Wix.com Ltd. (NASDAQ: WIX) — Betting the Balance Sheet on Vibe Coding at Exactly the Wrong Price

Independent equity research · Published 2026-08-01 · Initiating coverage Price at analysis: $55.03 (2026-07-31) · Shares outstanding 41,849,511 (2026-05-11) · Market cap ~$2.30B · EV ~$3.05B

⚡ Claude’s Take

This block is the author’s own subjective opinion. It is general information, not investment advice, and no recommendation is implied. The analytical body that follows it takes no position and carries no price target — it discusses valuation only as embedded expectations and scenarios.

Verdict: HOLD / AVOID here — do not chase the 36% bounce off the June low; the price that interests me is the low $40s, not $55. Not a short. Low-to-medium conviction. At $55.03 Wix trades at roughly 1.4x EV/sales and about 7.3x its own guided 2026 free cash flow, and that headline is the most seductive number in the story. It is also the wrong one. Guided FCF of ~$420m is struck before ~$230m of annual stock compensation, before $30–35m of restructuring cash, and before the Base44 earnout and retention payments that FY2025’s cash flow accrued but did not pay. Expense stock comp honestly and owner free cash flow is closer to $190m, which puts the enterprise at ~16x — an ordinary multiple for a business whose paying subscription base shrank from 6.2 million to 6.1 million in 2025 while an acquisition was being added to it, whose non-GAAP operating margin has gone 22% → 19% → 15% → 5% in five quarters, whose entire FY2025 GAAP profit was a $70.9m deferred-tax valuation-allowance release, and which now carries net debt and negative book equity for the first time in its life.

What tips this from “cheap enough” to “wait” is capital allocation, and the arithmetic is brutal. In March 2026 Wix sold 3.27m shares to Durable Capital Partners at $79.59; four weeks later it bought 17.5m shares — 30% of the company — at $92.00 in a Dutch auction it funded with those proceeds plus a $500m bank facility. Nine trading days after settlement the stock traded at $63.84, and today it is $55.03. Holding enterprise value constant, that tender cost continuing holders roughly $11 a share, about 20% of the current market capitalization. Zoom out and it is worse: since 2021 Wix has spent ~$3.21bn on buybacks — 1.4x its entire market capitalization today — to retire a net ~15.2m shares, because ~$230m a year of stock issuance kept refilling the bucket. That is roughly $195 of net cost per share retired against a $55 stock. Ten weeks after the board declared “continued confidence in the Company’s ability to drive strong cash flow generation,” management cut guidance and fired 20% of the workforce. My framing is explicitly not contrarian value: the tape agrees with me and disagrees with the bulls — Momentum loads −1.23 while Value loads +0.07, the stock is 84% off its 2021 peak with 89% three-month realized volatility, and 73% of its return variance is idiosyncratic. This is a falling knife that has bounced, not a value stock the market has recognized. Directional zone: my base case supports roughly $65–75, but the bear case ($12–25 if AI genuinely commoditizes the paid builder) is live enough, and the Q2 print lands on August 4 — three days from now — that paying $55 for a 12x-dispersion outcome ahead of a binary event is not a good use of risk.

Conviction: low-to-medium. Flips bullish if: the paying premium-subscription count stops shrinking (core Wix subs plus Base44 growing in units, not just ARPS), and Base44’s stand-alone gross margin is disclosed above ~60% with the $150m run-rate compounding. Flips bearish if: core Wix bookings growth ex-Base44 goes to zero or negative, or the Base44 earnout/retention accrual starts converting to cash faster than FCF grows. One-line tag: they bought 30% of the company at $92 to prove it was cheap; the market disagreed within a fortnight.

📈 Stock Price Action — Five-Year Event Map

Over the trailing five years Wix has fallen from a five-year high of $298.66 (2 August 2021) to a five-year low of $40.43 (23 June 2026), and now sits at $55.03 — 84.4% below the 2021 peak and 70.1% below the 52-week high of $184.24 set on 22 September 2025, with a 52-week range of $40.43–$184.24. The stock has bounced roughly 36% off the June low in five weeks, but trades 33% below its 200-day exponential moving average ($82.45). Year-end closes tell the round trip plainly: $157.79 (2021), $76.83 (2022), $123.02 (2023), $214.55 (2024), $103.89 (2025), $55.03 today. In each row below, the price move is a FACT; the attributed driver is INTERPRETATION.

Cycle narrative. (1) Wix was a canonical COVID beneficiary — 2020 revenue grew 30% as the world moved online — and gave nearly all of it back through 2022 as rates rose and the market stopped paying for unprofitable growth; FY2022 produced a $285.4m GAAP operating loss and a $424.9m net loss. (2) The 2022–24 recovery was a genuine self-help story: gross margin rose from 62.1% to 67.9%, operating income turned positive in FY2024 at $100.1m, and management bought back $825m of stock across three years while Wix Studio opened the agency/Partners channel. (3) 2025 reversed it. The 21 May 2025 (−16.2%) and 19 November 2025 (−19.9%) reactions were both to the same disclosure: management was raising bookings guidance while conceding that non-GAAP gross margin would fall to 68–69% “due to higher AI costs” and non-GAAP operating expenses would rise to ~50% of revenue to fund Base44 marketing. The market read growth bought with margin. (4) On 4–5 March 2026 the stock rallied 25% on a package — better-than-feared Q4, a $250m (upsized to $260m) private placement led by Durable Capital Partners “with an extended lockup,” a $2bn repurchase authorization, and a $1.75bn Dutch auction at $80–$92. (5) The auction cleared at the top of its range, $92.00, for ~17.5m shares (~$1.6bn); the day after expiry the stock fell 9.4%, and within nine sessions it traded at $63.84 — 31% below what the company had just paid. (6) The 13 May Q1 print was the single largest five-year down day, −27.1%: selling and marketing rose 78.9% year over year to $199.6m, total operating expenses rose 48.7% on 14.3% revenue growth, and non-GAAP operating margin collapsed from 21% to 5%. (7) On 28 May management told staff 1,000 jobs (20%) would go; the 8 June 6-K formalized it, cut FY2026 bookings growth to low-teens, and disclosed “a more pronounced slowdown, beyond our previous expectations, in the growth of our Partners business.” The stock bottomed at $40.43 on 23 June. (8) The subsequent 36% bounce reflects the $150m run-rate cost saving and a $420m FCF guide; it was interrupted by Morgan Stanley’s Overweight-to-Equal-weight downgrade (target $112 → $60) around 21 July and Raymond James’ 30 July downgrade on valuation, which took 9.4% off in a session.

1. Executive Summary

Wix.com is the largest website-building platform in the world — roughly 45% of the global builder market, 304 million registered users, 6.1 million paying premium subscriptions, and $1.99bn of 2025 revenue growing 13%. It sells a freemium subscription (Creative Subscriptions, 71% of revenue, 83% gross margin) attached to a lower-margin bundle of resold and owned business services (Business Solutions, 29% of revenue, 32% gross margin), collects cash upfront, and carries $903m of deferred revenue as interest-free float. For most of the last decade the debate about Wix was whether a company that never earned a GAAP profit could convert its cash generation into durable economics. In 2026 that debate has been replaced by a harder one.

The business is being re-underwritten in real time as an AI company, and the transition has destroyed its reported economics in five quarters. Non-GAAP operating margin went from 22% (FY2024) to 19% (FY2025) to 15% (Q4-2025) to 5% (Q1-2026). GAAP operating income for the whole of FY2025 was $1.8 million on $1,993m of revenue. The entire $50.6m of FY2025 GAAP net income — and more — came from a $70.9m release of the deferred-tax valuation allowance; without it the year was an approximate $20m loss. Q1-2026 produced a $57.5m GAAP net loss with selling and marketing up 78.9% year over year. The proximate causes are two AI initiatives: Base44, a “vibe-coding” app-generation platform bought in June 2025 for $92.2m of headline consideration that has scaled from zero to ~$150m of ARR by mid-May 2026, and Wix Harmony, an AI site builder now running on Wix’s own proprietary LLM. Both are being funded with front-loaded marketing (a >$20m Super Bowl campaign in Q1 alone) and inference costs that are a genuinely new variable cost in a business that has run at near-zero marginal cost for twenty years.

Underneath the AI story sits an uncomfortable fact: the paying unit base is shrinking. Premium subscriptions were approximately 6.2 million at 31 December 2024 and approximately 6.1 million at 31 December 2025 — including Base44. Registered users grew 7.7% over the same period. All revenue growth is price and mix (“ARPS”), a structure published research on GoDaddy has identified as the industry-wide signature of a franchise monetizing a declining base. Wix’s own FY2025 20-F now lists Claude Code, Cursor, Replit, Lovable, Google AI Studio, Bolt.new and v0 as competitors — an admission that the boundary of its market is being redrawn by parties with vastly larger R&D budgets and no need to earn a return on the SMB-presence franchise.

And the capital allocation has been, on the evidence to date, value-destructive at scale. Since 2021 Wix has spent ~$3.21bn on repurchases — 1.4x its entire current market capitalization — to retire a net ~15.2m shares, roughly $195 per net share retired against a $55.03 stock, because ~$230m a year of stock compensation continuously refilled the count. The 2026 sequence is the sharpest example: a $260m equity placement at $79.59 on 5 March, funding part of a $1.61bn Dutch auction at $92.00 four weeks later, followed ten weeks after the board’s stated “continued confidence” by a guidance cut and a 20% headcount reduction. Wix now carries negative shareholders’ equity of $102m, $1,150m of 0.00% converts struck at $210.49, a drawn $500m bank facility with a Bank-Debt/FCF ≤ 2.0x covenant and NIS 1bn of pledged collateral, and — for the first time in its history — net debt.

At $55.03 the enterprise is capitalized at roughly $3.05bn, or 1.37x guided 2026 sales and 7.3x guided free cash flow. Fully expensing stock compensation gives owner free cash flow of roughly $190m and a multiple nearer 16x. That is neither obviously cheap nor obviously expensive; it is a wide-dispersion outcome resting on a single unresolved question — whether AI application generation is a market Wix wins or a solvent that dissolves the paid website builder. A securities class action (class period 19 February 2025 to 12 May 2026) alleges the company overstated Base44 and Harmony’s competitiveness and understated their cost. Q2-2026 results land on 4 August 2026, three days after this report.

This memo takes no position and sets no price target. It concludes that Wix is a genuine category leader with a real brand asset, an impressive product organization, and a demonstrably poor record of turning either into shareholder value — now attempting the most expensive strategic pivot in its history from a weakened balance sheet it weakened itself.

2. Business Overview

What Wix sells. Wix operates a freemium software-as-a-service platform on which a non-technical person can create, host and operate a website or, since 2025, a functional web application. A user registers for free, builds a site with a drag-and-drop editor (Wix Editor), an AI-driven generator (Wix ADI since 2016, now Wix Harmony), or a professional workspace (Wix Studio, aimed at agencies and freelancers), and then pays a monthly, annual or multi-year subscription to remove Wix branding, attach a custom domain, and unlock commerce, scheduling and marketing functionality. As of 31 December 2025 there were 304.2 million registered users and approximately 6.1 million premium subscriptions, implying a conversion rate of about 2.0%. The company has 5,340 employees and contractors, over 60% of them in Israel, with material R&D and customer-care operations in Ukraine (594 contractors and 11 employees at year-end 2025).

The two reporting segments.

• Creative Subscriptions — FY2025 revenue $1,409.7m, +11.4% y/y, 70.7% of total, GAAP gross margin 83.2%. Wix premium plans, domain registration and renewal, hosting, and — importantly for the current debate — “other subscription based offerings, such as Base44.” This is the annuity: subscriptions renew, cash is collected upfront, and the incremental cost of serving an existing site has historically been close to nothing. Creative Subscriptions ARR reached $1.52bn at year-end 2025 (+13%).

• Business Solutions — FY2025 revenue $583.3m, +17.7% y/y, 29.3% of total, GAAP gross margin 31.6%. Payments (Wix Payments, POS, shipping, multi-channel commerce and gift cards), resold Google Workspace, email marketing, paid ads, and third-party apps from the Wix App Market. Transaction revenue — the payments piece — was $255.0m in FY2025 (+19%). The low margin is structural: this segment is largely a pass-through of interchange and third-party licence cost, and its economics improve only with scale in Wix’s own payment stack. Business Solutions ARR was $312.4m at year-end 2025 (+18%). GPV grew 12% y/y to $3.8bn in Q1-2026, which management describes as “soft” on SMB macro pressure.

A third cut that matters more than the segments: Partners. Wix reports “Partners revenue” — revenue generated through agencies and freelancers who build for others, plus B2B resellers such as LegalZoom and Vistaprint — across both segments. It was $750.3m in FY2025, +23% y/y (37.6% of total revenue), the fastest-growing part of the business and the strategic bet of the 2023–25 period (Wix Studio was built for it). In Q1-2026 Partners revenue was $203.4m, +19% — a deceleration management called “larger-than-expected” — and the 8 June 6-K then disclosed “a more pronounced slowdown, beyond our previous expectations, in the growth of our Partners business during the second half of May and early June.” The cause management gives is self-inflicted: “smaller partners cohorts added over the past few quarters as we pulled back partners focused marketing spend,” redirecting that spend to Base44 and self-creators. Whether professional web builders are leaving because Wix stopped marketing to them or because they now use Cursor and Claude Code is the most consequential open question in the business.

How the money actually flows — the cohort/bookings model. Wix does not primarily manage to revenue. It buys user cohorts with marketing spend and manages to a time-to-return-on-investment (TROI) on that spend, measured against bookings (cash collected plus committed, before revenue deferral). In Q1-2026 it spent approximately $90m on acquisition marketing (excluding the Super Bowl and AI costs attributed to free users) against a blended TROI target of 7–9 months, with Base44’s TROI running “less than 12 months.” The 6.4-million-user Q1-2026 cohort generated $52m of bookings in its first three months, +46% versus the Q1-2025 cohort. Cash received upfront and recognized over the subscription term produces the $903.3m deferred revenue balance (current $778.6m + long-term $124.6m) that funds working capital.

The model has one important consequence for how a reader should treat the accounts: because Wix expenses acquisition marketing immediately and recognizes the resulting revenue over years, an acceleration in marketing spend mechanically destroys near-term reported profitability even if the returns are good. That is precisely the defence management offers for Q1-2026. It is a legitimate defence — and it is also exactly what a company with deteriorating returns would say. The way to adjudicate it is the paying subscription count and the cohort bookings series, both discussed in the Growth section below.

Verdict. A well-understood, genuinely global, asset-light subscription business (FY2025 capex $9.9m on $1,993m of revenue) with a real float, a high-margin core and a low-margin attach layer. The reporting is clear at the segment level but has become materially less informative in 2026 because Base44 — a business with a completely different cost structure — is consolidated inside Creative Subscriptions with no stand-alone disclosure.

3. Industry Dynamics

Structure. The market for SMB web presence is large, fragmented and — critically — has no barrier to entry at the retail layer. Registrars, hosts, builders, e-commerce platforms and now AI code generators all sell overlapping bundles to the same buyer: a non-technical person who wants to exist online. Wix’s own FY2025 20-F names, in a single competition paragraph, Shopify and BigCommerce (commerce), Mindbody (vertical scheduling), WordPress.org and Drupal (open-source CMS), GoDaddy (registrar/host), plus SEM/SEO/email vendors — and then adds the category that did not exist in the FY2023 filing: “‘vibe coding’ solutions for developers, such as Claude Code, Cursor and Replit, and AI-powered low-code/no-code application builders, such as Lovable, Google AI Studio, Bolt.new, and V0.”

Market shares. Third-party estimates of global website-builder share put it at Wix ~45%, Squarespace ~16–18%, GoDaddy ~10%. Wix is the clear category leader and has been for years. Two facts qualify that leadership. First, Squarespace — the #2 — exited the public market, taken private by Permira for $7.2bn in October 2024, two years before the AI de-rating; the most comparable asset was sold at a premium by owners who did not have to live through 2026. Second, category leadership in builders has not translated into leadership in the adjacent profit pools: Shopify dominates commerce at a scale Wix does not approach, and GoDaddy dominates the domain/registrar annuity.

The capital cycle is running hot, and it is running against the incumbent. High perceived returns in AI-native creation have attracted an extraordinary inflow of capital and talent in roughly eighteen months. Lovable reached ~$500m ARR. Base44 went from founding to $150m of ARR in about a year. Vercel (v0), StackBlitz (Bolt.new), Google (AI Studio), Anthropic (Claude Code), Cursor and Replit are all funded, all shipping, and — with the partial exception of Lovable — none of them needs the SMB-presence business to earn its cost of capital. The capital-cycle prediction is not subtle: returns in AI application generation will be competed away. The only question is whose returns disappear first.

Wix is unusually exposed because it sits on both sides of the cycle simultaneously. It is the incumbent being attacked in site building (where its 45% share and 83% gross margin are the prize), and it is an entrant attacking in app generation (where it must outspend Lovable and the hyperscalers). It therefore pays the cost of the capital cycle twice — defensive marketing on the core and offensive marketing on Base44 — which is a precise description of what happened to the P&L in Q1-2026.

The economics of the substitute have changed in a way that matters. For twenty years the marginal cost of serving a Wix site was hosting and support — near zero, falling, and fixed-cost-like. AI generation is different: inference is a variable cost that scales with usage, and the heaviest usage occurs exactly when a user is building and has not yet paid. Management is explicit that Base44’s AI costs are “front loaded as new users consume more AI inference bandwidth during their initial build phase,” and that consolidated non-GAAP gross margin fell from 69% to 66% year over year on “elevated investments in BASE44.” Wix’s answer — build proprietary models — is credible (Harmony now runs on a Wix-built LLM at what management calls “marginal cost” versus frontier models) but is not yet applied to Base44, where the cost actually sits, and management would give no timeline.

Regulation. Light but rising and asymmetric. The EU Digital Services Act imposes take-down and content obligations on a platform hosting hundreds of millions of user sites it does not monitor. Emerging AI regulation is unquantified. US state sales/use tax nexus is a live and recurring item — a “sales tax accrual” appears in Wix’s non-GAAP bridge every single quarter. None of these is thesis-determining today; all of them are one-way costs.

Verdict: a structurally poor industry that has just got worse. There is no industry-wide barrier to entry, substitutes are abundant and multiplying, the buyer is a price-sensitive small business, and capital is flooding in at the exact point in the value chain where Wix earns its margin. Wix’s returns have come from being the scale winner within a bad industry — brand, funnel and product breadth — not from the industry’s structure. That is a survivable position when the capital cycle is quiet. It is a dangerous one when it is not.

4. Competitive Position

Name the moat. Under Bruce Greenwald’s taxonomy the three genuine advantages are supply/cost, demand/customer captivity, and economies of scale reinforced by captivity. Wix has:

• No supply-side cost advantage. Hosting is a commodity; inference is a commodity purchased from third parties (and now partly self-supplied, which changes the cost curve but does not create an advantage others cannot copy — every serious competitor is fine-tuning open models). Wix’s FY2025 gross margin of 68.1% is respectable but is lower than it was in 2018 (79.0%) and 2019 (74.3%), the direction of travel being set by the Business Solutions mix and now by AI.
• Real but modest customer captivity. A live website with a custom domain, an SEO history, a payments integration and a booking calendar is genuinely painful to move. This is the moat, and it is why retention has been good enough to support twelve years of ARPS increases. But it is captivity over a published site, not over the act of building — and AI attacks the building step, which is where the customer is acquired.
• No network effect. A Wix site does not become more valuable because another Wix site exists. The App Market and the Partner/agency marketplace are genuinely two-sided, but neither is dominant and neither has been shown to raise pricing power.
• A real intangible: brand and funnel. This is Wix’s best asset and management is right to emphasize it. The 20-F states that “our investments in building a global, scaled brand since our inception have made Wix synonymous with relevant general keywords on the internet,” which is the honest description of a decade of Super Bowl-scale advertising converted into organic search intent. It is why Wix can run acquisition marketing at a 7–9 month TROI where a startup cannot.

Now run the tests.

Market-share stability — mixed, and the failure is in the number that matters. At the category level Wix passes: ~45% builder share, sustained for years, with the #2 taken private. At the unit level it fails: premium subscriptions went from approximately 6.2 million (31 December 2024) to approximately 6.1 million (31 December 2025) — a decline of roughly 1.6% — while an acquisition (Base44) was being folded into the count and the registered-user funnel grew 7.7%. The company does not disclose a core-Wix-only subscription number, which means the core decline is at least as large as the consolidated one and probably larger. A franchise with genuine captivity does not lose paying units into a growing funnel.

The ROIC test — fails outright. Return on equity and price-to-book are not computable: shareholders’ equity is negative $102.4m and has been negative since at least 2024, a consequence of $1.6bn of treasury stock and an $908.4m accumulated deficit. On the underlying operations, GAAP operating income was $1.8m in FY2025 (0.09% margin), $8.2m in FY2023, and negative $285.4m in FY2022 and $325.5m in FY2021. Only FY2024 ($100.1m, 5.7%) produced a materially positive GAAP operating result. Cumulative GAAP earnings across Wix’s entire life as a public company are deeply negative. Adjusting to something defensible — GAAP operating income plus acquisition-related expense and amortization, but with stock compensation fully expensed — gives roughly $144m of FY2025 operating income, a 7.2% margin. Against roughly $2.9bn of cumulative capital raised and retained (paid-in capital $2.40bn plus $1.15bn of converts, less cash), that is a return in the low single digits. On Greenwald’s own test, there is no moat here, because the claimed moat has never shown up in a return on capital.

Head to head.

• Versus GoDaddy: Wix wins decisively on product, design and builder share (45% vs ~10%); GoDaddy wins decisively on economics (~22% ROIC, $1.58bn of FCF, an 89%-recurring registrar annuity). GoDaddy is, on its own numbers, a genuinely good business priced as a slowly-dying one. Wix is the mirror: a better product priced as a broken business — because its P&L is broken.
• Versus Shopify: not a real contest in commerce. Wix GPV is $3.8bn a quarter against Shopify’s order of magnitude more.
• Versus Squarespace (private): comparable design-led positioning, smaller, and now unobservable.
• Versus Lovable, Bolt.new, v0, Claude Code, Cursor, Replit: the new front. Base44 is, per management, “the largest AI-powered app creation platform in North America by market share” (source: Similarweb desktop unique visitors, October 2025 — a traffic proxy, not a revenue measure). It is genuinely winning on that metric. It is also being funded at a cost that took the consolidated operating margin to 5%.

Is the proprietary LLM a moat? Management frames it as “a flywheel that we believe is unmatched.” The claim has two parts. The cost part is credible and important: owning the model for Harmony removes third-party inference cost and puts unit economics back under Wix’s control — the CFO’s language is that it is “marginal cost” relative to frontier models. The differentiation part is weaker. Every serious competitor can fine-tune an open model on domain data; what Wix uniquely has is user feedback at scale on published Wix sites, which is a real but narrow data asset. And the model has not been applied to Base44, where the AI cost actually is, with no timeline offered because “it’s a bigger or more complex undertaking.”

Verdict: a real but narrow and non-widening moat — brand and modest switching costs over a commodity core — that has never produced a return on capital, and that is now being attacked at the point of customer acquisition rather than the point of retention. Wix is the leader of a category whose defensibility is being tested for the first time. Leadership is worth something; on the evidence of the last five quarters, it has not been worth a profit.

5. Growth History and Forward Opportunities

The record. Revenue: $603.7m (2018) → $757.7m (2019, +25.5%) → $984.4m (2020, +29.9%) → $1,269.7m (2021, +29.0%) → $1,387.7m (2022, +9.3%) → $1,561.7m (2023, +12.5%) → $1,760.7m (2024, +12.7%) → $1,993.0m (2025, +13.2%). The COVID spike and its hangover are visible; the post-2022 record is a remarkably stable low-teens.

But the composition has changed fundamentally. In the growth years, Wix added paying subscriptions. Now it does not.

All growth is ARPS. The 20-F is explicit: revenue growth “was also driven by an increase in average revenue per premium subscriptions (‘ARPS’)… driven by the onboarding and retention of high quality users that purchase higher priced packages, better adopt Business Solutions, and drive compounding GPV as well as the price increase implemented in early 2024.” That is three levers — mix, attach and price — applied to a base that shrank. Two of the three are finite. A price increase can be taken once every few years before it accelerates the churn that is already shrinking the base; the deliberate shift to “high-intent users” flatters the average by shedding the tail, which improves ARPS while reducing units, exactly the pattern in the data. Only attach (Business Solutions into the existing base) compounds — and Business Solutions carries a 32% gross margin, so attach growth dilutes the blended margin even when it works.

The cohort data is genuinely good and deserves weight. The Q1-2026 cohort of 6.4 million users produced $52m of bookings in its first three months, up 46% on the Q1-2025 cohort’s $36m. Management attributes the improvement to “noticeably improved year-over-year and quarter-over-quarter conversion of new users into paid subscriptions, powered by Harmony,” with the caveat that Base44 is a “meaningful contribution.” Stripping Base44 out, management says core Wix new-cohort bookings grew “nearly as strong as what we saw in the Q1 2025 cohort last year despite a slightly smaller user base” — i.e. roughly flat cohort quality on a smaller funnel. That is a fair-to-decent core result and a very strong consolidated one. It is also unauditable: the split is not disclosed.

The three forward engines.

1. Base44 — the genuine success. Acquired 13 June 2025 for $92.2m of headline consideration; $50m ARR target for year-end 2025 (raised mid-year), $100m achieved in early March 2026, $150m by mid-May 2026. Two million users by November 2025 and “more than one thousand new paying subscribers joining every day.” Management claims North American market leadership among AI app builders on a Similarweb traffic basis. This is the fastest ramp in Wix’s history by a wide margin and it materially expands the addressable market from websites to software applications. It is also the direct cause of the margin collapse, and its stand-alone gross margin is undisclosed.
2. Wix Harmony — the AI site builder, rolled out across main geographies in late January 2026, now running on Wix’s own proprietary LLM. Management reports “strong early conversion and monetization.” Independently, management also conceded on the Q1 call that partners “are pointing out to us specific holes, if you may, or missing capabilities” in Harmony “because we build Harmony for self creators and not in a view of partners” — which is both an honest disclosure and a partial explanation for the Partners slowdown.
3. Partners / Wix Studio — the engine of 2023–25 (+23% in FY2025 to $750.3m) that has now stalled. Growth decelerated to +19% in Q1-2026 and then slowed further in late May and early June. Management’s explanation is a deliberate marketing pullback plus product delays caused by the war (“more than 60% of Wix employees located in Israel… we did experience a headwind to the productivity of our team over the past couple of months. As a result, certain product time lines for our partners’ audience have been pushed out”). The alternative explanation — professional builders migrating to Cursor, Claude Code and Lovable — is not one management can disprove, and its own 20-F now lists exactly those tools as competitors.

Verdict: low-quality growth of adequate magnitude. Thirteen percent revenue growth is respectable; growth achieved entirely through price and mix on a contracting paying base, with the fast-growing new product carrying a structurally worse cost structure and the previously fastest-growing channel decelerating, is not the same thing as thirteen percent growth from adding customers. The cohort data argues the funnel still works. The subscription count argues the base does not hold. Until those two reconcile, the growth is not underwritable.

6. Financial Quality

This is the section where the story is decided, and it has four parts.

6.1 The income statement: profitability that has never actually existed

Three observations.

First, the FY2025 profit is an accounting artifact. Pre-tax income was $1.089m. Net income was $50.646m because of an income tax benefit of $51.047m, of which Note 14 discloses $70.920m as the release of the deferred-tax valuation allowance (total allowance $72.814m → $0). Without the release FY2025 was an approximate $20m net loss. The cash-flow statement carries the corresponding signature — changes in deferred income taxes, net of −$91.742m, backed straight out of operating cash flow.

The basis for the release deserves scrutiny. Management justified it on “transition to sustained profitability in recent years, which provides objective and verifiable positive evidence, together with expectations of future taxable income.” It was booked as of 31 December 2025 — in a quarter that reported a GAAP operating loss of $72.6m and a net loss of $40.2m — and was followed by a $57.5m loss in Q1-2026. Two consecutive quarterly GAAP losses immediately after asserting sustained profitability creates live re-establishment risk on a $144.8m gross deferred-tax asset (of which $65.2m relates to share-based compensation, itself a function of a stock price that has fallen 84%). This is a genuine open question, not a rhetorical one.

Second, the margin trajectory is the fact that matters. Non-GAAP operating margin: FY2024 22% → FY2025 19% → Q4-2025 15% → Q1-2026 5%. The Q1-2026 bridge: operating expenses rose 48.7% year over year against revenue growth of 14.3%, driven by S&M +78.9% ($111.6m → $199.6m) and R&D +39.8% ($127.5m → $178.2m). Roughly $24m of the S&M step-up was two Super Bowl advertisements (Harmony and Base44), which management says will not repeat, and approximately $90m was acquisition marketing at a stated 7–9 month blended TROI. The R&D step-up is substantially acquisition-related expense: the 20-F attributes $113.8m of FY2025’s $150.2m R&D increase to acquisition-related costs “mainly attributed to our acquisition of Base44,” and FY2025 total acquisition-related expense was $131.6m (with $90.0m of that in Q4 alone) versus $37.9m in Q1-2026.

Third, stock compensation is a real and enormous cost that the non-GAAP presentation removes. SBC has been essentially flat at $221–241m for five straight years — 11.9% of FY2025 revenue. It is not a start-up ramp; it is a permanent 12% tax on revenue, and it is the reason the buyback has retired so few net shares (see Capital Allocation). Any analysis of Wix that adds SBC back and then celebrates the resulting margin is not analysis.

6.2 Gross margin: a structural change, not a blip

Creative Subscriptions cost of revenue rose 40.0% year over year in Q1-2026 ($56.1m → $78.5m) on Creative revenue growth of 13.2%. That is the AI inference cost arriving. Management’s framing — “Core Wix Creative Subscriptions gross margin was stable as AI costs remained minimal” — is defensible only because Base44 sits inside the Creative Subscriptions segment, as the 20-F’s own definitions confirm. If the core is genuinely stable at ~83%, then Base44 at a ~$150m ARR run-rate is dragging nearly four points off a $1.5bn segment, which implies a Base44 gross margin materially below breakeven at the point of the disclosure. The CFO’s own language supports this: “we already positive in terms of the overall gross margin for base,” and “we managed to reduce the cost significantly.” Already positive is not a phrase used about an 80%-margin business.

This is the single most important undisclosed number in the company. Wix’s twenty-year economic identity is a near-zero-marginal-cost hosting business. Base44’s is a metered AI product with a variable input cost. They are not the same business and consolidating them into one segment line obscures exactly the thing an investor needs to know.

6.3 Cash flow: real cash, questionable durability

FY2025 free cash flow of $573.0m — 28.8% of revenue — is the number the bull case rests on, and it is genuinely cash that arrived in the bank. But its composition is unusual. Within the $582.9m of operating cash flow sit +$204.2m from “increase in accrued expenses and other current liabilities” (FY2024: +$3.1m) and +$45.4m from operating lease liabilities (FY2024: −$33.1m). On the balance sheet, “other long-term liabilities” went from $16.0m (2024) to $200.1m (2025), and Note 9 shows accrued expenses rising from $63.2m to $146.7m including two brand-new lines: “accrued employee compensation related to business combination” of $29.5m and current contingent consideration of $5.8m. The Base44 contingent-consideration liability itself was $89.5m at 31 December 2025, up from $74.1m at acquisition.

The read: a material portion of FY2025’s headline free cash flow is an accrual build for Base44 earnout and retention obligations that Wix has expensed but not yet paid. This is not an inference imposed on the accounts; management said it twice. The Q3-2025 release attributed the ~$600m FCF guide to “our anticipated increase in bookings, higher operating expenses, and continued working capital benefits.” The CFO said on 13 May 2026 that the Q1 S&M step-up “resulted in a large onetime step-up in working capital benefits.” A working-capital benefit driven by not yet paying for something is a timing difference, and timing differences reverse. Q1-2026 operating cash flow of $78.5m against $145.5m a year earlier — down 46% on revenue up 14% — is the reversal beginning.

The correct owner-earnings figure. Guided FY2026 free cash flow “excluding acquisition and restructuring costs” is ~$420m. Subtract $230m of stock compensation and the owner figure is roughly $190m. Further deductions are coming and are not in that number: $30–35m of restructuring cash (mostly 2H-2026), the Base44 earnout and retention cash as it converts, and roughly $100m per year of new interest cost and foregone interest income that management itself quantified on the Q1 call as the price of the tender.

6.4 Balance sheet: a deliberate weakening

At 31 March 2026 (pre-tender settlement): cash, deposits and marketable securities $2,022.7m; convertible notes net $1,127.1m; total shareholders’ deficiency $(102.4)m; treasury shares $1,600.2m; accumulated deficit $(908.4)m; deferred revenue $903.3m.

Post-tender, management guided pro-forma Q2 cash to “about $900 million”, of which approximately NIS 1bn (~$300m) in Bank of Israel Bills plus a $120m deposit is pledged as collateral to Hapoalim. Against that sit $1,150m of 0.00% converts due September 2030 (conversion price $210.49 — the stock is 74% below it, so these are pure debt) and a drawn $500m facility carrying a Bank Debt / Free Cash Flow ≤ 2.0x covenant. The CFO’s own summary on 13 May: “following the completion of our $1.6 billion tender offer in early April, we are now in a net debt position… our goal is to return to a net cash position efficiently.”

The covenant deserves a moment. Bank Debt / FCF ≤ 2.0x on a $500m facility requires trailing free cash flow (as defined, excluding one-time items) of at least $250m. Guided 2026 FCF is $420m, so there is headroom — but the headroom is one bad year deep, and the definition excludes “one-time/non-recurring expenses,” which is a negotiated term rather than a bright line.

Verdict: economics that do not improve with scale. Revenue is up 47% since 2022 and GAAP operating income is $1.8m. Gross margin peaked in 2019 at 74.3% and is now 65.3%. Stock compensation has been a flat ~$230m for five years regardless of revenue. Free cash flow is real but is currently flattered by accruals for obligations that will be paid, and the company has just voluntarily added ~$100m a year of financing cost. The honest summary is that Wix converts a good gross margin into cash and then spends all of it — on marketing, on stock compensation, and now on interest — leaving nothing that compounds for the owner.

7. Capital Allocation

This is the weakest part of the company and the reason the memo’s tone is what it is.

7.1 The 2026 recapitalization, in sequence

The internal contradiction. On 5 March Wix sold stock to a new institutional holder at $79.59. Four weeks later it bought stock back at $92.00 — 15.6% higher — using, among other sources, the proceeds of the sale it had just made and a bank loan. Both transactions cannot represent the same judgement about intrinsic value. Either the March placement was underpriced (in which case the company sold cheap) or the April tender was overpriced (in which case it bought dear). On the evidence of the subsequent nine trading days — the stock closed at $63.84 on 14 April, 30.6% below the tender price — it was the latter.

What it cost continuing holders. Hold today’s enterprise value constant and reverse the tender: add back ~$1.61bn of cash, remove the ~$500m of drawn facility, restore ~17.5m shares. The implied price is roughly $66 versus the actual $55.03 — the tender transferred on the order of $11 per share, ~$460m, about 20% of the current market capitalization, from continuing holders to tendering ones. (ASSUMPTION: enterprise value is invariant to the financing decision. Management’s own ~$100m per year of foregone interest and new interest expense makes the true cost worse.) This is presented as arithmetic under a stated assumption, not as a claim of certainty.

7.2 The long record: $3.21 billion for 15 million shares

Basic weighted-average shares were 57.00m in FY2021; shares outstanding are 41.85m today. The net reduction is ~15.2m shares against $3.21bn of gross repurchases less $260m of placement proceeds — roughly $195 of net cost per share retired, against a $55.03 stock. The gap between gross buyback and net reduction is stock compensation: ~$230m of issuance every year, forever, quietly re-diluting what the buyback retires. Wix has spent 1.4 times its entire current market capitalization on repurchases and its share count has fallen by 27%.

To be fair to management: buying back stock at $62–$76 in 2022 was good, and the idea of retiring 30% of the equity of a cash-generative business is not wrong in principle. The execution is what fails. Repurchases were smallest in the cheapest year (2023, $127m at $75–$123) and largest in the most expensive (2026, $1.61bn at $92). That is the textbook procyclical pattern.

7.3 M&A

Base44 (13 June 2025). Total purchase consideration $92.158m — $18.058m cash plus $74.100m of liability-classified contingent consideration, allocated to $80.893m goodwill and $14.130m intangibles. Separately, the founder and certain employees are eligible for $42.988m of cash retention bonuses through 2028 and $8.0m of RSUs/PSUs, expensed as post-combination compensation. The earnout runs on revenue-related metrics for calendar 2025 through 2028 and is marked to fair value each quarter via a Monte Carlo model; it had risen to $89.5m by 31 December 2025, and the CFO told analysts on 13 May that “every quarter, we are going to have increased the provision for the M&A… up to a certain cap.”

As a product purchase this looks outstanding: ~$150m of ARR for a headline $92m. As a cost forecast it has been poor — the associated charges are the single largest driver of FY2025’s 30% R&D increase and a meaningful share of the FY2025 free-cash-flow accrual build. And there is an under-appreciated feature: the better Base44 performs, the larger the earnout and the greater the future cash drain. The bull case and the cash-flow bridge fight each other.

Hour One AI Ltd. (May 2025). Acquired; terms not separately disclosed; not material to the accounts.

7.4 Insiders and incentives

Wix is a foreign private issuer, so its officers and directors were exempt from Section 16 until the Holding Foreign Insiders Accountable Act took effect — the thirteen Form 3s in the corpus are all dated 18 March 2026. There is therefore no pre-2026 Form 4 record, and the absence of recorded open-market purchases during the 84% drawdown is partly mechanical, not purely behavioural. That caveat matters and is stated plainly.

What is observable is the Form 144 record. Across 45 notices from May 2023 to September 2025, 1,823,773 shares were noticed for sale at a weighted-average price of $176.47, aggregating ~$321.8m. Named sellers include CEO Avishai Abrahami (150,000 shares at ~$152.32 in June 2025), President Nir Zohar (~110,225), CFO Lior Shemesh (~133,549), CMO Omer Shai (~102,403), Shelly Meyer (~51,357), Yaniv Even Haim, and directors Allon Bloch and Francesco de Mojana. The largest prints came at the top: 81,000 shares at $214.55 on 2 January 2025 and 50,000 at $238.89 on 3 February 2025.

The last Form 144 was filed on 19 September 2025 at $180.99. The 52-week closing high of $184.24 came three trading days later. Nothing has been filed since. Insiders stopped selling within a week of the peak and — on the record available — have not bought a share on the way down.

Beneficial ownership at 31 January 2026: Abrahami 2,014,244 shares (3.6%, of which 1,060,337 are options/RSUs); all officers and directors as a group 4,098,074 (7.2%). For a twenty-year-old founder-led company that is modest, and the fact that more than half the CEO’s disclosed stake is unvested equity means his exposure is more to the option on the stock than to the stock.

Verdict: management has not allocated capital intelligently. The company has produced roughly zero cumulative GAAP profit in its public life, issued ~$230m of stock a year, spent $3.21bn buying that stock back at an average net cost of ~$195 a share, executed the largest single repurchase in its history at $92 four weeks after selling stock at $79.59, and cut guidance and 20% of the workforce ten weeks after the board declared “continued confidence.” Whatever else is true about Wix, this is the part of the record that should carry the least benefit of the doubt.

8. Changes and Headwinds — Last Two Years

2025-05 — Hour One AI Ltd. acquired. Small AI-video/avatar capability; terms undisclosed; immaterial financially, indicative strategically.

2025-06-13 — Base44 acquired. The pivot. $92.2m headline, earnout to 2028, and the beginning of a cost structure Wix had not previously carried.

2025-08-15 — 2025 Convertible Notes ($575.0m) fully repaid, followed in September 2025 by the issuance of $1,150.0m of 0.00% converts due 2030 at a $210.49 conversion price. At issuance the stock was near $180; the notes were struck 17% above the 52-week high the market would ever see. They are now pure five-year debt at a zero coupon — arguably the single best financing decision in the sequence, since the company borrowed $1.15bn at 0% and the option it sold is worthless.

2025-11-19 — Q3-2025 results, stock −19.9%. Bookings guidance raised to $2,060–2,078m; revenue guidance narrowed; but non-GAAP gross margin guided down to 68–69% “due to higher AI costs as a result of better than anticipated demand and increasing usage volume for Base44,” non-GAAP operating expenses guided to ~50% of revenue, and FY2025 FCF guided to ~$600m helped by “continued working capital benefits.” The market decided it was paying for growth with margin, and repriced accordingly.

2026-01-28 — $2bn repurchase authorization, with the notable disclosure that it may be funded “from raising additional capital, including through the issuance of debt, equity or equity-linked securities.”

2026-03-03/04/05 — the recapitalization: $500m Hapoalim facility, $260m Durable Capital placement at $79.59, $1.75bn Dutch auction launched. Stock +25% over two sessions.

2026-04-01 — tender clears at $92.00 for ~17.5m shares. Share count −30%; net debt for the first time.

2026-05-13 — Q1-2026, stock −27.1%. S&M +78.9%; operating margin 21% → 5%; GAAP loss $57.5m; FCF halved. Management maintained full-year guidance.

2026-05-28 / 2026-06-08 — the reset. A 20% workforce reduction (1,000 employees) communicated internally on 28 May and disclosed by 6-K on 8 June, together with “the scaling down and/or discontinuation of certain activities, initiatives, products, and subsidiaries.” FY2026 bookings growth cut from mid-teens to low-teens; revenue growth to low-to-mid teens; ~$50m of bookings and ~$25m of revenue removed; ~$70m of in-year and ~$150m of run-rate cost savings identified; $30–35m of pre-tax restructuring charges, “a substantial majority… cash expenditures for the payment of severance.” FCF guidance raised $20m to ~$420m. Cause given: the realignment plus “a more pronounced slowdown, beyond our previous expectations, in the growth of our Partners business during the second half of May and early June.”

2026-07 — the sell side capitulates and the plaintiffs’ bar arrives. Morgan Stanley cut Wix from Overweight to Equal-weight with a price target reduced from $112 to $60, noting Base44 and Harmony “did not improve the company’s overall growth profile.” Raymond James cut from Strong Buy to Outperform on 30 July with a target of $70 purely on valuation after a 32% one-month rally; the stock fell 9.4%. Separately, Hagens Berman announced a securities class action with a class period of 19 February 2025 to 12 May 2026 (lead-plaintiff deadline 22 September 2026), alleging Wix “overstated the competitiveness and performance” of Base44 and Harmony and “understated the costs associated with developing and promoting them.”

The war. More than 60% of Wix’s employees are in Israel. Management disclosed on 13 May that a conflict beginning in late February 2026 caused “a headwind to the productivity of our team over the past couple of months” and pushed out partner-facing product timelines. Wix also has 594 contractors and 11 employees in Ukraine. This is a recurring, unhedgeable operational exposure, and it is compounded by a $64m FY2026 expense headwind, net of hedging, from shekel strength — a currency exposure that runs against the company precisely because its cost base is Israeli and its revenue is dollar.

Verdict: these developments weaken the thesis materially. The two-year record is an expensive, half-completed strategic pivot; a balance sheet deliberately levered at the top of the price range; a growth channel (Partners) that has stalled; a 20% headcount reduction that is genuine cost discipline but also an admission that the January plan was wrong; and now litigation. The one clear positive is Base44’s ramp. The one clear structural positive is the 0% converts.

9. Risk Analysis

The two risks that dominate are the first and third, and they are linked. If AI commoditizes the builder, the paying base accelerates its decline and no amount of cost cutting saves the multiple. If the free cash flow that anchors the valuation turns out to have been an accrual timing benefit, the 7.3x EV/FCF that makes the stock look cheap was never real. Both are testable within two to four quarters.

10. Valuation Discussion (Embedded Expectations)

No price target and no recommendation appear in this section.

10.1 Capital structure and enterprise value

10.2 Multiples, and what the stock’s own history says

FY2026 revenue on post-cut guidance is roughly $2.20–2.26bn (low-to-mid-teens growth on $1,993m, less the ~$25m realignment impact). Guided free cash flow excluding acquisition and restructuring costs is ~$420m.

A necessary correction to the percentile. AZI’s valuation_index reports TTM sales per share of $36.4347, which implies ~54.7m shares — the pre-tender count. On the actual 41.85m shares, TTM sales per share is ~$47.62 and P/S is ~1.16x, materially cheaper than the 1.51x AZI reports. So the 1.69th-percentile reading, already the cheapest in the stock’s ten-year history, is if anything understated on a per-share basis. But this is precisely where per-share cheapness misleads: the share count fell because the company spent $1.61bn of cash and borrowed $500m to make it fall. EV/sales does not benefit from the buyback, and EV/sales is the honest measure. The P/E and P/B percentiles are null and correctly so — GAAP EPS is negative and book equity is negative.

10.3 Embedded expectations

At ~$3.05bn of enterprise value against ~$190m of normalized owner free cash flow, a single-stage reverse test at a 10% cost of capital implies required perpetual owner-free-cash-flow growth of roughly 3.8%. Set against reported bookings growth of ~15% and ARR growth of ~15%, that looks undemanding — which is the entire bull case in one line.

It looks less undemanding once three things are held in view simultaneously. First, the ~15% growth is delivered on a paying subscription base that shrank, so it is price and mix, and price and mix are finite. Second, the $190m is itself flattered: it does not deduct restructuring cash, the Base44 earnout as it converts, or the possibility that FY2025’s accrual build reverses. Third, the business has an entirely new variable cost (inference) whose steady-state level is unknown even to management, who declined to give a Base44 gross-margin figure or a timeline for extending the proprietary model to it.

What the market appears to be underwriting correctly: that low-teens bookings growth continues; that the $150m of run-rate cost savings is real; that the 0% converts are cheap money; that Base44 has genuine product-market fit.

What the market may be underwriting incorrectly, in either direction: the durability of free cash flow (the accrual question), the terminal gross margin of an AI-heavy revenue mix, and — the largest — whether the paid website builder is a category with a future. Nothing in the current price distinguishes between “Wix is a 13%-growth SaaS business with a temporary investment year” and “Wix is a melting franchise that has borrowed to buy back stock.”

10.4 Scenarios (2027 exit; per share on 41.85m shares; net debt held at ~$650m)

Assumptions. Bear: AI commoditizes the builder; the paying base declines mid-single-digits; growth falls to low single digits; margin recovers only to the pre-Base44 core level and the multiple compresses to a terminal-decline rate. Base: growth settles at 8–10% as Partners stabilizes and Base44 matures; the $150m run-rate saving holds; FCF margin returns to the low-20s ex-acquisition; a 14x owner-FCF multiple, roughly the market multiple for a low-growth software annuity. Bull: Base44 exceeds $400m ARR, Harmony converts durably, the proprietary model is extended to Base44 and inference cost collapses, and the market re-rates to a growth multiple.

The dispersion is the finding. Bear to bull spans roughly 15x. That is not an artifact of aggressive assumptions; it is the honest consequence of a business whose central question — is AI generation a market I win or a solvent that dissolves me — has not been answered and cannot be answered from the outside today. A valuation range this wide is a statement that the security is a call option on an unresolved technology transition, not a cash-flow asset that can be discounted with confidence.

11. Variant Perception

Consensus. The sell side has capitulated to roughly the current price rather than defending a higher one: Morgan Stanley cut to Equal-weight with a target reduced from $112 to $60; Raymond James cut on valuation on 30 July with a $70 target after a 32% one-month rally. The prevailing view is that Wix is a structurally sound low-teens-growth business having an expensive investment year, that the 20% workforce cut restores the margin path, and that Base44 is a genuinely valuable asset acquired for almost nothing — but that there is no reason to own it until core growth is visible. That is a “cheap but no catalyst” consensus.

The strongest bull case. Wix is the #1 website builder on earth, with 45% share, 304 million registered users, $1.9bn of ARR growing 15%, a $903m float, and an asset-light structure (capex 0.5% of revenue). It has just retired 30% of its equity and can retire more. Its converts cost 0.00% and will never convert. It has demonstrated, in Base44, that it can build or buy a genuinely new $150m-ARR business inside twelve months — something almost no incumbent software company has done in the AI era. It has just removed $150m of run-rate cost. At 7.3x guided free cash flow and 1.37x sales — the cheapest sales multiple in its history — an investor is paying nothing for Harmony, nothing for Base44’s option value, and less than a market multiple for the core. If the AI panic is wrong, this is a multi-bagger, and the March placement by Durable Capital Partners at $79.59 with an extended lockup is a sophisticated, informed investor saying exactly that.

The strongest bear case. Wix has never earned a real profit. Cumulative GAAP earnings across its public life are negative; the accumulated deficit is $908m; FY2025’s entire GAAP profit was a $70.9m tax-allowance release booked in a loss-making quarter. Its paying subscription base is shrinking even with an acquisition folded in, so all growth is price on a melting base. Its new growth engine carries a variable inference cost the old business never had, and its stand-alone margin is undisclosed — which, on the disclosure record of this management team, should be read as unfavourable. Its free cash flow is flattered by a ~$200m accrual build for obligations it will have to pay. It has just levered a business facing technological disruption in order to buy 30% of its own stock at a price 67% above where it now trades, four weeks after selling stock 15% cheaper — and then cut guidance and fired a fifth of its workforce ten weeks later. Book equity is negative. It is now in net debt with a covenanted bank facility and pledged collateral. And its own annual report names Anthropic, Cursor and Replit as competitors.

The three to five assumptions that actually matter.

1. Does the paying premium-subscription count stabilize? Everything else is second order. 6.2m → 6.1m with an acquisition is the most important disclosed number in the company.
2. What is Base44’s stand-alone gross margin, and does it converge on core Wix? Management asserts “similar margins to our core business in the long-term.” This is unverified and unverifiable from outside.
3. Is the ~$420m FY2026 free cash flow durable, or a working-capital timing benefit? The Base44 earnout and retention obligations are accrued and must be paid by 2028.
4. Is the Partners slowdown a self-inflicted marketing pullback (recoverable) or professional migration to AI dev tools (structural)? Partners is 37.6% of revenue.
5. Will management stop repurchasing stock procyclically? Roughly $390m of the $2bn authorization remains.

Falsification. The bull is falsified if the paying subscription count declines again in FY2026 with Base44 included; if core Wix bookings growth ex-Base44 goes to zero; or if free cash flow undershoots $420m as accruals convert. The bear is falsified if premium subscriptions grow in units; if Base44’s disclosed gross margin exceeds ~60% and rises; or if Partners bookings reaccelerate on the delayed product releases.

The factor-positioning input. The tape is not currently on the bull’s side, and it is important to say so precisely. In FactorsToday’s Base+Sector+Industry model (R² 31.3%), Wix loads Momentum −1.231 and Value +0.071 — it has been expelled from the momentum cohort and has not been adopted by the value cohort. Market beta is 1.399, Cloud Computing 0.859, SmallSize 0.812, LowVolatility −0.507. Risk-adjusted performance is dismal on every horizon: one-year −59.5% with a Sharpe of −0.88 and a −78.1% maximum drawdown; five-year −29.0% annualized with an −86.7% drawdown; three-month −78.3% annualized on 88.8% volatility. Idiosyncratic volatility is 56.9% annualized against a model R² of 26.5% — roughly 73% of the variance is company-specific, which is the statistical way of saying the market is not trading Wix as a software beta but as a single-name binary. Its factor-similar peers (NCNO 0.95, PD 0.94, QTWO 0.93, NABL 0.93, CXM 0.93, BRZE 0.92) are de-rated mid-cap SaaS, not internet marketplaces.

GoDaddy’s factor profile shows the identical signature — momentum expelled, value not yet loading. That is the no-man’s-land between de-rated growth and recognized value, where contrarian re-ratings often begin — but the same data warns that the trend has not turned. The reading applies to Wix with more force, because Wix’s drawdown is deeper (−84% versus −59%), its volatility is nearly triple, and its cash flows are far less certain. This is where consensus may be offsides in both directions at once: under-owned by growth investors who have left, and not yet owned by value investors because the earnings are not yet legible. As a positioning fact that is reportable; as a prediction it is interpretation, and it is regime-dependent.

12. Fact vs. Interpretation

13. Open Questions

1. What is Base44’s stand-alone gross margin? Not disclosed. The CFO’s phrase — “we already positive in terms of the overall gross margin for base” — implies a low figure. Without it, the terminal margin of the consolidated business cannot be estimated.
2. What is the core-Wix-only premium subscription count? Not disclosed. The consolidated 6.2m → 6.1m decline with Base44 included means the core fell by more, but by how much is unknown and it is the single most decision-relevant number.
3. What is the cash-payment schedule of the Base44 earnout and the $42.988m of retention bonuses through 2028? The 20-F gives liability balances ($89.5m contingent consideration at year-end 2025, $29.5m of accrued business-combination compensation) but no payment profile. This determines how much of the ~$420m FCF guide survives.
4. How much of the Hapoalim facility is actually drawn — $500m or $600m? The 20-F documents a $500m commitment; an analyst asserted $600m on the Q1 call and was not corrected on the amount. Resolvable on 4 August.
5. Does the deferred-tax valuation allowance survive? It was released on 31 December 2025 citing “sustained profitability,” immediately after a $72.6m quarterly operating loss and immediately before a $57.5m one.
6. Is the Partners slowdown recoverable? Management attributes it to its own marketing pullback and war-related product delays. If it is instead professional migration to Cursor/Claude Code/Lovable, 37.6% of revenue is structurally impaired.
7. When, if ever, is the proprietary LLM extended to Base44? That is where the inference cost sits. Management would give no timeline: “it’s a bigger or more complex undertaking.”
8. Will the remaining ~$390m of the $2bn authorization be spent, and at what discipline? The stated funding policy explicitly contemplates further debt or equity issuance.
9. Why did the CEO not present on the Q1-2026 call? Disclosed as the birth of his son, which is a complete and reasonable explanation. Worth confirming his presence on 4 August.
10. What are the actual TROI outcomes on the Q1-2026 marketing cohort? Management projects 7–9 months on ~$90m. That is a forecast, not a result, and it will be observable in the Q3 and Q4 cohort bookings series.

14. What Must Be True

Bull case — what must be true

1. The paying subscription base must stop shrinking. Premium subscriptions must be flat or growing at year-end 2026 with Base44 included, and ideally management must begin disclosing the core-only figure.
2. Base44 must converge toward core gross margins. Management’s claim of “similar margins to our core business in the long-term” must be evidenced by a rising, disclosed segment margin — most plausibly via extension of the proprietary LLM to Base44.
3. The ~$420m FY2026 free cash flow must be delivered and survive the earnout conversion, with the $150m run-rate saving visible in the Q3 and Q4 non-GAAP operating margin.
4. Partners must reaccelerate on the delayed product releases, demonstrating that the slowdown was a marketing decision rather than a migration.
5. Capital allocation must become countercyclical — the remaining authorization spent at prices below intrinsic value, or not at all.

Falsification test for the bull: if FY2026 ends with premium subscriptions below 6.1 million, or free cash flow below ~$380m, or Partners bookings growing slower than the corporate average, the “temporary investment year” story is wrong and the business is a melting franchise that has levered itself.

Bear case — what must be true

1. AI generation must genuinely substitute for the paid builder, not merely coexist with it — visible as continued unit decline, falling retention, or ARPS growth that stalls because price increases start driving churn.
2. Base44’s inference cost must remain structurally high, so that revenue growth does not convert into gross profit growth.
3. The FY2025 free-cash-flow accrual must reverse materially, taking reported FCF meaningfully below guidance in 2026–27.
4. The Partners channel must be structurally, not cyclically, impaired.
5. Management must continue to allocate capital procyclically, spending the remaining authorization into strength.

Falsification test for the bear: if Wix reports growing premium subscriptions, a disclosed Base44 gross margin above ~60% and rising, free cash flow at or above $420m with the earnout being paid in cash out of it, and Partners bookings reaccelerating, then the AI-disruption narrative is wrong, the 2026 investment year was exactly what management said it was, and a business at 1.37x EV/sales with 15% bookings growth is materially mispriced.

The honest summary of both. The bull and bear cases are not two readings of the same evidence; they are two different futures for the category, and the evidence available today is genuinely consistent with both. That is why the scenario range in the Valuation section spans roughly 15x, and why the next two to four quarterly prints — starting on 4 August 2026 — carry more information value than any further work on the current disclosure.

15. Source Appendix

See Appendix B below for the full source list with URLs and access dates.

This article contains no investment recommendation and no price target; the labelled Claude's Take block at the top is the author’s own subjective view. All third-party analyst price targets referenced are reported as facts about the market and are not adopted. Nothing here is investment advice.

APPENDIX A — Standard Diligence Questionnaire

Wix.com Ltd. (NASDAQ: WIX) · Published 2026-08-01 · Price $55.03

A standard diligence questionnaire applied to Wix. Labels: Fact / Interpretation / Assumption.

General

What thoughtful questions have other investors asked about this company?

The questions asked on the Q1-2026 call (13 May 2026) map closely onto the memo’s open questions, which is itself informative — the buy side is asking exactly the right things and not getting answers.

• Brad Erickson (Arvest): “investors are worried here about the competitive risk… notice the lack of visible customer growth, even when you put together the core and BASE44 — what would you guys point investors to… why that competitive risk is not the case?” The answer did not address the unit-growth point. This is the single best question asked of Wix in 2026 and it went unanswered.
• Deepak Mathivanan (Cantor): “when do you see the potential to deploy this [proprietary LLM] in BASE44, which is having significant influence [inference] cost?” Answer: no timeline; “a bigger or more complex undertaking.”
• Trevor Young (Barclays): “$1.6 billion in cash went out here in 2Q, $600 million came in from the credit facility… pro forma here in 2Q, are we around $900 million of cash?” Answer: “yes, it’s about $900 million.” The $600m figure was not corrected against the $500m in the filing.
• Needham: “on the acquisition earn-outs… I think there was $37 million in the first quarter.” Answer: “every quarter, we are going to have increased the provision for the M&A… up to a certain cap.”
• Multiple analysts asked whether Wix Partners are adopting Base44 or leaving for third-party AI tools. Management confirmed “a decent amount of partners usage on BASE44 — it’s not marginal” and separately that partners “are using some AI platforms.”
• The securities class action (Hagens Berman, class period 19 February 2025 – 12 May 2026) is in effect the same question in legal form: did management overstate Base44/Harmony competitiveness and understate their cost?

Cyclicality & Earnings Nature

Are earnings at a cyclical high or low? Interpretation: at a low, but it is a self-inflicted investment trough rather than a demand trough. Non-GAAP operating margin has fallen from 22% (FY2024) to 5% (Q1-2026) while revenue grew 14%. The compression is entirely spend-side: S&M +78.9% and R&D +39.8% year over year in Q1-2026. A 20% workforce reduction with ~$150m of run-rate savings is now in place, so reported margin should recover mechanically in 2H-2026. Whether it recovers to 22% depends on Base44’s gross margin, which is undisclosed.

Driven by the external environment or internal actions? Overwhelmingly internal: the Base44 acquisition, the Harmony launch, a >$20m Super Bowl campaign, and ~$90m of quarterly acquisition marketing. External factors are secondary but real — SMB macro softness (GPV +12%, described as “soft”), a $64m FY2026 shekel FX headwind, and war-related productivity loss with >60% of employees in Israel.

How stable are revenues? Structurally very stable. This is a subscription business collecting cash upfront with $903.3m of deferred revenue, ~6.1m paying subscriptions, and $1.90bn of ARR. Revenue growth has been in a 12–14% band for four consecutive years. The instability is in profitability, not revenue.

Outlook for products/services? Bifurcated. Base44 is the fastest-ramping product in company history (0 → ~$150m ARR in ~11 months). Wix Harmony shows “strong early conversion.” The Partners/Wix Studio channel — 37.6% of FY2025 revenue at +23% — has decelerated to +19% and then slowed further in late May/early June 2026.

How big will this market be — growing, shrinking, domestic or international? Global and majority non-US. The website-builder market is mature and growing at low-double-digits; the AI application-generation market is new, growing at triple digits, and being contested by parties with far larger balance sheets. Interpretation: the total pool is growing; the profit pool is being competed away (see the capital-cycle discussion in the Industry Dynamics section).

Business Quality & Competitive Moat

Is the industry getting more or less competitive? Decisively more. Wix’s FY2025 20-F added an entirely new competitor category absent from prior filings: “‘vibe coding’ solutions for developers, such as Claude Code, Cursor and Replit, and AI-powered low-code/no-code application builders, such as Lovable, Google AI Studio, Bolt.new, and V0.” Lovable is at roughly $500m ARR. The #2 builder, Squarespace, was taken private by Permira for $7.2bn in October 2024.

How profitable is the business (ROIC, ROE)? ROE and P/B are not computable — shareholders’ equity is negative $102.4m. GAAP operating income was $1.8m on $1,993m of revenue in FY2025 (0.09%). Adjusting for acquisition-related expense and amortization but fully expensing stock compensation gives roughly $144m of operating income, a 7.2% margin. Against ~$2.9bn of capital raised and retained, that is a low-single-digit return. Cumulative GAAP earnings across Wix’s entire public life are negative (accumulated deficit $908.4m). This is the answer that most damages the moat claim.

How profitable is the industry — how many competitors, what barriers to entry? Poor and worsening; effectively no barrier to entry at the retail layer. GoDaddy earns ~22% ROIC on the registrar annuity, not on building. Shopify earns on commerce scale. Nobody earns an attractive return on website building itself, which is the definitional problem.

Can the business be easily understood? Yes — freemium SaaS, cash upfront, cohorts bought with marketing at a target payback. It is one of the more transparent business models in software. The 2026 complication is that Base44, with a different cost structure, is consolidated inside Creative Subscriptions with no stand-alone disclosure, which has made the accounts materially harder to read.

Can it be undermined by foreign low-cost labour? Not by labour — Wix’s own labour is largely Israeli and expensive (a $64m FX headwind proves the point). It can be undermined by cheap intelligence: AI generation that makes site building free at the margin.

Do brands matter? Yes — this is Wix’s best asset. The 20-F states that a decade of global brand investment has “made Wix synonymous with relevant general keywords on the internet,” which is why the company can run acquisition marketing at a 7–9 month payback where a startup cannot. It is a real intangible advantage. It is also the only one.

What is the nature of competition? Product, design and brand — not price on the core, and increasingly capability on the AI front. Competition is now asymmetric: Anthropic, Google and Vercel do not need SMB web presence to earn their cost of capital, so they can subsidize the exact capability Wix must monetize.

Customers’ switching costs? Real but moderate, and — critically — asymmetric across the lifecycle. A published site with a custom domain, SEO history, payments and bookings is genuinely painful to move (this is why retention supports repeated price increases). But there is essentially no switching cost at the moment of creation, and creation is exactly where AI competes. That asymmetry is the structural heart of the bear case.

Financial Condition & Balance Sheet

Assets not fully recognized on the balance sheet? Yes, and they are substantial: the Wix brand and the organic search position it commands; 304.2 million registered-user relationships; the Partner/agency ecosystem; the App Market; and — newly — a proprietary LLM built at what management calls “quite small” cost and carried at essentially nothing.

Off-balance-sheet liabilities? Few in the classic sense. Operating leases are on balance sheet ($392.6m ROU asset; $459.6m of lease liabilities, largely the Tel Aviv campus). The contingent items are the ones to watch: the Base44 earnout runs on revenue metrics through 2028, marked to fair value each quarter ($74.1m at acquisition → $89.5m at 31 December 2025) and explicitly expected by the CFO to keep rising “up to a certain cap”; and $42.988m of cash retention bonuses through 2028. Also outstanding: warrants on 816,674 shares at ~$104.73 (deeply out of the money) and the $1,150m 0.00% converts at a $210.49 conversion price.

How conservative is the accounting? Interpretation: mixed, tilting aggressive on presentation. Points against: (i) FY2025 GAAP net income of $50.6m was entirely a $70.9m deferred-tax valuation-allowance release, booked as of a quarter-end that reported a $72.6m operating loss; (ii) the non-GAAP presentation removes $237m of stock compensation and $131.6m of acquisition-related compensation, producing a “19% operating margin” against a GAAP 0.09%; (iii) headline free cash flow was helped by a ~$200m accrual build that management itself twice described as “working capital benefits”; (iv) Base44 is consolidated inside Creative Subscriptions without stand-alone margin disclosure. Points in favour: the earnout is expensed rather than capitalized; the segment cost lines are given; the auditor issued a clean opinion; and management has been candid in prose about AI cost pressure — the Q3-2025 release volunteered the gross-margin cut.

How CapEx-hungry is the business? Barely at all: FY2025 capex of $9.9m on $1,993m of revenue (0.5%), down from $63.0m in FY2023 when the Tel Aviv headquarters was being built. This is a genuinely asset-light model. The relevant “capex” is now marketing and AI inference, both of which run through the P&L, which is why reported operating income and reported free cash flow diverge so violently.

Capital Allocation & Management

How much FCF does the business generate, how does management use it, what is the philosophy? FY2025 free cash flow was $573.0m (28.8% of revenue); FY2026 is guided to ~$420m excluding acquisition and restructuring costs. Essentially all of it — and more — goes to share repurchases. The stated philosophy, from the January 2026 authorization, is to fund buybacks “with cash on hand and future cash generated from its operations or from raising additional capital, including through the issuance of debt, equity or equity-linked securities.” That is an unusually explicit statement of willingness to lever for buybacks, and it was acted on within eight weeks.

Significant acquisitions recently? Base44 (13 June 2025), $92.158m total consideration ($18.058m cash + $74.100m contingent), plus $42.988m of retention cash and $8.0m of equity awards through 2028. Hour One AI Ltd. (May 2025), terms undisclosed. Interpretation: Base44 has been an excellent product acquisition and a poorly forecast cost — it drove $113.8m of the FY2025 R&D increase and is the proximate cause of the margin collapse.

Buying back shares? Aggressively, and — on the record — badly timed. ~$3.21bn spent since 2021 (2021 $200m, 2022 $232m, 2023 $127m, 2024 $466m, 2025 $575m, 2026 ~$1,610m). Basic weighted-average shares were 57.00m in FY2021 and outstanding shares are 41.85m today: a net reduction of ~15.2m shares for a net ~$2.95bn, roughly $195 per share retired, against a $55.03 stock. The 2026 tender cleared at the top of its $80–$92 range; nine trading days after expiry the stock closed at $63.84. Spending was smallest in the cheapest year and largest in the most expensive.

Issuing large amounts of new shares to insiders? Yes — $237.4m of stock compensation in FY2025, 11.9% of revenue, and essentially flat at $221–241m for five consecutive years. This is the reason $3.21bn of buybacks retired only ~15.2m net shares. Separately, $260.0m of ordinary shares plus warrants were issued to Durable Capital Partners and others at $79.59 on 5 March 2026.

Compensation policy of directors/management? As a foreign private issuer Wix files no DEF 14A, so there is no US-style CD&A, no PSU metric table, and no say-on-pay disclosure — a genuine and material transparency gap relative to a domestic filer, and one that prevents the standard “read the incentive metrics” diagnostic. What is observable: officers and directors as a group held 4,098,074 shares (7.2%) at 31 January 2026, and the CEO’s disclosed 2,014,244 shares include 1,060,337 options/RSUs — i.e. more than half of the founder-CEO’s stake is unvested equity, so his economic exposure is more to the option than to the equity.

Motivations of management? Interpretation: founder-led (Abrahami, Zohar and Kaplan since 2006) and evidently product-motivated — the willingness to build a proprietary LLM and to buy Base44 in weeks reflects genuine technical ambition. The record on shareholder orientation is weaker: ~$322m of Form 144 sales at a $176.47 weighted average between 2023 and September 2025, with the last notice filed three trading days before the 52-week high, and no recorded open-market purchases through an 84% drawdown (subject to the FPI caveat that Section 16 applied only from 18 March 2026).

Valuation & Market Data

Is the stock an ADR, MLP, or K-1 issuer? None of these. Wix ordinary shares trade directly on Nasdaq under WIX; there is no ADR ratio and therefore none of the per-share/units mismatch that corrupts vendor data on many foreign issuers. It is an Israeli company and a foreign private issuer, filing 20-F annually and 6-K for interim results. Practical consequences for a US holder: no proxy statement, no Regulation FD, insider Section 16 reporting only from March 2026, and Israeli withholding considerations on any future distribution.

Dividend policy? None. Wix has never paid a dividend and the converts/facility make one improbable. All shareholder return is via repurchase.

How profitable is the business? Restating the answer above because it is the crux: FY2025 GAAP operating margin 0.09%; adjusted for acquisition items but with stock compensation expensed, ~7.2%; company non-GAAP, 19% falling to 5% in Q1-2026. Gross margin 68.1% (FY2025), 65.3% (Q1-2026), versus 79.0% in 2018.

Is net income diverging from cash from operations? Massively, and in both directions over time. FY2025: net income $50.6m against operating cash flow $582.9m — an 11.5x ratio. The bridge is $237.4m of stock compensation, $91.7m of deferred-tax movement, $103.9m of deferred-revenue growth, and roughly $250m of accrued/lease/other liability build. Interpretation: the deferred-revenue portion is a genuine, recurring feature of a prepaid subscription model and should be credited. The accrued-liability portion — largely Base44 earnout and retention obligations expensed but unpaid — is a timing difference that reverses, and Q1-2026 operating cash flow falling 46% year over year is that reversal beginning.

Own-history valuation context. AZI’s valuation_index puts Wix at the 1.69th percentile of its own ~10-year price-to-sales history — the cheapest it has ever been on sales. P/E and P/B percentiles are null (negative TTM EPS, negative book value), correctly so. Two caveats: (i) AZI’s sales-per-share uses a pre-tender ~54.7m share count, so the true post-tender P/S is ~1.16x rather than the reported 1.51x — cheaper still; (ii) per-share cheapness is misleading here because the share count fell only because $1.61bn of cash left and $500m was borrowed. EV/sales of ~1.37x is the honest measure and it does not benefit from the buyback.

Risks & Downside

What factors would cause the stock to decline? Continued decline in the paying subscription count; free cash flow undershooting the ~$420m guide as the Base44 earnout converts to cash; further Partners deceleration; a disclosed Base44 gross margin that is low and not improving; re-establishment of the deferred-tax valuation allowance; escalation in Israel; further procyclical use of the remaining ~$390m authorization; and — most immediately — the Q2-2026 print on 4 August 2026, the first quarter reflecting the workforce cut, the restructuring charge and the post-tender capital structure, into 89% three-month realized volatility.

Risk of a catastrophic loss? Interpretation: moderate and higher than it was twelve months ago, because the company chose to make it so. The balance sheet has gone from ~$1.1bn of net cash to net debt of roughly $750m; book equity is negative $102m; the Hapoalim facility carries a Bank Debt/FCF ≤ 2.0x covenant requiring at least ~$250m of defined free cash flow against a $420m guide — one bad year of headroom; and ~$420m of the ~$900m cash is encumbered as collateral. Offsetting this materially: the $1,150m of converts is zero-coupon and not due until September 2030, so there is no near-term maturity wall and no cash interest on the largest liability.

Chance of a total loss? Interpretation: low. Wix generates several hundred million dollars of cash annually from a $1.9bn ARR base with $903m of deferred revenue, 45% share of its category, no near-term debt maturity, and capex of 0.5% of revenue. Even in the bear case the business is a cash-generative, slowly-declining annuity worth something meaningful — the bear scenario above is roughly $12–28 per share, not zero. Permanent capital impairment from $55 is entirely plausible; a wipeout is not.

Recent News & Events

Has the business environment changed recently? Yes, fundamentally, and twice. First, generative AI moved from a feature Wix could adopt (ADI in 2016) to a substitute technology that both created a new market Wix could enter (Base44, ~$150m ARR) and lowered the cost of the thing Wix sells. Second, and self-inflicted, Wix recapitalized itself in March–April 2026 — $500m facility, $260m equity placement, $1.61bn tender — converting a net-cash balance sheet into a net-debt one at the top of the year’s price range, then cut FY2026 guidance and 20% of the workforce ten weeks later.

Significant acquisitions? Base44 (June 2025) and Hour One AI (May 2025). See above.

Change in accounting policies? No policy change. Two significant estimate changes: the release of the entire $72.8m deferred-tax valuation allowance at 31 December 2025 (a $70.9m benefit), and the recurring fair-value remeasurement of the Base44 contingent consideration ($74.1m → $89.5m), which the auditor identified as an area of significant judgement involving a Monte Carlo simulation on projected revenue metrics.

Recent changes — new markets, facilities, management? New market: application generation via Base44, a genuine TAM expansion beyond websites. New product: Wix Harmony, rolled out across main geographies in late January 2026, now powered by Wix’s own first proprietary LLM. New facilities: the Tel Aviv campus buildout completed (capex fell from $63.0m in FY2023 to $9.9m in FY2025). Organization: a 20% workforce reduction (1,000 employees) communicated 28 May 2026, including “the scaling down and/or discontinuation of certain activities, initiatives, products, and subsidiaries” — a material narrowing of the company’s scope that has not yet been detailed product by product. Management continuity is otherwise intact: Avishai Abrahami (co-founder, CEO), Nir Zohar (co-founder, President), Lior Shemesh (CFO). Abrahami did not present on the Q1-2026 call, disclosed as being due to the birth of his son.

Litigation. A securities class action announced by Hagens Berman Sobol Shapiro LLP with a class period of 19 February 2025 to 12 May 2026 and a lead-plaintiff deadline of 22 September 2026, alleging Wix overstated the competitiveness and performance of Base44 and Harmony and understated their development and promotion costs.

Sell-side actions. Morgan Stanley: Overweight → Equal-weight, price target $112 → $60 (~21 July 2026). Raymond James: Strong Buy → Outperform, target $75 → $70 (30 July 2026), explicitly on valuation after a 32% one-month rally; the shares fell 9.4% that day. These are recorded as facts about market expectations, not adopted as a view here.

APPENDIX B — Source Appendix

Wix.com Ltd. (NASDAQ: WIX) · Published 2026-08-01 · All sources accessed 2026-08-01 unless noted

Wix.com Ltd. is an Israeli foreign private issuer. It files an annual report on Form 20-F and furnishes interim results on Form 6-K; there are no 10-Ks, 10-Qs or DEF 14A proxy statements. Substantive content in each 6-K is carried in the co-filed Exhibit 99.1, not the cover-page primary document. Section 16 insider reports (Forms 3/4/5) begin only on 18 March 2026, following the extension of Section 16 to foreign private issuers; Form 144 notices are the only pre-2026 insider record.

Primary — Company filings (SEC EDGAR, CIK 0001576789)

Primary — Management commentary

Quantitative data sources

Secondary — press and third-party

Analytical frameworks applied

Two analytical systems were applied throughout: Greenwald & Kahn (Competition Demystified) for the moat taxonomy (supply/cost, demand captivity, scale-plus-captivity), the market-share-stability test and the ROIC test — both applied explicitly in the Competitive Position section; and Marathon Asset Management (Capital Returns) for the supply-side capital-cycle reading of AI application generation, where an unusually rapid capital inflow is the operative fact.

All figures in this article are in US dollars, Wix’s reporting currency. Where a company-defined non-GAAP measure is used it is labelled as such and reconciled to the GAAP figure. This article contains no investment recommendation and no price target; the labelled Claude's Take block is the author’s own subjective view. Nothing here is investment advice.