Okta, Inc. (NASDAQ: OKTA) — The Independent Identity Platform: Healed, Profitable, and Boxed In by Microsoft

An independent equity research note. Report date: 2026-06-19. Fiscal year ends January 31; “FY26” = year ended 31-Jan-2026. All dollar figures USD. Primary sources: SEC filings (10-K FY22–FY26, Q1-FY27 10-Q, DEF 14A 2026, Form 4 corpus, earnings 8-Ks), the Q1-FY27 earnings call (28-May-2026), and third-party market/factor data, each cited inline.

⚡ Claude’s Take

This block is the author’s own independent opinion and general information only — not investment advice. The analytical body that follows is deliberately position-free and carries no price target; this block is the one exception.

Verdict: HOLD at ~$118 — a de-risked, fairly-priced quality compounder, not a bargain. Accumulate on weakness toward the high-$80s–$95 zone (~16–18x EV/FCF); not a short. Conviction: medium. Tag: “The Switzerland of Identity — healed, profitable, and standing in Microsoft’s doorway.”

Okta is a genuinely repaired business that the market has already partly re-embraced. The thesis is not the moat and it is not the growth — both are contested. The thesis is valuation against a confirmed inflection: a net-cash ($2.2B), GAAP-profitable, ~30%-FCF-margin identity platform whose revenue growth decelerated for five straight years (56%→43%→22%→15%→12%) but whose net-retention rate just inflected back up to 107% and whose operating margin swung from −20% to +5% GAAP / ~25% non-GAAP in two years. At ~$118 it trades at ~6.3x EV/sales, ~21x EV/FCF, ~31x forward non-GAAP P/E — the 36th percentile of its own decade of sales multiples, having round-tripped from a $294 ZIRP bubble to a $44 trough to here. A reverse-DCF says the market is pricing ~5% perpetual FCF growth: neither a growth stock nor a melting ice cube. That is roughly fair, with mildly favorable asymmetry.

What the market is pricing correctly: the deceleration is real and Microsoft Entra ID — free-to-near-free inside the M365 E5 bundle — is a permanent ceiling on Okta’s new-logo economics, which is exactly why the growth rate halved and halved again. What it may be under-pricing: Okta is uniquely the independent, multi-cloud identity layer at a moment when AI agents and non-human identities are exploding the number of “things” that need to authenticate — a category Okta has a credible right to win precisely because it is not Microsoft, Google, or AWS. That option is real but unmonetized; management itself calls it “not material… heavily discounted in the guide.” The framing from the tape supports patience over chase: factor data shows an abandoned former-growth name (5-year return −13%/yr, −83% max drawdown, momentum loading negative) at an early, unconfirmed FCF/NRR inflection that the momentum crowd has not yet re-bought. You are not paying for a crowd. Bull flip: NRR holds >107% and agentic/machine identity becomes a named, material revenue line → re-rate toward ~9–10x sales. Bear flip: a third security breach, or NRR rolls back over as Entra accelerates share capture → terminal growth re-rates to mid-single digits and the 31x forward multiple compresses. At today’s price both are roughly balanced; the attractive entry is a growth-scare drawdown, not the post-beat high.

📈 Stock Price Action — Five-Year Event Map

Okta has completed a violent round-trip. From a February-2021 all-time high of ~$294, the stock collapsed ~85% to a ~$44 trough on 4-Nov-2022, ground sideways-to-up through a profitability repair, then spiked ~30% in a single session on the 28-May-2026 Q1-FY27 print to a 52-week high of $142.35 (1-Jun-2026) before pulling back ~17% to $117.81 (18-Jun-2026 close). It remains ~60% below its 2021 peak five years on — a stock that has compounded negatively over five years even after the recent rally (annualized 5-yr total return ≈ −13%).

Cycle narrative. (1–2) Okta was a marquee ZIRP-era SaaS name that round-tripped the entire bubble; the de-rating was a multiple event (sales multiple fell from ~30–40x to ~6x) layered on a genuine fundamental deceleration. (3–4) Two security incidents — Lapsus$ (Jan-2022) and the far more damaging Oct-2023 support-system breach in which threat actors accessed customer HAR files and session tokens — were uniquely corrosive because Okta sells trust; they accelerated the new-logo slowdown and forced the “Secure Identity Commitment.” (5) The 2024–25 recovery was earned on margins, not growth: opex was held flat while revenue grew, producing the first GAAP profits and a re-rating off the trough. (6) The 28-May-2026 print was the first genuine growth positive in years — NRR ticking back up to 107% after a multi-year decline — and the stock gapped ~30%. (7) The subsequent pullback is positioning/macro, not news. All price moves are facts from the price series; attributed causes are interpretation cross-referenced to filing/earnings dates and the news feed.

1. Executive Summary

Okta is the leading independent identity-and-access-management (IAM) platform — the neutral “Switzerland” alternative to identity bundled inside a hyperscaler’s stack. It runs two clouds: the Okta Platform (workforce identity — single sign-on, adaptive MFA, Universal Directory, lifecycle management, governance, privileged access) and the Auth0 Platform (developer-first customer identity / CIAM, acquired for ~$6.5B of stock in 2021). The model is high quality: ~98% subscription revenue, ~80% subscription gross margin, asset-light (capex <0.5% of revenue), >20,000 customers with no concentration, and a $1.7B deferred-revenue float.

The investment debate has three moving parts. First, the business has healed. After a brutal 2022–24 — a SaaS de-rating, Auth0 integration pain, and two security breaches — Okta crossed into GAAP profitability (FY25 first positive net income; FY26 net income $235M, GAAP EPS $1.31, non-GAAP EPS $3.50), expanded operating margin from −20% to +5% GAAP / ~25% non-GAAP in two years on flat opex, and generated $875M of free cash flow (30% margin). It is net-cash $2.2B, is cash-settling its last convertible, and has begun a first-ever $1B buyback. Second, growth has decelerated to a low-double-digit base — revenue +11.8% in FY26, guided +9–10% for FY27 — and the central question is whether ~10% is the durable floor or the next step down. The NRR inflection to 107% in Q1-FY27 (after falling from ~123%) is the most important single data point in the bull case. Third, the moat is real but narrowing. Okta’s switching costs and 7,000+ pre-built integrations make the installed base genuinely sticky (106% NRR survived a breach and a recession), but Microsoft Entra ID’s bundling economics are competing away the share of new growth — the deceleration is the evidence.

Valuation sits at fair-to-mildly-cheap: ~6.3x EV/sales, ~21x EV/FCF, ~31x forward non-GAAP P/E, a wide discount to faster-growing peers (CRWD ~26x sales, NET ~33x) that is partly deserved (half their growth) but extreme given Okta’s superior FCF margin and net-cash balance sheet. A reverse-DCF implies the market prices only ~5% perpetual FCF growth. The quality is now visible and largely in the price; the embedded expectations are modest. This is a business to own at a growth-scare discount, not to chase at a post-beat high. No recommendation or price target follows in the body below.

2. Business Overview

What Okta does. Okta sells cloud-delivered identity — the software that decides who (or what) is allowed to access which application, system, or piece of data, and proves it. Identity is the control plane of “zero-trust” security: as enterprises abandoned the network perimeter for cloud apps and remote work, the question “is this login legitimate?” became the primary security decision, and Okta is the largest pure-play vendor answering it. The company’s mission framing — “the world’s identity company” — maps onto two product clouds that go to two different buyers.

The Okta Platform (Workforce Identity). Sold to IT and security teams to manage employee and contractor access. Core SKUs: Single Sign-On (one login to thousands of apps), Adaptive Multi-Factor Authentication (risk-based step-up auth), Universal Directory (a cloud system-of-record for user/device/app profiles), and Lifecycle Management (automated provisioning/deprovisioning as employees join, move, and leave). Layered on top is a growing security-and-governance portfolio that is the company’s expansion engine: Okta Identity Governance (OIG) — access certification/requests, competing with SailPoint; Okta Privileged Access (OPA) — vaulting/just-in-time access to servers, competing with CyberArk; Identity Threat Protection (ITP) with Okta AI — continuous post-authentication risk evaluation; Identity Security Posture Management (ISPM) (from the Spera acquisition); and Device Access. This “platform beyond SSO” cross-sell is the heart of the forward growth narrative.

The Auth0 Platform (Customer Identity / CIAM). Sold to developers who need to embed login, signup, and authorization into the apps and websites their own customers use. Acquired in 2021, Auth0 is API-first and developer-led — a different motion, buyer, and competitive set (AWS Cognito, Microsoft, Google, and CIAM upstarts Transmit/Stytch). Auth0 also anchors Okta’s newest thrust: Auth for GenAI / “Auth0 for AI Agents” (GA April-2026), authentication and fine-grained authorization for AI agents acting on a user’s behalf.

Business model. Revenue is ~98% subscription (term licenses billed annually in advance), with a small, deliberately-shrinking professional-services line that Okta is handing to global system integrators (a ~1-point FY27 revenue headwind, but margin-accretive because services run at a gross loss). Pricing is a mix of per-user (seat) and per-SKU/edition — customers land on SSO/MFA and expand by adding users, products, and editions. This produces the classic SaaS land-and-expand dynamic measured by dollar-based net retention (NRR) and remaining performance obligations (RPO/cRPO). FY26 revenue was $2,919M (subscription ~$2,855M; services ~$64M); cRPO ~$2.5B; total RPO ~$4.8B. [FACT — FY26 10-K; Q1-FY27 10-Q/transcript]

Customers and end-markets. >20,000 customers across every vertical and geography; ~5,100 customers with >$100k annual contract value (up from ~4,800 and ~4,485 the prior two years), and ~85% of ACV from these large customers. ~20% of revenue is international. There is no customer concentration (no single customer is material). End-markets skew enterprise and mid-market, with a growing, strategically-important US public-sector business riding FedRAMP authorizations. [FACT — FY26 10-K]

Verdict. A high-quality, highly-recurring, asset-light subscription model with genuine scale (>$2.9B revenue, the pure-play identity leader) and clean unit economics. The model itself is not in question; the growth rate and the durability of the moat against Microsoft are. Those are.

3. Industry Dynamics

Structure. Identity is one sub-pool of the broader cybersecurity market, which is structurally attractive: spending is non-discretionary (a board-level, regulation-and-breach-driven priority), recurring (subscription/SaaS), high-margin (software economics), and growing faster than IT budgets. Within cyber, identity is arguably the most strategic layer — “identity is the new perimeter” is not marketing; in a cloud/remote world, compromised credentials are the dominant attack vector, so identity controls sit upstream of everything else. That is the good news. The bad news is structural too: identity is also the cyber sub-pool most exposed to Microsoft, which gives away capable identity (Entra ID) inside bundles enterprises already buy.

Market size and the TAM honesty gap. Okta markets a “$80B+ TAM” by summing workforce IAM, CIAM, IGA, PAM, and identity-security adjacencies. The directly-addressable, like-for-like IAM/PAM/IGA market that Okta actually competes for is more credibly ~$24–26B today, growing to ~$40–42B by 2030 (~10% CAGR) per third-party sizing. The gap matters: Okta’s “$80B+” includes adjacencies it barely monetizes (machine identity, posture management) and double-counts overlap. The realistic market grows at roughly the rate Okta is now growing — meaning the company is, at best, holding share in a ~10% market, not taking it. [FACT/INTERPRETATION — 10-K TAM claim vs third-party market sizing]

Sub-markets. (a) Workforce IAM — Okta’s core; the most contested vs Microsoft. (b) CIAM — Auth0’s domain; developer-led, fragmented, faster-growing, but commoditizing at the low end. © Identity Governance (IGA) — SailPoint (re-IPO’d 2025) is the incumbent leader; Okta’s OIG is a credible challenger and a key cross-sell. (d) Privileged Access (PAM) — CyberArk (being acquired by Palo Alto Networks for ~$25B) dominates; Okta’s OPA is early. (e) Machine / non-human identity — the fastest-growing frontier (AI agents, service accounts, workloads), today small and unmonetized, and the strategic battleground.

Regulatory and structural tailwinds. US federal zero-trust mandates (OMB M-22-09), NIST frameworks, cyber-insurance MFA requirements, and a steady drumbeat of credential-based breaches all push identity spend structurally higher and favor scaled, certified platforms (FedRAMP High is a real barrier-to-entry in public sector). These are genuine tailwinds for the category; they do not resolve which vendor captures the spend.

Capital-cycle read (Marathon lens). Cybersecurity is late in its capital cycle: VC funding has poured in (~$14B/yr at the peak), and a wave of mega-M&A (Google–Wiz ~$32B, Cisco–Splunk ~$28B, Palo Alto–CyberArk ~$25B) signals incumbents buying growth at high multiples — a classic late-cycle, deteriorating-supply-side signal. For Okta specifically, the relevant supply-side pressure is not a flood of new identity startups (that has ebbed) but the near-zero marginal price of Microsoft’s bundled supply — a structural, not cyclical, headwind.

Verdict: a structurally GOOD industry — non-discretionary, recurring, high-margin, regulation-driven — but identity is the cyber sub-pool with the most adverse supply dynamics, because the largest competitor’s marginal price is ~zero. The category will grow; the open question is whether Okta grows with it or slower than it.

4. Competitive Position

The moat — name the mechanism. In Greenwald’s taxonomy, Okta’s advantage is primarily demand-side customer captivity (switching costs), reinforced by a niche economies-of-scale-in-neutrality effect from network/integration breadth. Once an enterprise wires Okta into its identity fabric — every app federated to Okta SSO, every joiner/mover/leaver workflow automated through Lifecycle Management, MFA policies tuned, the Okta directory as the system-of-record — ripping it out means re-integrating thousands of applications and re-architecting access for the whole workforce. That is expensive, risky, and rarely worth it. The proof is in the retention: net retention held ~106–107% straight through a recession, an Auth0 integration mess, and two security breaches — a business with weak captivity would have bled. The second mechanism is the Okta Integration Network (7,000+ pre-built app integrations): because Okta is independent, ISVs build to it, and the breadth of pre-built connectors is itself a switching cost and a reason to choose the neutral platform over a cloud-locked one. [FACT — NRR history per filings/transcripts; OIN integration count per 10-K]

The threat — Microsoft Entra ID. Okta’s 10-K names Microsoft as a principal competitor, and it is the bear case. Microsoft Entra ID (formerly Azure AD) ships inside M365/Entra suites that enterprises already license; at the E5 tier, capable SSO/MFA/conditional-access is effectively free at the margin. A CIO standardizing on Microsoft can often get “good-enough” identity for “free” rather than pay Okta a standalone subscription. Microsoft’s security business (>$20B revenue) can cross-subsidize identity indefinitely. This is why Okta’s growth halved and halved again: the moat protects the installed base and the category position, but Entra’s bundling economics are competing away the incremental/new-logo share. The deceleration from 56% to ~10% is, in large part, the sound of Microsoft setting the price of new identity. [INTERPRETATION — supported by the growth trajectory and the 10-K competition disclosure]

Where Okta still wins. (1) Heterogeneous / multi-cloud / Microsoft-skeptical enterprises that refuse to hand their identity control plane to the same vendor that runs their OS, cloud, and email — neutrality is a feature, especially for security-conscious and regulated buyers. (2) Best-of-breed depth — Okta’s workforce capabilities, integration breadth, and CIAM (Auth0) are generally regarded as more complete than Microsoft’s for complex needs. (3) Developers (Auth0) — a different buyer Microsoft serves less well. (4) The agentic/non-human frontier — where Okta’s independence is, again, a right-to-win.

Head-to-head map. Microsoft — principal competitor, bundling-driven, owns the low end and the Microsoft-standardized majority. Ping Identity / ForgeRock (merged, Thoma Bravo–owned, private) — legacy enterprise, less cloud-native. CyberArk (→ Palo Alto) — PAM leader, adjacent. SailPoint — IGA leader, adjacent and overlapping with OIG. AWS / Google IAM — cloud-native identity for their own stacks. CIAM upstarts (Cognito, Transmit, Stytch) — pressure Auth0 at the low end. Okta is the largest independent generalist; everyone else is either a hyperscaler (bundled) or a point-solution (narrower).

Verdict: a REAL but NARROWING moat. The switching-cost captivity is genuine and durable for the installed base (the retention proves it), and the neutrality advantage is structural for a meaningful segment of the market. But the moat defends the base and the category, not the share of growth — Microsoft’s bundling is the permanent ceiling that turned a hypergrowth company into a ~10% grower. The correct characterization is “a sticky, defensible franchise that has lost pricing power over new demand,” not “a widening-moat compounder.” Whether ~10% is the floor depends on whether the next frontier (machine/agent identity) lets Okta re-expand the pie faster than Microsoft can bundle it.

5. Growth History and Forward Opportunities

The deceleration is the dominant fact. Revenue growth: FY22 +56% → FY23 +43% → FY24 +22% → FY25 +15% → FY26 +11.8%, guided +9–10% for FY27. This is one of the cleanest deceleration curves in software — five consecutive years of step-downs. Some of FY22–23’s growth was Auth0 (acquired May-2021, inorganic), and some of the deceleration is the law of large numbers, but the larger driver is competitive (Microsoft) plus self-inflicted (the breaches and a 2022–23 go-to-market reorganization that split the Okta and Auth0 sales forces and disrupted productivity). [FACT — IS history]

Net retention — the swing variable. Dollar-based NRR fell from ~123% (FY22 peak) through the low-110s and into the ~106% range by FY25–early FY26 as upsell slowed and seat counts came under macro pressure. In Q1-FY27 it inflected back up to 107% — the first uptick in years, and the single most important bull data point. Management attributes it to the security/governance cross-sell (OIG, OPA, ITP) and rising strategic relevance from the AI narrative. Caveat: a one-quarter move from 106% to 107% is an inflection, not a trend; durability above 107% is the falsification test. [FACT — NRR figures; INTERPRETATION — durability]

Forward opportunities — ranked by credibility.

1. Security/governance cross-sell into the installed base (highest credibility). Selling OIG, OPA, ITP, and ISPM to >20,000 existing customers is the proven, in-flight engine — it is what just nudged NRR up. This is real and quantifiable.
2. Public sector / FedRAMP High (high credibility). A structurally-advantaged, certification-gated market where Okta’s neutrality and accreditations are a moat; growing.
3. Auth0 / CIAM re-acceleration (medium). After integration pain, CIAM is stabilizing; Auth for GenAI gives it a new wedge.
4. Agentic AI / non-human identity (real option, low near-term credibility). The biggest potential TAM expansion — every AI agent and workload needs to authenticate and be authorized, and Okta’s independence is a genuine right-to-win. Partners include ServiceNow, Google, AWS, OpenAI, and Anthropic. But management is unusually candid that this is “not material” today, “a pipeline, not in the guide,” and that monetization is unsolved (“we don’t get paid for pipeline”). It is a 2027–2029 call option, not an FY27 driver — and Microsoft can bundle “Entra Agent ID” just as it bundled workforce identity.
5. International (steady). ~20% of revenue with room to grow, but no step-change.

Quality of growth. What growth Okta has is high-quality in character (recurring, sticky, expansion-led, software-margin) but low-quality in trajectory (decelerating, and increasingly dependent on selling more to existing customers rather than winning the category’s new demand). The bull needs the cross-sell engine plus the agent option to put a floor under and then re-expand the growth rate; the bear needs only for Microsoft to keep doing what it is doing.

Verdict: a low-quality (decelerating) growth history now stabilizing into modest-but-profitable growth, with a real — but unmonetized and contested — AI/machine-identity option on top. The NRR inflection is the evidence the floor may be forming; the agent narrative is the evidence the ceiling could lift. Neither is yet proven in the numbers. This is the fulcrum of the entire thesis.

6. Financial Quality

The headline: operating leverage, and it is real. Okta’s two-year P&L is a textbook fixed-cost-absorption story. Against +$656M of incremental revenue (FY24→FY26), the company held operating expenses roughly flat — R&D actually fell ($656M→$639M), and S&M and G&A grew far slower than revenue — producing incremental operating margins of ~70%+ and swinging GAAP operating margin from −20.3% (FY24) → −2.4% (FY25) → +5.2% (FY26). GAAP net income went −$355M → +$28M (first-ever positive) → +$235M, diluted EPS −$2.17 → +$0.16 → +$1.31. Gross margin climbed from ~71% to 77.4% GAAP (~80%+ subscription). Non-GAAP operating margin: ~14% (FY24) → 22% (FY25) → ~24.5% (FY26), guided 25–26% FY27. Non-GAAP EPS: $2.81 (FY25) → $3.50 (FY26), guided $3.74–$3.82 (FY27). [FACT — FY26 10-K; Q4-FY26 8-K EX-99.1]

Free cash flow is genuine and large — but read the composition. FCF: $503M (FY24) → $742M (FY25) → $875M (FY26), a 30% margin, on essentially zero capex (asset-light). FY27 guide is a 27–28% FCF margin (~$870–900M). This is the strongest single feature of the financial profile. Two important caveats on quality: (1) ~60% of reported FCF is the stock-based-compensation add-back. SBC was $544M in FY26; FCF excluding SBC is only ~$330M (~11% “owner” cash margin). SBC is a real economic cost (it dilutes shareholders), so the honest “cash earnings to owners” figure is closer to ~11% than 30%. (2) FCF is flattered by the deferred-revenue float (annual upfront billing); as growth slows, the float’s tailwind to FCF fades. Both caveats temper, but do not negate, a real cash-generative inflection — and crucially, SBC is declining as a share of revenue (30%→18.6%), so the gap between reported and owner FCF is narrowing.

Quality-of-earnings flags. (1) Interest income is a large, non-operating, rate-sensitive prop on reported profit. FY26 pretax income was $255M, of which ~$110M was interest income on the ~$2.6B cash pile — i.e., ~42% of pretax income is not from the identity business. Operating income alone is ~$150M. As rates fall and the cash pile shrinks (buyback + convert settlement), this tailwind reverses — management already flags a ~1-point FY27 FCF-margin headwind from it. (2) The GAAP tax rate is ~7.8%, far below statutory, driven by NOLs/valuation-allowance release; this flatters GAAP net income by ~$30M+ and will normalize upward (the non-GAAP rate is already set at 21% for FY27). (3) Billings/RPO are tracking ahead of revenue — RPO +14.5% and cRPO +12% vs revenue +11.8% — a positive QoE signal that bookings are not deteriorating faster than the P&L. [FACT — 10-K/10-Q; transcript]

Returns on capital. Reported ROIC/ROE are low (~2% and low-single-digits) and not the right lens, because the balance sheet is dominated by $5.49B of Auth0 goodwill (59% of equity) — i.e., returns are depressed by the price paid for an acquisition, not by the economics of the underlying business. The underlying operating business is capital-light and, on an unlevered cash basis, generates high returns on the tangible capital actually employed. The honest read: the acquired capital (Auth0) has not yet cleared its cost (sub-2% ROIC on the deal), but the organic business has excellent incremental economics. [INTERPRETATION — goodwill-adjusted]

Balance sheet: a fortress. $2.59B cash + short-term investments; total debt just $411M ($350M 2026 convertible + $61M leases); net cash ~+$2.18B. The 2026 convert is being cash-settled in June-2026 (eliminating that dilution source). Current ratio 1.43; deferred revenue $1.7B. Dilution has decelerated sharply: diluted shares 127M (FY21) → 179M (FY26), but the FY26 increase was only ~2.4%, and the new buyback is now reducing share count. [FACT — Q1-FY27 10-Q]

Verdict: economics improve emphatically with scale — the operating-leverage story is real and the FCF generation is large — BUT the quality is now substantially in the price, and headline profitability is flattered by (a) interest income, (b) a sub-statutory tax rate, and © the SBC add-back in FCF. Strip those and you have a ~$330M owner-FCF business growing ~10% — good, not spectacular. The trajectory (margin still expanding, SBC still falling, dilution reversing) is the encouraging part.

7. Capital Allocation

The defining decision — and the original sin — is Auth0. In May-2021, at the bubble peak, Okta acquired Auth0 for ~$6.5B in an all-stock deal (~$5,671M booked) — roughly 13x forward sales, financed by printing Okta equity at >$250/share (the stock is now ~$118). With hindsight, Okta paid a bubble multiple in a bubble currency for an asset whose integration then stumbled. It left $5.49B of goodwill (never impaired, but “not impaired” ≠ “value created”), and because Okta refuses to break out Auth0/CIAM revenue, the deal’s ROIC is unverifiable from the outside — a genuine disclosure gap and a fair criticism. The charitable read: Auth0 bought Okta the entire developer/CIAM franchise and the technical foundation for Auth for GenAI; the skeptical read: it was an expensive, dilutive deal whose returns management won’t let investors measure. Both are true. [FACT — deal terms; OPEN QUESTION — Auth0 standalone economics]

Everything since Auth0 has been materially better disciplined:

• SBC discipline. SBC cut to $544M / 18.6% of revenue (FY26) from 30% — the lowest of the OKTA/CRWD (~23%)/ZS (~25%) identity-adjacent cohort. Net dilution fell to ~2.4%/yr.
• First-ever capital return. A $1B buyback authorization: ~$79M repurchased in FY26, and ~$241M (≈3M shares) already in Q1-FY27 — notably near the lows, ~$680M remaining. Okta is the only one of the three independent identity peers running a real buyback, funded comfortably from FCF and the cash pile.
• De-leveraging / de-dilution. Cash-settling the $350M 2026 convertible removes a dilution overhang.
• R&D/S&M intensity normalizing. S&M remains the largest cost (sales-intensive enterprise software) but is growing far slower than revenue; R&D held flat in absolute dollars. This restraint is exactly what produced the margin inflection.

Incentive alignment — a management-certified read. The 2026 proxy shows the FY26 annual bonus paid on 60% Revenue / 40% non-GAAP Operating Income, funded at 98.7% of target; the CRO is on a Net ARR commission (paid 150% — net ARR beat); and long-term incentives are relative-TSR performance shares (FY26 achievement 114%, capped at 100%). The plan pays squarely on profitable growth and relative shareholder return — not vanity metrics — and the near-target funding is consistent with the operating turnaround being genuine rather than narrative. This is a credible alignment signal. [FACT — DEF 14A 2026]

Governance overhang — dual class. Founders/insiders control ~31.6% of votes on ~4.9% of economics (CEO McKinnon ~26.7% of votes on ~3.9% economic) via 10-vote Class B shares. The structure entrenches the founders and warrants a governance discount, though Class B converts over time and co-founder Kerrest has stepped back to a disengaged Executive Vice Chair role. There is no activist lever here.

Verdict: MIXED but IMPROVING to GOOD. The Auth0 deal was a value-destructive, bubble-priced, dilutive acquisition whose returns remain unmeasurable — a real black mark. But the post-2022 record is genuinely good: SBC discipline, the first buyback (executed counter-cyclically near lows), convert settlement, and an incentive plan that pays on profitable growth. The trajectory of capital allocation is clearly positive; the legacy of the one big swing is the drag.

8. Changes and Headwinds — Last Two Years

The two-year arc is a credible recovery from a trust crisis. The defining events:

• Security breaches and the “Secure Identity Commitment.” The Jan-2022 Lapsus$ intrusion (via a third-party support contractor) and the far more damaging Oct-2023 support-system breach — in which threat actors accessed the customer-support case-management system and customers’ uploaded HAR files containing session tokens — were uniquely corrosive for an identity-and-trust vendor. Okta responded with a public “Secure Identity Commitment,” security re-architecture, and leadership focus on its own security posture. The related securities class action settled for $60M (D&O-funded) and was dismissed Nov-2024. The incidents are contained, but they are a permanent line in the bear case: the company whose product is trust has been breached twice. [FACT — 8-Ks, 10-K legal proceedings]
• Strategy pivot: growth-at-all-costs → profitable growth. Restructurings/RIFs, opex discipline, and the GTM re-org (separating then re-stabilizing the Okta and Auth0 sales motions) drove the margin inflection. This is the source of the entire 2024–26 re-rating.
• The agentic-AI / non-human-identity pivot. “Okta for AI Agents” and “Auth0 for AI Agents” (GA April-2026), with hyperscaler and AI-lab partnerships (ServiceNow, Google, AWS, OpenAI, Anthropic). Strategically central, financially immaterial today.
• Go-to-market shift to GSIs. Handing professional services to global system integrators — a ~1-point FY27 revenue headwind but margin-accretive and a sign of platform maturity.
• Public-sector / FedRAMP High push and product breadth (OIG, OPA, ITP, ISPM via Spera) expanding the cross-sell surface.
• Q1-FY27 inflection (28-May-2026). Revenue +12%, NRR up to 107%, guide raised — the first growth-positive print in years; stock to a 52-week high. Sell-side price targets were raised to a ~$100–$150 cluster, but the ratings mix is telling: a wall of Overweight/Buy alongside several pointed Neutral/Equal-Weight/Sector-Perform holds (Wells Fargo, Scotiabank, Susquehanna, Piper) — i.e., the Street reads it as “fair, improving, not a screaming buy,” consistent with this note. [FACT — news/analyst feed, 28-May to 09-Jun-2026]

Verdict: NET STRENGTHEN. The breaches were a real wound but are contained and litigated-out; everything else over two years moved the thesis the right way — profitability, retention inflection, capital return, and a credible (if unmonetized) AI strategy. The headwinds that remain are structural (Microsoft, deceleration), not fresh deterioration.

9. Risk Analysis

Catastrophic-loss assessment. A total loss is highly unlikely — net cash $2.2B, $875M FCF, GAAP-profitable, no refinancing risk. The realistic severe downside is fundamental erosion: a multi-year grind where Microsoft commoditizes new demand, NRR slips back below 100%, growth fades to low-single-digits, and the multiple compresses toward a no-growth software level (~3–4x sales), implying meaningful (but not catastrophic) capital loss from here. The single event that could step-change the downside is a third major breach.

10. Valuation Discussion (Embedded Expectations)

Where the stock trades (live, $117.81; ~179.3M diluted shares). Market cap ~$21.1B; net cash ~+$2.18B; EV ≈ $18.9B. On that base:

• EV/TTM sales ≈ 6.3x (TTM revenue ~$3.0B); EV/FY27E sales ≈ 5.9x.
• EV/TTM FCF ≈ 21.6x; EV/FY27E FCF ≈ ~21x (27–28% margin on ~$3.2B).
• GAAP P/E ≈ 90x (distorted by the low tax rate and amortization); non-GAAP P/E ≈ 33.7x trailing / ~31x forward ($3.50 → $3.74–$3.82).
• AZI own-history percentile: P/S at the 36th percentile — a de-rating from the 22–40x bubble-era sales multiple to ~6–7x. (Ignore the P/B percentile — book is goodwill-distorted and near-zero; meaningless.)

Cross-sectional comp set.

Okta is the cohort’s slowest grower but also, with ZS, the cheapest FCF-generative identity-security platform — and uniquely net-cash, GAAP-profitable, and buying back stock. The ~4–5x-sales discount to CRWD/NET is partly deserved (Okta grows at roughly half their rate) but extreme given Okta’s superior ~30% FCF margin and fortress balance sheet. The market is paying up for growth (CRWD/NET) and discounting Okta’s profitability — a defensible stance if you believe Microsoft caps Okta’s terminal growth, and an over-reaction if you believe ~10% is the durable floor.

Reverse-DCF / embedded expectations. At a ~10% discount rate, $18.9B EV against ~$875–900M FCF implies the market is underwriting only ~5% perpetual FCF growth. Put differently: the price requires Okta merely to hold low-double-digit growth fading to mid-single digits — not to re-accelerate. It is priced as neither a growth stock nor a terminal-decliner. A two-stage 10-year DCF brackets the value:

• Base (~$20.0B EV, ≈ current): ~9–10% revenue growth fading to ~5%, FCF margin holding ~28–30%. The market’s current view.
• Bull (~$34.8B EV, +~85%): NRR durably >107%, cross-sell + agentic identity re-accelerate revenue to low-teens, FCF margin to ~33%. Requires the moat to re-widen via the AI frontier.
• Bear (~$12.3B EV, −~35%): Microsoft grinds growth to mid-single digits, NRR drifts toward 100%, multiple compresses to a no-growth ~3–4x sales.

Embedded-expectations conclusion: at ~21x EV/FCF and ~5% implied perpetual growth, the bar is low — undemanding rather than cheap. You are not paying for re-acceleration; you are paying for stability. The asymmetry is mildly favorable (bull +85% vs bear −35%) conditional on the deceleration having found its floor — which is precisely the contested point. No price target is set.

11. Variant Perception

Consensus belief. A formerly-hypergrowth identity name that has structurally decelerated to a ~10% grower, now valued reasonably on improving free cash flow; a “quality at a fair price” stabilization story with a Microsoft overhang that caps the upside. Sell-side targets cluster ~$100–150 with a genuinely split ratings book (lots of Buys, several pointed Neutrals) — i.e., consensus is “fine, not exciting.”

The strongest bull case. The deceleration has bottomed and the inflection is underway: NRR turned back up to 107%, the security/governance cross-sell (OIG/OPA/ITP) is a proven engine reaching >20,000 sticky customers, and agentic AI / non-human identity is a real TAM-expansion option that plays directly to Okta’s one durable edge — independence. Stack on ~25–28% FCF margins, a net-cash balance sheet, a counter-cyclical buyback, declining SBC, and a 36th-percentile-of-history sales multiple, and you have an abandoned former-growth name at an early, unconfirmed inflection that the momentum crowd hasn’t re-bought (5-yr return −13%/yr, momentum factor loading negative). If ~10% is the floor and agentic identity lifts the ceiling, the stock re-rates from a depressed base — and you weren’t paying for a crowd.

The strongest bear case. The deceleration is not over — it is a five-year unbroken trend (56→43→22→15→12→~9–10%), and Microsoft Entra’s bundling economics ensure the terminal rate keeps drifting toward mid-single digits, at which point identity is “a feature, not a platform” and a ~31x forward multiple compresses hard. NRR’s one-quarter uptick is noise, not signal; the agentic narrative is immaterial hype that Microsoft will bundle away just as it bundled workforce identity (“Entra Agent ID”); and the balance-sheet/FCF strengths are partly accounting artifacts (interest income, SBC add-back, a 7.8% tax rate). The breach history is a latent re-rating risk. You’re paying ~31x forward earnings for a decelerating ~10% grower with a structurally-capped TAM and a dominant competitor that gives the product away.

The 3–5 assumptions that matter most:

1. Is the Microsoft-Entra ceiling a hard cap or a manageable headwind? (The single most important variable. Bear: hard cap → mid-single-digit terminal. Bull: manageable, because neutral/multi-cloud demand and the AI frontier are real and growing.)
2. Is the 107% NRR inflection durable? (Falsifies the bear if it holds/rises for 2–3 quarters; falsifies the bull if it rolls back below 106%.)
3. Does agentic / machine identity become a named, material revenue line? (The bull’s TAM-expansion engine. Today: zero. Watch for management to start quantifying it — that is the trigger.)
4. Does FCF margin hold ~25–28% as the interest-income and deferred-revenue tailwinds fade? (Tests whether the cash economics are operating or financial.)
5. No third breach. (A latent binary; not in anyone’s model.)

Factor-positioning evidence (where consensus may be offsides). The tape says abandoned, not crowded: Cybersecurity-industry beta ~1.2, market beta ~1.05, momentum loading negative (−0.35), high idiosyncratic vol (~43%), 5-yr drawdown −83%, 5-yr return −13%/yr — yet a sharp recent inflection (Q1 quarter ~+40% actual). This is the signature of a value/recovery setup in a growth stock’s clothing: the crowd left years ago and has not returned despite the inflection. That supports the bull’s asymmetry argument (you’re early, not late) — but the same high idiosyncratic vol warns that the floor can break violently on any single growth miss. Net variant read: the risk/reward is balanced-to-mildly-favorable, and the edge is in patience (buying a growth scare) rather than chasing the post-beat move.

12. Fact vs. Interpretation Table

13. Open Questions

1. What are Auth0/CIAM’s standalone revenue, growth, and margin? Okta does not break them out — the single biggest disclosure gap, and the only way to judge whether the $6.5B deal is earning its keep.
2. What is the NRR trajectory over the next 2–3 quarters? One quarter up is an inflection; three is a trend. The whole bull case hinges here.
3. Can Okta quantify agentic/machine-identity revenue, and when? Until management names a number, it is an option, not a business.
4. How fast is Microsoft actually taking new-logo share — and is Okta’s win-rate in competitive deals stable, improving, or eroding? (Not disclosed; channel checks needed.)
5. What is the durable FCF margin once interest income and the deferred-revenue float normalize? Is the cash machine operating or financial?
6. Does the public-sector / FedRAMP business become a needle-mover, and is it insulated from the broader deceleration?
7. Where does SBC settle? 18.6% and falling — does it reach the low-teens (further closing the reported-vs-owner FCF gap), or stall?

14. What Must Be True

For the bull case to be right:

• NRR holds at/above 107% and the cross-sell engine sustains — the deceleration floor is real. Falsification: NRR drifts back below 106% over the next two quarters, or FY27 revenue growth guides/prints below 9%.
• Agentic / non-human identity becomes a named, material revenue contributor by FY28–29, validating the TAM-expansion and proving Okta’s independence is a durable right-to-win. Falsification: management still calls it “immaterial / not in the guide” a year from now, or Microsoft’s “Entra Agent ID” demonstrably wins the agent-identity standard.
• FCF margin holds ~25–28% as interest-income and float tailwinds fade — confirming operating, not financial, cash generation. Falsification: FCF margin compresses toward the high-teens as the cash pile is deployed.

For the bear case to be right:

• Microsoft Entra grinds Okta’s terminal growth to mid-single digits — identity becomes a feature, the new-logo motion keeps eroding, and ~10% proves to be a way-station, not a floor. Falsification: revenue growth re-accelerates (NRR > 110%, growth back to low-teens) for two-plus quarters.
• The cash economics are partly artifact — strip interest income, the SBC add-back, and the low tax rate and the “owner” earnings power is far below the ~30% FCF headline, justifying a lower multiple. Falsification: owner-FCF (ex-SBC) margin expands and SBC keeps falling toward low-teens.
• A latent re-rating event (a third breach, or an Auth0 impairment) re-prices the trust premium. Falsification: a clean multi-year security record and stable goodwill.

The single cleanest test of the whole thesis: the NRR print over the next two quarters. Sustained >107% validates the bull’s “floor is in”; a roll-back below 106% validates the bear’s “deceleration is unbroken.” Everything else — the AI option, the margin debate, the multiple — is downstream of whether the installed base is re-expanding or still leaking to Microsoft.

15. Source Appendix

See the separate Source Appendix (Appendix B below) for the full, dated, primary-source citation list. Principal sources: Okta FY22–FY26 Forms 10-K and the Q1-FY27 Form 10-Q (SEC EDGAR, CIK 0001660134); the Q4-FY26 and Q1-FY27 earnings releases (Form 8-K EX-99.1, 04-Mar-2026 and 28-May-2026); the Q1-FY27 earnings-call transcript (28-May-2026); the 2026 Proxy Statement (DEF 14A, 07-May-2026); the Form 4 insider-transaction corpus; third-party market/valuation/factor data (own-history valuation percentiles, factor loadings and risk-adjusted return history, and the AZI price series) used as cross-checks and reconciled to filings. All quantitative figures driving verdicts are reconciled to the underlying 10-K/10-Q; third-party aggregated data is treated as a starting point, not primary.

APPENDIX A — Standard Diligence Questionnaire — Okta, Inc. (NASDAQ: OKTA)

Report date 2026-06-19. Supplemental to the research note. Fact/Interpretation/Assumption labels applied where material. FY ends Jan 31.

General

What thoughtful questions have other investors asked about this company? The recurring institutional debates: (1) Is the deceleration structural or cyclical? — i.e., is Microsoft Entra permanently capping growth at ~10%, or did the breaches + GTM reorg + macro create a temporary trough now inflecting? (2) What is Auth0 actually worth four years after a $6.5B all-stock peak-multiple deal that Okta won’t break out? (3) Is 30% FCF margin “real” or an SBC/interest-income/deferred-revenue artifact? (4) Can Okta monetize machine/agent identity before hyperscalers bundle it? (5) Is the NRR inflection to 107% durable? These map directly to this note’s swing variables.

Cyclicality & Earnings Nature

Are earnings at a cyclical high or low? Interpretation: Margins are at an all-time structural high and still expanding (op margin −20%→+5% GAAP / ~25% non-GAAP in two years), but this is a secular operating-leverage story, not a cyclical peak — there is further margin runway as SBC falls and opex stays disciplined. Revenue growth is near a cyclical/secular low (~10% vs ~56% peak). So: earnings power rising, growth rate troughing. Driven by external environment or internal actions? Predominantly internal (cost discipline, GTM repair, cross-sell) for the margin story; external (Microsoft competition, macro seat counts) for the growth deceleration. How stable are revenues? Very — ~98% subscription, multi-year contracts, ~107% net retention, $4.8B RPO / $2.5B cRPO providing forward visibility. Among the most predictable revenue bases in software. Outlook for products/services? Core SSO/MFA is mature/commoditizing; growth shifts to governance (OIG), privileged access (OPA), identity threat protection (ITP), CIAM (Auth0), and the agentic-identity frontier. How big will this market be? Fact/Interpretation: Okta markets “$80B+ TAM”; the credibly-addressable IAM/IGA/PAM market is ~$24–26B today → ~$42B by 2030 (~10% CAGR). Growing, global (~20% of revenue international with room to expand), not shrinking — but growing at roughly Okta’s own current rate.

Business Quality & Competitive Moat

Is the industry getting more or less competitive? More — Microsoft Entra’s bundling intensifies at the low/mid end; CIAM upstarts pressure Auth0; CyberArk (→PANW) and SailPoint contest the cross-sell adjacencies. Late capital cycle. How profitable is the business (ROIC, ROE)? Reported ROIC ~2% / ROE low-single-digits — depressed by $5.49B Auth0 goodwill (59% of equity), not by operating economics. The organic business has high incremental returns (~70%+ incremental operating margins). Interpretation: underlying operating returns are strong; acquired-capital returns are sub-cost. How profitable is the industry; barriers to entry? High gross-margin (software) industry; barriers = integration breadth, certifications (FedRAMP), switching costs, brand/trust. But the dominant competitor’s marginal price is ~zero (bundled), which caps industry pricing for the contested segment. Can the business be easily understood? Yes — a subscription identity platform; clean model. The hard part is the competitive judgment (Microsoft), not the accounting. Undermined by foreign low-cost labor? No — it’s a software/IP business; not labor-arbitrage exposed. Do brands matter? Yes, acutely — Okta sells trust, so brand and security reputation are core assets (and the reason the 2023 breach was so damaging). “Neutral/independent” is itself a brand position. Nature of competition? Bundling economics (Microsoft) vs best-of-breed/neutrality (Okta) vs point-solutions (CyberArk/SailPoint). Won on platform breadth, integrations, multi-cloud neutrality; lost on “good-enough-and-free.” Customers’ switching costs? High — identity is wired into thousands of app integrations and HR/joiner-mover-leaver workflows; ripping it out is expensive and risky. Evidenced by ~106–107% NRR through a breach and a recession.

Financial Condition & Balance Sheet

Assets not fully recognized on the balance sheet? The installed base / switching-cost franchise and the Okta Integration Network (7,000+ integrations) are valuable intangibles not on the balance sheet. Conversely, $5.49B of Auth0 goodwill is on the balance sheet and arguably over-stated. Off-balance-sheet liabilities? None material beyond standard operating leases (already capitalized) and the (small) convertible. No pension, minimal debt. How conservative is the accounting? Reasonably conservative on revenue (subscription, ratable). Flags: GAAP profit flattered by a ~7.8% tax rate (NOL-driven, will normalize) and ~$110M non-operating interest income; non-GAAP excludes SBC (a real cost). Goodwill un-tested by segment disclosure. How CapEx-hungry? Not at all — capex <0.5% of revenue; asset-light SaaS. This is a core quality positive (FCF ≈ OCF).

Capital Allocation & Management

How much FCF, and how is it used? ~$875M FY26 (30% margin). Uses: a first-ever $1B buyback (~$320M executed through Q1-FY27, counter-cyclically near lows), cash-settling the $350M 2026 convert, and building the cash pile. No dividend. Interpretation: a sensible, shareholder-friendly shift now that growth no longer consumes all cash. Significant acquisitions recently? Auth0 (2021, $6.5B, all-stock) is the defining deal — value-destructive at the price paid; small tuck-ins since (Spera → ISPM). No large deals recently — a positive given the Auth0 experience. Buying back shares? Yes — newly, and counter-cyclically. First capital return in company history. Issuing large amounts of stock to insiders? SBC $544M (18.6% of revenue) — large but declining (from 30%); 48M-share plan reserve is the overhang. Net dilution down to ~2.4%/yr and now offset by buyback. Compensation policy / incentive metrics? Fact (DEF 14A 2026): annual bonus = 60% Revenue / 40% non-GAAP Operating Income (funded 98.7%); CRO on Net ARR commission (paid 150%); LTI = relative-TSR PSUs (114% achievement, capped 100%). Pays on profitable growth + relative TSR — well-aligned. Motivations of management? Founder-led (McKinnon CEO/co-founder; Kerrest Exec Vice Chair, disengaged); dual-class gives founders ~31.6% of votes on ~4.9% economics. Long-tenured, mission-driven; entrenchment risk but no evidence of self-dealing.

Valuation & Market Data

ADR, MLP, or K-1 issuer? No — US C-corp common stock (Class A, NASDAQ: OKTA). No K-1. Dividend policy? None; returns capital via buyback only. How profitable? GAAP net margin ~8%; non-GAAP operating margin ~25%; FCF margin ~30% (reported) / ~11% ex-SBC. Inflecting positive. Net income diverging from cash from operations? Yes, favorably — OCF ($884M) >> GAAP NI ($235M), driven by the SBC add-back and deferred-revenue float. This is normal/healthy for SaaS but means GAAP NI understates cash generation while reported FCF overstates “owner” earnings (the SBC cost is real). Read both.

Risks & Downside

What factors would cause the stock to decline? A growth miss / guide-down below ~9%; NRR rolling back below 106%; a third security breach; multiple compression in high-multiple software; evidence of accelerating Microsoft share capture; agentic-identity proving unmonetizable. Risk of catastrophic loss? Low at the enterprise level — net cash $2.2B, $875M FCF, GAAP-profitable, no refinancing risk. The realistic severe case is fundamental erosion (Microsoft commoditizes → mid-single-digit growth → de-rate to ~3–4x sales), a meaningful but non-terminal capital loss. Chance of total loss? Negligible — fortress balance sheet, large recurring cash flows, sticky installed base.

Recent News & Events

Has the business environment changed recently? Yes, positively at the margin: Q1-FY27 (28-May-2026) revenue +12%, NRR inflected up to 107%, FY27 guide raised → stock to a 52-week high ($142.35, 1-Jun); analyst targets raised to a ~$100–150 cluster (split ratings). The agentic-AI strategy (Auth0 for AI Agents, GA April-2026) is the new narrative. Significant acquisitions? None recent of size (Spera tuck-in for ISPM). Change in accounting policies? Non-GAAP tax rate changed to 21% effective FY27 (from 26%) — a modeling note, not a red flag. Recent changes — new markets, facilities, management? Eric Kelliher as President/COO; deepening public-sector/FedRAMP push; GTM shift moving professional services to global system integrators (~1-pt FY27 revenue headwind, margin-accretive); continued opex discipline.

APPENDIX B — Source Appendix — Okta, Inc. (NASDAQ: OKTA)

Report date 2026-06-19. Primary sources first. All figures driving verdicts reconciled to SEC filings; third-party aggregated data used as cross-check and labeled. SEC EDGAR CIK 0001660134.

Primary — SEC filings (EDGAR, CIK 0001660134)

1. Form 10-K, FY2026 (period ended 31-Jan-2026), filed 05-Mar-2026 — okta-20260131.htm. Business (Item 1), risk factors, MD&A, financial statements/notes, segment/customer disclosure, competition, RPO/cRPO, SBC, goodwill.
2. Form 10-K, FY2022–FY2025 (periods ended 31-Jan), filed 2022-03-07 / 2023-03-03 / 2024-03-01 / 2025-03-05 — multi-year revenue, margin, SBC, dilution, and breach/litigation history.
3. Form 10-Q, Q1-FY2027 (period ended 30-Apr-2026), filed 29-May-2026 — okta-20260430.htm. Q1 revenue +12%, NRR/cRPO, balance sheet (cash $2.59B, debt $411M, net cash $2.18B), buyback execution, convert settlement intent.
4. Form 8-K / EX-99.1 — Q4 & FY2026 earnings release, filed 04-Mar-2026 — okta-1312026_ex991.htm. FY26 revenue $2,919M; GAAP EPS $1.31; non-GAAP net income $646M, non-GAAP diluted EPS $3.50; Q4 non-GAAP EPS $0.90; FY27 guidance: revenue +9–10%, non-GAAP op margin 25–26%, FCF margin 27–28%, non-GAAP EPS $3.74–$3.82 (~185M shares, 21% non-GAAP tax rate).
5. Form 8-K / EX-99.1 — Q1-FY2027 earnings release, filed 28-May-2026 — okta-20260528.htm. Revenue +12%, NRR 107%, cRPO +12%, raised FY27 guide.
6. DEF 14A — 2026 Proxy Statement, filed 07-May-2026 — okta-20260507.htm. Executive compensation, FY26 incentive metrics (60% Revenue / 40% non-GAAP Op Income, funded 98.7%; CRO Net ARR 150%; relative-TSR PSUs 114%→capped 100%), dual-class voting (founders ~31.6% of votes, ~4.9% economic), equity-plan reserve.
7. Annual Report to Shareholders (ARS) FY2026, filed 07-May-2026.
8. Form 4 / Form 144 insider-transaction corpus (EDGAR full-text & XML, FY ongoing) — ~329 Form 4s (predominantly 10b5-1/RSU-vest M/F/S codes; one notable open-market purchase, code P, by a director near the 52-week low) and routine Rule 144 sale notices.
9. 8-K material-event timeline (2021–2026) — earnings releases, the Oct-2023 support-system (HAR-file/token) security-breach disclosure, Jan-2022 Lapsus$ disclosure, the $60M securities-class-action settlement (D&O-funded; dismissed Nov-2024), buyback authorization, and convertible-note matters.

Primary — Company materials

10. Q1-FY2027 earnings-call transcript (28-May-2026) — McKinnon (CEO), Tighe (CFO), Kelliher (President/COO); guidance philosophy, NRR inflection, agentic-AI/non-human-identity commentary (“not material,” “heavily discounted in the guide”), GSI services shift, interest-income/FCF headwind notes. (Saved locally: output/OKTA/transcripts/OKTA_Q1FY27_2026-05-28.txt.)
11. Okta investor-relations supplemental commentary (posted with Q1-FY27 release) — KPI detail (customer cohorts >$100k ACV, RPO/cRPO, international mix).
12. Okta product/security disclosures — “Secure Identity Commitment”; Auth0 for AI Agents (GA April-2026) and Okta AI/Identity Threat Protection product announcements.

Secondary — Market, valuation, and factor data (cross-check; reconciled to filings)

13. Third-party fundamentals/valuation aggregator (ROIC.ai) — multi-year income statement, balance sheet, cash flow, profitability ratios, enterprise value, and valuation multiples; used as a starting point and reconciled to the 10-K/10-Q (notably: ROIC’s snapshot EV was stale to an ~$74 basis and recomputed live to ~$18.9B).
14. Own-history valuation-percentile index — P/S at the 36th percentile of Okta’s ~10-year range (P/B percentile disregarded as goodwill-distorted/meaningless).
15. Factor / risk model (FactorsToday) — factor loadings (Cybersecurity industry beta ~1.2, market beta ~1.05, negative momentum loading), risk-adjusted return history (5-yr return −13%/yr, max drawdown −83%, recent-quarter inflection), idiosyncratic volatility (~43%), and factor-similar peers (RBRK, CRWD, NET, S, ZS).
16. Daily price/OHLCV series (AZI, split/dividend-adjusted) — five-year price history underpinning the Event Map: 2021 peak ~$294 (12-Feb-2021), trough $44.12 (04-Nov-2022), 52-week high $142.35 (01-Jun-2026), 18-Jun-2026 close $117.81.
17. News / analyst-action feed — post-Q1-FY27 sell-side target revisions to a ~$100–150 cluster with split ratings (Cantor, Morgan Stanley, JPMorgan, RBC, Truist, DA Davidson, Macquarie, UBS positive; Wells Fargo, Scotiabank, Susquehanna, Piper Sandler neutral).

Peer cross-reads (prior sector research notes, for industry framing and comps)

18. CRWD (CrowdStrike), ZS (Zscaler), PANW (Palo Alto Networks), NET (Cloudflare), S (SentinelOne) — used for cybersecurity industry structure, the Microsoft-bundling overhang, capital-cycle framing, and the cross-sectional valuation comp table; independent OKTA analysis conducted throughout.

Third-party market sizing

19. Industry IAM/IGA/PAM market-size estimates (~$24–26B today → ~$42B by 2030, ~10% CAGR) used to pressure-test Okta’s “$80B+ TAM” claim; cyber M&A reference points (Google–Wiz, Cisco–Splunk, Palo Alto–CyberArk) for the Marathon capital-cycle read.

Note on methodology: management commentary is treated as hypothesis and validated against filings, financials, and external evidence. Facts, interpretations, assumptions, and open questions are labeled throughout the note. No price target or buy/sell recommendation appears outside the clearly-labeled “Claude’s Take” block.