is $2,435M against $3,875M of guided FY2026 revenue, i.e. ~63% of revenue is on a recurring run-rate. Deferred revenue is $970.7M total. That is genuinely resilient by industrial standards. Two qualifications. FACT: ARR as Trimble defines it is a single quarter annualized and includes term-licence contract value, which under ASC 606 is recognised largely upfront — it is not a contracted-backlog measure; total RPO is ~$2.0B, below reported ARR. FACT: the remaining ~37% of revenue is hardware, which management says is “the one that has — we have the least visibility with.”

Outlook for products/services? FACT (guidance): FY2026 revenue $3.875B (+~8%), ARR growth 13%, EBITDA margin 29.7%, non-GAAP EPS $3.55. Q2-2026: revenue $950M (+7.5%), EPS $0.80, EBITDA margin 27.7% (+30bps y/y). By segment the guide implies AECO low-to-mid-teens revenue growth, Field Systems low-to-mid single-digit organic growth (against +12% in Q1), and T&L continuing modest growth off a weak freight base.

How big will this market be — growing, shrinking, domestic or international? INTERPRETATION: growing, with construction-technology penetration still low and two durable tailwinds — a North American construction labour shortage of “hundreds of thousands of workers” and infrastructure plus data-centre construction demand, both named by management as Field Systems drivers. Geographically Trimble is North-America-centric (“centric to capturing the North American opportunity”) with faster European AECO growth in Q1-2026 and a Q1-2026 APAC launch of Trimble Construction One. OPEN QUESTION: Trimble does not publish a TAM figure by segment, and the AI question makes any TAM estimate unusually unstable in both directions — management argues AI expands it, the market is pricing compression.

Business Quality & Competitive Moat

Is the industry getting more or less competitive? INTERPRETATION: more competitive in AECO, stable in Field Systems, more competitive in T&L. In AECO, Autodesk’s AECO segment generated $3,583M (+22%) — 2.4x Trimble’s entire AECO segment growing more than twice as fast — and Bentley owns the infrastructure-engineering niche at 81.5% gross margins. In Field Systems the OEM/dealer oligopoly with Topcon and Hexagon/Leica is stable and slow-moving. In T&L, project44, Samsara, Motive and Platform Science (in which Trimble now holds up to 32.5%) compete in a crowded, cyclically weak market. Overlaying all three, generative AI is a new and unquantified entrant threat at the authoring layer.

How profitable is the business (ROIC, ROE)? FACT: ROIC 6.92% (FY2025), down from 9.24% (FY2021), with a 4.71% trough in FY2024. Return on capital 9.79%; ROE 11.87%; ROA 4.51%. The FY2024 ROE of 48.6% is an artifact of the $1,687.9M Ag divestiture gain and should be disregarded. INTERPRETATION: this is the single most damaging number in the file. A 69%-gross-margin business earning 6.9% on invested capital is telling you the capital was mis-priced, not that the operations are bad. Peer ROICs: Autodesk ~23%, PTC ~15%, Bentley 11.6%.

How profitable is the industry — how many competitors, what barriers to entry? FACT: the industry is highly profitable for the best-positioned participants — gross margins of 81.5% (Bentley), ~84% (PTC), ~91% (Autodesk) against Trimble’s 69.1%. Barriers differ by layer. INTERPRETATION (Greenwald): at the format/authoring layer barriers are very high and Trimble does not own them (Autodesk’s Revit does, reinforced by government BIM mandates). At the machine-control/positioning layer barriers are high and Trimble does own them — factory-fit across 24 construction OEMs, a global independent-dealer network for sales/service/calibration, and a proprietary grade-control engine. At the freight-network layer barriers are moderate and density-driven, and Transporeon has real European density. At the workflow-bundling layer (Trimble Construction One) barriers are the weakest and most contested.

Can the business be easily understood? Partially, and less easily than it should be. The three-segment structure and the physical-to-digital value proposition are intuitive. What is genuinely hard is the numbers: three different FY2025 revenue growth rates circulate (−3%, +6%, +10%), two ARR growth rates (+6% reported, +14% organic), non-GAAP EPS runs 78–88% above GAAP, the comparison base changes with divestitures every year, FY2024 GAAP EPS contains a $1,687.9M gain, and a 53rd week plus January-1 licence renewals distort the year-over-year. INTERPRETATION: reconstructing Trimble’s true growth and margin trajectory takes real work, and the material weakness in revenue controls means the reconstruction cannot be fully verified.

Can it be undermined by foreign low-cost labour? Largely no. The software (AECO, T&L) is not labour-arbitrageable in a way that threatens pricing. The hardware (Field Systems — GNSS receivers, total stations, laser scanners) faces low-cost Asian competition at the commodity end, but Trimble’s position is defended by OEM factory-fit, dealer service and calibration networks, and centimetre-grade accuracy requirements rather than by manufacturing cost. INTERPRETATION: the more relevant substitution threat is not cheap labour but cheap intelligence — generative AI at the authoring layer.

Do brands matter? Moderately, and unevenly. SketchUp is a genuine consumer-and-professional brand with a very large installed base (management notes “the bulk of the user count” is consumer/maker, monetised “at the professional grade level”). Tekla and Trimble in survey and machine control carry real professional trust — a surveyor’s or grade operator’s tool choice is a reputational decision. In construction ERP and project management the brand matters less than the data and the integrations. INTERPRETATION: brand is a supporting rather than a primary advantage; it lowers customer-acquisition cost but does not itself defend price.

What is the nature of competition? Feature/workflow breadth and channel reach in AECO; OEM design-win cycles and dealer coverage in Field Systems; network density and integration count in T&L. Notably not primarily price — Trimble does not describe discounting pressure in either recent call, and gross margin is rising.

Customers’ switching costs? INTERPRETATION: real but unquantifiable, and lower than management implies. High for construction ERP and estimating (historical job-cost data, finance integrations), high for embedded machine control (re-tooling machines and retraining operators), moderate for design authoring, moderate-to-low for project management and point solutions. The decisive disconfirming datapoint is management’s own: only 20% of customers buy more than one product after roughly five years of a strategy premised on ecosystem lock-in. FACT: Trimble discloses no net revenue retention, gross retention, churn or seat count anywhere — so switching costs cannot be measured from outside, and the standard Greenwald market-share-stability test cannot be run.

Financial Condition & Balance Sheet

Assets not fully recognised on the balance sheet? Yes, and they are meaningful. INTERPRETATION: (a) the installed base and OEM relationships — 24 OEM factory-fit relationships and “hundreds of thousands of instruments… in the real physical world” — carry no balance-sheet value; (b) the Trimble Connect data asset (30M+ projects, 50M+ users since inception, 130+ marketplace extensions) is expensed as incurred; © internally developed software and the AI organisation (approaching 10% of development resources) are largely expensed — R&D was $630.7M in FY2025, 17.6% of revenue. Conversely, FACT: the acquired intangibles very much are on the balance sheet at $5,213.6M of goodwill plus $872.1M of other intangibles, and they are the reason ROIC is 6.9%.

Off-balance-sheet liabilities? FACT: modest and disclosed. Fixed operating-lease payment obligations of $208.8M ($46.7M within twelve months) at FY2025 year-end. Unrecognised tax benefits of $79.7M in other non-current liabilities including interest and penalties (FY2025), $39.1M at Q1-2026. Rights to acquire additional Platform Science equity (up to 32.5% total) create a potential future funding obligation of undisclosed size. No material pension obligation, no securitisation, no meaningful guarantee structure disclosed.

How conservative is the accounting? INTERPRETATION: this is the central concern of the entire engagement, and the honest answer is that it cannot be verified. FACT: KPMG issued an adverse opinion on the effectiveness of internal control over financial reporting dated 2026-02-25; management states material weaknesses in ITGCs and in “revenue and related accounts, income taxes, and other financial reporting processes” “exist as of January 2, 2026”; CEO and CFO concluded disclosure controls were “not effective” at FY2025 year-end and again in the Q1-2026 10-Q. This is the third consecutive unremediated year, following a 2024 episode in which EY identified the weakness while preparing for a PCAOB inspection, the FY2023 ICFR opinion was withdrawn, three 10-Qs were filed late in a single batch, Nasdaq issued listing-rule notices, and Form S-3 eligibility was lost until April 2026. EY — auditor since 1986 — was replaced by KPMG in 2025.

The mitigants are real and should be stated: KPMG’s opinion on the financial statements themselves is unqualified, the 10-K states “the control deficiencies did not result in a material misstatement,” no restatement of FY2025 has occurred, and management documented genuine 2025 progress including effective ITGCs over “our core ERP system, revenue management system and key reporting tool.”

Separately on presentation conservatism: FACT — non-GAAP operating income exceeds GAAP by $396.1M (1,100bps) and non-GAAP diluted EPS exceeds GAAP by 78% (88% in Q1-2026); $151.5M of the add-back is stock-based compensation (a real cost) and $53.5M is “restructuring and other” that has risen from $32.4M and is structurally recurring for a company doing 38 portfolio transactions in six years. FACT — $616.8M of equity investments are Level 3, marked on “unobservable inputs, including discounted cash flow projections, market comparables, and an option pricing model.” INTERPRETATION: the presentation is aggressive within the bounds of normal software-industry practice; the controls are not, and they are the reason a wider-than-normal discount is warranted.

How CapEx-hungry is the business? FACT: barely at all. Capital expenditure was $25.3M in FY2025 — 0.7% of revenue — down from $69.0M in 2019. Property and equipment net is $180.8M against $8,991.0M of total assets. This is a genuinely capital-light model at the operating level. INTERPRETATION: the crucial distinction is that Trimble is not capex-hungry, it is acquisition-hungry — ~$4.9B of acquisition cash outflow since 2018 versus ~$338M of cumulative capex over the same period. The capital intensity is entirely in M&A, which is precisely why ROIC is the right metric and why it is 6.9%.

Capital Allocation & Management

How much FCF does the business generate, how does management use it, what is the philosophy? FACT: reported FY2025 free cash flow $360.9M, which normalizes to ~$668M once the one-time $277.4M final Ag divestiture tax payment plus ~$30M of related costs are added back (management’s own framing: “$361 million when considering the $307 million of tax payments and other costs primarily related to divestitures”). Q1-2026 reported FCF $275M. FY2026 guidance: FCF ≈ 1x non-GAAP net income, i.e. ~$824M — approximately $680M on an owner basis after deducting ~$145M of after-tax stock-based compensation. Prior years’ reported FCF: $704.4M (2021), $348.0M (2022), $555.1M (2023), $497.8M (2024).

Stated philosophy: “at least 1/3 of our free cash flow to be used for repurchasing shares,” with M&A “focused on strengthening our core market positions… primarily in construction software with an emphasis on tuck-in acquisitions,” and leverage held well below a 2.5x target (currently 1.1x). No dividend.

INTERPRETATION: the philosophy is coherent and the execution is mixed. Deleveraging from $2,836.8M of net debt (2023) to $1,178.7M was well done. The buyback is formulaic rather than valuation-contingent, and it has been executed materially above the current price.

Significant acquisitions recently? FACT: 13 acquisitions and 25 divestitures since 2020 — 38 portfolio transactions in about six years. Acquisition cash outflow: $1,763.5M (2018, Viewpoint/e-Builder), $220.8M (2019), $201.9M (2020), $236.1M (2021), $373.5M (2022), $2,088.9M (2023, Transporeon), $22.0M (2024), $8.8M (2025) — ~$4.9B cumulative. Recent activity is small: Document Crunch (announced 2 April 2026, price undisclosed, AI construction-contract risk management) and a small Field Systems divestiture in Q1-2026.

INTERPRETATION: Transporeon is the transaction on which the capital-allocation verdict turns. ~$2.09B paid in 2023, debt-funded (net debt to $2,836.8M, net-debt/equity 63%), and three years on T&L is the smallest (15% of revenue), lowest-margin (21.9%) and slowest-organically-growing (+2% FY2025) segment, with reported revenue down 30% in FY2025 and 4.4% in Q1-2026. The network and the >$100M of transactional revenue are genuine strategic assets; the price was not earned. Against that, the Ag divestiture was genuinely good capital allocation — $1.9B cash plus a retained 15% stake for a low-margin hardware business, at a price AGCO set, with roughly half the proceeds retiring debt.

Buying back shares? FACT: yes, and aggressively. $195.1M (2021), $408.3M (2022), $93.3M (2023), $181.5M (2024), $863.4M (2025), $317M (Q1-2026), with $608M remaining authorized. Shares outstanding 250.9M (2021) → 236.0M (2025), with Q1-2026 diluted weighted-average at 236.9M and the AZI-implied current count 221.8M. INTERPRETATION: funding-wise smart, price-wise poor. The FY2025 programme deployed residual Ag proceeds across a year in which the stock traded roughly $62–$85, and a further $317M was spent in Q1-2026, against $52.91 today. Roughly 17% of the FY2025 repurchase merely offset ~$150M of annual SBC issuance. This is programmatic capital return, not counter-cyclical repurchasing.

Issuing large amounts of new shares to insiders? FACT: stock-based compensation was $146.5M (2025), $158.6M (2024), $145.4M (2023) — roughly 4.1% of FY2025 revenue and ~1.2% of market capitalisation annually. From the Form 4 corpus (2025-01-15 to 2026-06-17): grants (code A) 261,876 shares, option exercises (M) 518,802 shares, shares withheld for taxes (F) 212,932 shares. INTERPRETATION: SBC is high in absolute terms but not egregious for software, and the buyback more than offsets it in share count. The problem is not the quantity of issuance but what it is paid for.

Compensation policy of directors/management? FACT (2026 DEF 14A): the annual cash bonus (Trimble OneBonus / “TOP”) uses revenue, adjusted EBITDA and annualized recurring revenue at the corporate level, with CEO target at 200% of salary (400% at maximum, 440% with the ±10% modifier). Long-term incentives are CEO 60% PRSUs / 25% RSUs / 15% options (other NEOs 50/37.5/12.5), with PRSUs vesting 50% on relative TSR versus the S&P 500 over three years and 50% on ARR growth over three years. TSR weighting moved 25/75 (2023) → 50/50 (2024, “based largely on stockholder feedback”) → 50/50 (2025); a new 300% maximum TSR payout was added in 2025 for a top-of-index outcome, with a 250% maximum combined average. 2025 corporate financial measures paid out at 138.6% of target; the non-financial modifier was not applied to any NEO. Outside directors must hold 8x the annual cash retainer (~$520,000) and all comply; director pay is ~$350,000–$400,000 with 3,974 unvested RSUs each. Say-on-Pay support was 93.5% in 2025.

FY2025 compensation: Painter $19,129,944 (salary $950,000; stock $13,311,864; options $2,230,637; non-equity incentive $2,633,443; $21.6M of stock-award value at maximum PRSU achievement); Sawarynski (CFO) $4,453,662; Schwartz $4,900,270; Bisio $4,572,608; Keating $2,208,469.

INTERPRETATION: the structural defect is what is absent — there is no return-on-capital metric and no free-cash-flow metric anywhere in the plan. A search of the 2026 proxy returns zero instances of free cash flow as a compensation measure. Management is paid on revenue, adjusted EBITDA and ARR — three metrics wholly indifferent to the price paid for acquired revenue — at a company that spent ~$4.9B building $5.2B of goodwill and earns 6.9% on invested capital. The plan cannot register the one thing that has gone wrong. A second, sharper problem: ARR appears in both the bonus and the long-term plan, and ARR is produced by the revenue processes carrying an unremediated material weakness and an adverse ICFR opinion. To management’s credit, the shift to 50% relative TSR is a genuine improvement and was made in response to shareholder feedback.

Motivations of management? FACT: zero insider open-market purchases (code P) between 2025-01-15 and 2026-06-17, a window in which the shares fell roughly 40%. Open-market sales totalled 256,358 shares / ~$20.83M, of which CEO Robert Painter sold 216,347 shares for ~$17.96M across five transactions; other sellers were Schwartz (~$1.12M), Keating (~$1.01M), Allison, Large, Dalton and Bisio. All 16 sale lines appear in filings referencing Rule 10b5-1 plans, which appropriately blunts the signal from the sales — and does nothing to blunt the signal from the absence of purchases. Following the 14 April 2026 sale of 7,500 shares at $66.94, press coverage reports Painter directly owned 11,897 shares, roughly $630,000 at the current price.

INTERPRETATION: the alignment picture is poor. On 6 May 2026 Painter told analysts “nothing has fundamentally changed in the market” and raised guidance; the stock then fell 19.5% over eight sessions into a rising software tape; and no insider bought. The instructive contrast is Autodesk, whose shares fell ~40% on the same AI fear and whose CEO, CFO and a director bought ~$1.8M of stock on the open market in 2026 at the lows (code P, no 10b5-1 plan), per Autodesk’s Forms 4. That is the difference between management viewing a decline as an opportunity and management viewing it as weather.

Valuation & Market Data

Is the stock an ADR, MLP, or K-1 issuer? No. FACT: Trimble Inc. is a Delaware corporation with common stock listed on Nasdaq under TRMB (CUSIP 896239100, SEC file number 001-14845). Ordinary US common stock, Form 1099 reporting, no K-1, no ADR structure, no partnership units. One structural note: FACT — as a consequence of the delayed 2024 filings the company lost Form S-3 eligibility for twelve months, which ended in April 2026; the 10-K warned this “could make accessing the capital markets during this period more costly or less efficient.” That constraint has now lapsed.

Dividend policy? FACT: no dividend. Trailing annual dividend yield 0.0%; no dividends appear in the AZI price history’s dividend column. All capital return is via repurchase, at a stated “at least 1/3 of free cash flow.” INTERPRETATION: appropriate for the growth profile, and it means the entire shareholder-return case depends on the buyback being executed at sensible prices — which in FY2025 and Q1-2026 it was not.

How profitable is the business? FACT (FY2025): gross margin 69.1% GAAP / 71.5% non-GAAP; GAAP operating margin 16.5% ($592.0M); non-GAAP operating margin 27.5% ($988.1M); adjusted EBITDA $1.05B / 29.2%; net margin 11.8%; GAAP diluted EPS $1.76; non-GAAP diluted EPS $3.13. ROIC 6.92%; ROE 11.87%; ROA 4.51%. Segment operating margins: AECO 34.2%, Field Systems 31.1%, T&L 21.9%. INTERPRETATION: highly profitable at the operating line and unprofitable relative to its cost of capital — the two facts are reconciled by $5.2B of goodwill.

Is net income diverging from cash from operations? FACT: yes, in both directions, and both divergences are explained by the same non-recurring event. Cash-flow-to-net-income was 0.35x in FY2024 (net income $1,504.4M included a $1,687.9M non-cash divestiture gain, so earnings ran far ahead of cash) and 0.91x in FY2025 (net income $424.0M against CFO $386.2M, depressed by the $277.4M Ag tax payment). Working capital swung correspondingly: accounts payable and accruals +$327.8M in FY2024 and −$339.1M in FY2025, with income taxes payable falling $325.0M → $17.7M.

INTERPRETATION: neither year is representative and the normalized relationship is healthy. Normalizing FY2025 CFO to ~$693M against $424.0M of net income gives ~1.6x, consistent with the 2021 (1.52x) and 2023 (1.92x) readings and appropriate for a business with $199.9M of D&A, $146.5M of SBC and $970.7M of deferred revenue. There is no evidence of earnings quality deteriorating relative to cash. The caveat that must accompany this conclusion: both the revenue and the income-tax processes generating these figures carry an unremediated material weakness.

Valuation summary (context only — no price target). FACT at $52.91 (24 July 2026): market capitalisation ~$11,737M; net debt $1,178.7M; EV ~$12,916M. EV/TTM revenue 3.50x; EV/TTM gross profit 5.05x; EV/TTM EBITDA 15.1x; GAAP P/E ~27.7x; P/TBV not meaningful (tangible common equity −$327.6M). On FY2026 guidance: EV/EBITDA 11.2x, P/E 14.9x, headline FCF yield 7.0%, owner-FCF yield 5.8%. Own-history percentiles: P/E 21.0th, P/B 1.3rd, P/S 32.8th, composite 18.4th — versus Bentley 1.3rd, PTC 2.0th, Autodesk 3.0rd. Short-interest and detailed ownership data were not independently verified for this report.

Risks & Downside

What factors would cause the stock to decline? In descending order of assessed importance: (1) a fourth consecutive adverse ICFR opinion in the FY2026 10-K, or any restatement of FY2024/FY2025 revenue; (2) organic ARR growth decelerating through 10% (it fell from 14% in Q4-2025 to 12% in Q1-2026 against a 13% guide); (3) AECO segment margin remaining flat on organic growth, confirming that FY2025’s expansion was divestiture mix rather than leverage — AECO is ~62% of the sum-of-the-parts equity value; (4) evidence that AI compresses seat-based licensing — a disclosed seat decline or a shift in guidance language from “expanding TAM” to “transitioning pricing models”; (5) Field Systems’ low-to-mid single-digit organic guide proving to be the run-rate rather than conservatism; (6) the 2027 “3-4-30” targets being retired without a credible 2028 framework; (7) another large debt-funded acquisition at a Transporeon-like multiple; (8) a construction or freight downturn; (9) a goodwill impairment ($5,213.6M against $8,991.0M of assets); (10) a write-down of the $616.8M of Level 3 equity stakes.

Risk of a catastrophic loss? Low. FACT: leverage is 1.1x net-debt/EBITDA against a 2.5x target; the $1.25B five-year revolver entered 4 December 2025 is undrawn; principal debt is $1,400.0M; normalized FY2025 free cash flow was ~$668M against $25.3M of capex; revenue is diversified across three segments, multiple geographies and direct/distribution/OEM/enterprise channels with no material single-customer concentration; and each business pillar carries more than $230M of ARR. INTERPRETATION: the realistic bear outcome is a multi-year de-rating and dead money — the bear scenario above implies roughly −26% equity value from here — not impairment of the enterprise. The one path to something worse is escalation from an adverse control opinion to a material restatement, which is not the base case (KPMG’s financial-statement opinion is unqualified and no misstatement has been identified) but which cannot be excluded while the revenue-process controls remain ineffective into a fourth year.

Chance of a total loss? Negligible on any reasonable time horizon. The business generates roughly $680M of owner free cash flow against $1,178.7M of net debt — under one and a half years of owner cash flow — on a capital-light cost structure, with an undrawn revolver, no near-term maturity wall disclosed, three profitable segments and $970.7M of deferred revenue. Total loss would require simultaneous fraud, restatement and covenant failure. Nothing in the record supports that: the identified control deficiencies “did not result in a material misstatement,” and the financial statements carry a clean opinion.

Recent News & Events

Has the business environment changed recently? FACT: yes, in three respects, all named by management. (a) Geopolitical and trade: “the conflict we see in the Middle East and uncertainty around tariff policies,” plus “high oil prices, high memory prices” — cited as reasons for the cautious Field Systems guide. (b) Freight: T&L operates in “a constrained freight market” with “the macro environment remain[ing] challenged,” though “the North American market is beginning to show some signs of market recovery.” © AI: the most consequential change. Trimble launched SketchUp AI in Q4-2025 ($11.99/month credit-bundled add-on), announced an Anthropic Claude / SketchUp MCP integration in late April 2026, is “approaching our target to dedicate 10% of our development resources to an applied AI organization,” and states “the vast majority of new code is generated with AI tools.” Offsetting positives: infrastructure and data-centre construction demand, and a North American construction labour shortage of “hundreds of thousands of workers.”

Significant acquisitions? FACT: Document Crunch — agreement announced 2 April 2026, price undisclosed — AI-powered construction-contract and compliance analysis, to be bundled into Trimble Construction One and positioned as “a new AI-powered risk management category.” Management’s stated rationale rests on the industry’s dispute economics: “More than 80% of projects exceed budget… the average claim in North America tops $60 million.” Also FACT: a small Field Systems divestiture closed in Q1-2026, taking the running total to 13 acquisitions and 25 divestitures since 2020.

Change in accounting policies? FACT: no change in accounting policy — but a continuing failure in accounting controls. No new critical accounting policy or standard adoption of note is disclosed for FY2025 or Q1-2026. What has changed is the control and audit posture: KPMG replaced EY (auditor 1986–2025) in 2025 and issued an adverse ICFR opinion dated 2026-02-25; the material weaknesses in ITGCs, revenue and income taxes “exist as of January 2, 2026” and remained unremediated in the Q1-2026 10-Q. Separately, the OBBBA (signed 4 July 2025) permanently repealed the domestic R&D capitalization requirement, producing expected cash-tax reductions of ~$53M in 2025 and ~$53M in 2026; management believes it “will not have a material impact on our future effective income tax rate.”

Recent changes — new markets, facilities, management? FACT: new markets — Trimble Construction One launched in Asia-Pacific in Q1-2026; ProjectSight brought to Europe in recent quarters, with European AECO growth exceeding North America’s in Q1-2026. Products — SketchUp AI (Q4-2025); the Claude integration (April 2026); Field Systems extensibility to compact track loaders, dynamic-swing-boom excavators and ground-penetrating-radar-integrated asphalt compaction; FedRAMP certification in progress. Financing — a new $1.25B five-year unsecured revolver (4 December 2025, expandable to $1.75B, undrawn). Management and board — Phillip Sawarynski is CFO (David G. Barnes signed the May 2024 late-filing notification; no causal link to the control matters is disclosed and none is inferred); director James C. Dalton, a board member since 2020, is not standing for re-election at the 26 May 2026 annual meeting. Events — Trimble Dimensions user conference in Las Vegas, November 2026, at which management has suggested it may host the analyst community and possibly frame a post-2027 outlook.

APPENDIX B — Source Appendix

Trimble Inc. (NASDAQ: TRMB) · Report date: 2026-07-25 · CIK 0000864749 · CUSIP 896239100 · SEC file no. 001-14845

Sources are listed primary-first and are all publicly available. SEC documents were retrieved from the EDGAR archive and read in full; the trailing 60-month corpus comprises 591 enumerated filings. Accessed 2026-07-25 unless otherwise stated.

A. SEC filings — primary (EDGAR, https://www.sec.gov/cgi-bin/browse-edgar?action=getcompany&CIK=0000864749)

Corpus enumeration. SEC EDGAR full-filing history for CIK 0000864749 since 2021-07-01 returns 591 filings. Form census: 353 Form 4, 69 8-K, 67 Form 144, 15 10-Q, 13 DEFA14A, 12 Form 3, 8 Form 4/A, 6 Form 5, 6 SCHEDULE, 5 DEF 14A, 5 SD, 5 10-K, 4 NT, 4 ARS, 2 8-K/A, 2 424B5, 1 each 10-K/A, S-3ASR, S-8, FWP, 144/A. Documents 424B*, FWP and 144 are excluded from the review as structured-note and routine noise.

B. Earnings-call transcripts (ROIC.ai MCP, get_earnings_call_transcript)

Note: management commentary is treated throughout as hypothesis rather than evidence, and is validated against the filings above wherever the two can be compared. Non-earnings event transcripts (e.g. the J.P. Morgan 54th Annual Global Technology, Media and Communications Conference fireside chat, 2026-05-18) were identified in the news feed but not obtained; ROIC’s transcript coverage is earnings-call-centric.

C. Quantitative data services

D. Peer companies used for cross-read (public filings)

E. Analytical frameworks

• Bruce Greenwald & Judd Kahn, Competition Demystified — barriers to entry as the dominant consideration; the three genuine competitive-advantage types (supply/cost, demand/customer captivity, economies of scale plus captivity); the market-share-stability and ROIC tests; earnings power value versus asset value. Applied in the Competitive Position section to name the moat type by segment, and to record that the market-share-stability test cannot be run on Trimble’s disclosure because no share, seat, retention or churn data is published.
• Edward Chancellor (ed.), Capital Returns (Marathon Asset Management) — supply-side capital-cycle analysis and the asset-growth anomaly, i.e. that high returns attract capital and then mean-revert. Applied in the Industry Dynamics and Capital Allocation sections to the goodwill-up / ROIC-down pattern: goodwill grew from $3,540.0M (2018) to $5,239.7M (2025), +48%, while ROIC fell from 9.24% (2021) to 6.92% (2025).
• A methodological caution on own-history valuation percentiles: where GAAP earnings or book value are distorted by a large one-time item, the P/E and P/B percentiles must be discounted and P/S read instead. That caution is applied directly here — Trimble’s 1.3rd-percentile price-to-book reading is an artifact of the $1,687.9M FY2024 Ag divestiture gain inflating retained earnings by roughly $1.3B.
• A methodological caution on margin analysis: where consolidated margin is expanding, the segment-level margin series must be pulled, because a flat largest segment means the consolidated figure is a weighting statistic rather than evidence of operating leverage. That is precisely what the Financial Quality section found in AECO’s flat 34.2%.

F. Data gaps, limitations and disclosures

1. This article relies entirely on publicly available sources, all of which are listed above. No non-public or proprietary information was used.
2. No position. The author holds no position in TRMB and nothing here should be read as disclosing or implying one.
3. Current share count is not established in a filed document. AZI implies 221.8M shares as of 2026-07-24 (market capitalisation $11,737.2M ÷ $52.91); Q1-2026 diluted weighted-average shares were 236.9M; $317M was repurchased in Q1-2026 with $608M of authorization remaining. Valuation is run on the AZI market capitalisation with the 236.9M-share alternative disclosed in the Valuation section (EV $13,713M, EV/FY26E EBITDA 11.9x, owner-FCF yield 5.4%). Direction of every conclusion is unchanged. Resolves with the Q2-2026 10-Q.
4. The catalyst for the 2026-01-14→20 decline is not in the public record. No 8-K, guidance change or company disclosure exists in the window; the only 8-K between 2025-11-05 and 2026-02-10 is the 2025-12-05 revolver filing. Roughly 6pp of the 12.8% decline is attributable to a 1.26 beta on the concurrent 5.1% IGV move; the residual is unexplained and is not assigned a cause.
5. No retention, churn or seat disclosure exists. Net revenue retention, gross retention, churn and seat/customer counts appear in none of the FY2025 10-K, the Q1-2026 10-Q, the 2026 DEF 14A or either recent earnings call. Switching costs therefore cannot be measured from outside, and the Greenwald market-share-stability test cannot be run.
6. Form 4 parsing coverage. 122 of the 353 enumerated Form 4 filings were downloaded (all those filed 2025-01-01 onward); 46 contained machine-parsed non-derivative (Table I) transactions and the remainder are predominantly derivative-only grant filings served as SEC XSL-rendered HTML. The conclusion that matters — zero code-P open-market purchases — is robust, because a code-P purchase is by definition a Table I non-derivative transaction and none appears in any parsed filing.
7. The CEO’s 11,897-share direct holding is a secondary source (press reporting of the 2026-04-14 Form 4) and was not independently re-derived from the filing itself.
8. Non-earnings event transcripts were not obtained, including the 2026-05-18 J.P. Morgan technology-conference fireside chat identified in the news feed. ROIC’s transcript coverage is earnings-call-centric.
9. Short interest and detailed institutional ownership were not independently verified. 13F-derived items in the news feed (CalPERS, BNY Mellon, Barclays and others) were read as flow colour only and are not relied upon for any conclusion.
10. Third-party data authority. ROIC.ai, AZI and FactorsToday are third-party aggregated or statistically estimated data and are not primary. EDGAR and the 10-K/10-Q remain primary throughout; where a computed figure differs from the filing (e.g. FY2025 operating income), the filing governs and the difference is disclosed. No analyst price target, consensus estimate or third-party valuation model has been adopted as this article’s view. The overriding caveat on all of it: Trimble’s own revenue and income-tax reporting processes carry an unremediated material weakness and an adverse ICFR opinion as of 2026-01-02, so every figure derived from them — including the ARR and segment-margin series on which the investment case turns — carries an audit-quality qualification that no amount of third-party cross-checking can remove.

Compiled 2026-07-25. Every non-obvious factual claim in this article traces to a source listed above. This article is general information and independent opinion, not investment advice, and not a recommendation to buy or sell any security. The author holds no position in the securities discussed.