Chapter 3

The Pivot and the Promises

Squeezing a Flat Base described the machine as it runs today: one lever — ARPU on a flat customer base — feeding a capex-light, upfront-billed model that converts revenue to cash at a 31.8% margin. That machine did not assemble itself. It is the product of a single decision taken in 2019 and a run of promises management made, and mostly kept, while proving the decision worked. This act reads the record: what management said it would do, what it did, and where the two diverge.

The Free-Cash-Flow Reset

The formative decision was to stop optimizing for customers and start optimizing for cash per share. Under Aman Bhutani, chief executive since September 2019, management reframed the whole business around what it calls its "north star of maximizing free cash flow over the long-term" [1]. Everything downstream — the segment recut, the margin expansion, the buyback — is machinery in service of that one objective.

The reset shows up in the margin structure. Operating margin roughly doubled between FY2021 and FY2025, and the free-cash-flow margin climbed from just over a fifth of revenue to nearly a third — a widening gap that is the signature of a company harvesting a base rather than buying growth.

Source: operating margin and free-cash-flow margin, FY2021–FY2025, as reported; operating income per the FY2025 Consolidated Statements of Operations [2] and operating cash flow per the Consolidated Statements of Cash Flows [3].

The plumbing that turns that margin into per-share value is the 2022–2023 recut into two segments — the high-margin Applications and Commerce line and the durable Core Platform cash base introduced earlier — paired with a capital-return framework built around a multi-year share-reduction target. The segments generate the cash; the framework decides where it goes. What makes the reset legible as a track record, rather than a slogan, is that management then put numbers on it and let investors keep score.

The Said-Versus-Did Ledger

At its March 2024 investor day, management set three-year targets that are unusually specific for a company its size: 6% to 8% annual revenue growth, a normalized-EBITDA margin near 33% by 2026, more than $4.5 billion of cumulative free cash flow, and a 20%-plus compound growth rate in free cash flow per share [4]. Those targets are the yardstick. Held against them, the record splits cleanly by the kind of promise.

Sources: three-year targets, 2024 Investor Day [5]; share reduction and customer-growth commitment, Q3 FY2024 call [6]; FY2026 revenue guide and deferred pricing lever, Q4 FY2025 call [7]; FY2025 customer count, FY2025 10-K [8].

The hard financial promises were kept or beaten, and they were kept through the drawdown, not before it. The share-reduction target is the clearest example: management said it would cut gross shares about 20% over three years and reported a 23% reduction by the third quarter of 2024, "three points ahead of our three-year targeted reduction of 20%" [9]. Free cash flow reached $1.58 billion in FY2025, a 19% gain, keeping the three-year cumulative target on a path to clear $4.5 billion with room to spare. Revenue grew 7.5% and then 8.3% — the top half of the promised band — before the FY2026 guide came in "at approximately 6% at the midpoint," the bottom of it [10].

The gaps are all on the softer, narrative side. In late 2024, with divestitures and product migrations behind it, management told investors "we expect to return to customer growth in 2025" — the count then stood at 20.7 million [11]. The FY2025 count instead came in at 20.4 million, below that starting point [12]. Rather than deliver the headline, management reframed toward "high-intent" customers and a higher-spend cohort. Whether narrowing the funnel deliberately is the right strategic call is the subject of The Front Door and AI; as a stated promise, it was abandoned, not met. Pricing and bundling — a headline growth lever through 2024 — was pushed out behind the new product: "we have pricing and bundling initiatives that are going to have to wait because this new AI-based website builder is just ready to go" [13]. The commerce and payment-volume theme that filled the 2024 script faded from later calls in the same way.

The pattern is worth stating plainly: when GoDaddy attached a number to a promise, it delivered; when the promise was a narrative — more customers, a pricing cadence, a commerce story — it was reframed as conditions changed. That distinction matters for how much weight to put on management's word going forward, and it points at the two commitments still open: the buyback and the AI bet.

The Buyback and Its Timing

GoDaddy is a single-instrument capital allocator. It pays no dividend, spends almost nothing on capex, and has done no material acquisitions since the Host Europe era; nearly all of its cash goes to one place. Across FY2021–FY2025 the company generated about $5.5 billion of free cash flow and spent about $5.4 billion buying back stock — roughly 97 cents of every free-cash dollar — which the May 2026 investor deck frames as returning more than 95% of free cash flow to repurchases, some $5.1 billion cumulatively, for a 31% cut in gross fully diluted shares since January 2022 [14].

Source: buyback and free-cash-flow totals derived from reported cash-flow statements, FY2021–FY2025; gross-reduction figure per the Q3 FY2024 call [15] and 2026 overview deck [16].

The gross-reduction headline overstates the net economic return, which resolves a thread Squeezing a Flat Base left open. Equity-based compensation runs near $300 million a year — $317.8 million in FY2025, larger than all depreciation and amortization combined [17] — and that stock is re-issued to employees each year. So despite $1.3 billion to $1.6 billion of gross repurchases annually, the diluted share count fell only about 3% to 4% a year, from 151.5 million in FY2023 to 140.6 million in FY2025. The buyback is roughly one-fifth a dilution offset and four-fifths genuine shrink; the net accretion is real, but more modest than the gross number suggests.

The softer part of the record is not whether management returned the cash — it plainly did — but the price it paid. The average price of open-market repurchases rose every year, from about $73 a share in 2023 to roughly $125 in 2024 and $142 in 2025 [18], with accelerated repurchase agreements settled at higher marks still — $139.65 in 2024 and $176.02 in 2025 [19].

Source: average price derived from reported open-market repurchase dollars and share counts — FY2023–FY2025 per the FY2025 10-K repurchase table [20]; FY2022 per the FY2024 10-K [21].

The largest dollar year was the last and the most expensive. At about $96 a share on July 27, 2026 — down 55% from its early-2025 high — the entire 2025 program sits underwater, and the 2025 accelerated repurchase at $176 deeply so. Then the pattern inverted: in the first quarter of 2026, with the stock roughly halved and cash at $1.3 billion, GoDaddy bought back just 3 million shares for $280 million against $474 million of free cash flow [22] — spending less as the stock got cheaper, not more. Asked about it directly, management deflected to the long view: "don't look at any particular quarter; look at our history — it's a good indicator of how we approach this" [23]. The commitment to return capital has held for five years; the price discipline is the part of the record still under scrutiny.

Reading Earnings Through the Tax Line

There is one place where the reported numbers can mislead a reader who stops at the bottom line, and it is worth clearing before the valuation chapter. Diluted earnings per share fell from $9.08 in FY2023 to $6.45 in FY2024 to $6.22 in FY2025 [24]. Read alone, that looks like a business going backwards. It is the opposite.

Source: FY2025 Consolidated Statements of Operations [25] and Consolidated Statements of Cash Flows [26].

Operating income rose every year, from $547.4 million to $1,127.3 million, and operating cash flow climbed from $1,047.6 million to $1,599.4 million over the same span. Net income fell only because the tax line reversed. FY2023 net income of $1,375.6 million carried a $971.8 million income-tax benefit, driven by releasing most of the U.S. deferred-tax-asset valuation allowance; FY2024 carried a $171.5 million benefit, itself driven by a $267.4 million gain from the DNC (Desert Newco) restructure; FY2025 normalized to a $145.0 million tax expense [27]. GAAP earnings fell because two one-off tax tailwinds ran out, not because operations weakened. Across these three years the comparable series are operating income and cash, not net income or EPS — which is why the two FY2025 quarters that screened as earnings "misses" have to be read against the tax base effect, and why the market's growth-line concern, examined in Priced for the Fade, is a separate matter from earnings quality.

One caveat belongs on the same page. The largest of those estimates — the FY2023 valuation-allowance release — was booked in the very year GoDaddy disclosed a material weakness in the design of its controls over income taxes; management concluded its internal control over financial reporting was not effective as of December 31, 2023. The weakness has since been remediated, and controls were attested effective at year-end 2024 and 2025 [28]. Ernst and Young has audited the company since 2004 and issued a clean opinion, its sole critical audit matter the mechanical complexity of high-volume subscription billing rather than any judgment or estimate [29]. The control gap sat precisely on the account that produced the biggest one-off — a specific reason to lean on cash generation over reported earnings, which this model already invites a reader to do.

Oversight and Incentives

The governance backdrop is conventional and carries no red flag; it sits here as the frame around the delivery record rather than a finding of its own. The board runs under a separate independent chair, Brian Sharples, with the CEO the only member drawn from management [30]. Ownership is fully dispersed: there is no founder, sponsor, or control person, index managers Vanguard (14.3%) and BlackRock (10.7%) top the register, and the entire board-and-management group holds under 1% of the economics [31].

The one detail that reinforces the read-earnings-through-management theme is in the pay design. For FY2025 the annual bonus formula was narrowed from four corporate metrics to two — Normalized EBITDA and Bookings, weighted equally — dropping revenue and unlevered free cash flow [32]. Both surviving metrics are company-defined, non-GAAP measures that management itself sets and reconciles; the change removed the two GAAP- and cash-anchored checks from the bonus. Pay is large and equity-heavy — the CEO earned $23.0 million for 2025, 95% of it stock and none of it options [33] — and the say-on-pay vote passed with 92.4% support, solid but with a non-trivial 7.6% against [34]. None of this is a problem. It is worth knowing that the yardsticks management is paid against are ones it defines — the same reason the tax line above rewards a reader who checks the cash.

Where the Record Stands

Across an eighteen-month, 55% decline, GoDaddy met or beat every quantified financial target it set. Free cash flow, margin, and the roughly 20% share-reduction promise were all delivered or exceeded, and they were delivered while the stock fell, not before. On the evidence, this is a de-rating against a clean delivery record, not a company that missed its numbers.

The strongest fact against leaning on that record is that the promises which slipped are the growth-side ones — the return to customer growth, the pricing cadence, the commerce story — and the two live commitments now pull in opposite directions. The buyback, the more reliable of the two for five years, just slowed as the stock cheapened and cash built; the AI revenue bet that would justify retaining that cash is one management still declines to size. A reader would revise the "clean deliverer" read if the bookings line failed to recover through FY2026, or if cash conversion stopped running ahead of reported earnings — either would turn a multiple de-rating into the early signal of a delivery problem.

That leaves two questions the record cannot settle on its own. Whether the growth story the cash rests on survives the shift toward AI-driven discovery is taken up in The Front Door and AI. Whether the price the market now sets fairly reflects a clean deliverer that stopped growing is the subject of Priced for the Fade.