Cloudflare: A Long-Term Owner's Perspective

Hunting ten-year five-baggers among great growth stocks — pressing the upside question: "Can it get much bigger?"

• QHow large is its market ceiling? Is it expanding an existing pie, or creating an entirely new market?6/10

  A

  The ceiling is high enough, but Cloudflare is mainly taking share in an already large and still-expanding existing pie, while also making that pie thicker. It is not creating a wholly new market from scratch. It competes across connectivity, security, acceleration, Zero Trust/SASE, application delivery, and developer platforms, all of which are established pools with incumbents, then sells them as an integrated offering on one global network. What Baillie Gifford's LTGG lens cares about is the market depth needed for “5x in 10 years.” Cloudflare broadly clears that bar: the report defines it as “more than 60 cloud services on a unified platform across more than 100 countries,” with a naturally global, long-duration demand profile. But the element of “creating a new category” is limited. This is mostly penetration through software-izing and subscription-izing existing capabilities.

  External cross-checks also show the market is still growing. Cybersecurity and edge demand were expanding over the same period: Zscaler's FY2026 Q2 revenue rose +26% year over year to $815.8 million; Palo Alto Networks' FY2026 Q2 revenue rose +15% year over year to about $2.6 billion, while next-generation security ARR rose +33% to $6.3 billion; even the smaller Fastly reported Q1 revenue of $173 million, up +20%, with security revenue up +47%. The demand pool is getting larger, but that is exactly the point: a high ceiling does not mean a blue ocean. Cloudflare is fighting for profit in a large market crowded with heavyweight rivals, not enjoying an empty territory alone.

  The incremental opportunity that could push the ceiling from “large” to “very large” is AI application infrastructure (edge inference, Workers/AI, developer platform) and enterprise network replacement. The report argues that Cloudflare is penetrating “higher-value, more complex, harder-to-replace enterprise customers,” with evidence in large customers above $100,000 in annualized revenue rising to 4,416 in 2026Q1, up 25% year over year (about 3,527 a year earlier, according to the report). If this materializes, it is a real new growth surface that “expands the existing pie.” Still, it is an extension of existing demand structures, not the definition of a market that did not previously exist. The honest conclusion is therefore: the market ceiling is not a hard constraint on a “5x in 10 years” outcome. The constraint is whether Cloudflare can keep capturing a disproportionate share and profit pool inside a crowded large market, not whether the ceiling itself is high enough.

  Jun 11, 2026
• QCan its revenue at least double over the next five years? Will growth be driven mainly by volume, price, or new businesses?7/10

  A

  Very probably, and doubling is only the passing mark. Given Cloudflare's current scale and growth rate, a revenue doubling within five years looks relatively conservative. Growth is a three-part stack led by volume plus new businesses, with price playing a supporting role. The anchors are clear: full-year 2025 revenue was $2.1679 billion (up +30% year over year), and 2026 Q1 revenue was $639.8 million, up +34% year over year. Simple math: if growth slows from 30% to 15% over five years (about a 22% geometric average), revenue can still rise from roughly $2.2 billion to roughly $5.8 billion, well beyond a doubling. To miss a doubling, growth would have to quickly fall below 15% and stay there, which plainly conflicts with the current +34% reading. Even the report's DCF “conservative case” uses a 10-year CAGR of 13%-15%, which means a doubling (requiring about 14.9% annualized over 5 years) is close to the downside case.

  Structurally, three forces are doing the work, with different weights:

  • Volume (customer count plus usage expansion), the main engine. Large customers (annualized >$100,000) rose to 4,416 in 2026Q1, up 25% year over year (about 3,527 a year earlier, according to the report). Enterprise penetration is the core source of revenue growth. The report also stresses that this is not superficial growth piled up through huge numbers of low-price customers.
  • Price/expansion (existing customers buying more), the secondary engine. This shows up in dollar-based net retention: 118% in 2026Q1, up 7 percentage points year over year (111% for full-year 2025, according to the report). Existing customers alone naturally contribute growth in the low-teens percentage range each year. Here, “price” mostly means the same customer buying more modules, not simply a higher unit price.
  • New businesses (AI/Workers, Zero Trust/SASE, developer platform), the incremental accelerator. This is the key variable that pulls growth back up from natural decay, and it is the upside source both the report and the Baillie Gifford framework are underwriting.

  The honest risk note: whether the revenue doubling comes with real earnings is a separate question. The report's central concern is not whether revenue can double, but whether infrastructure costs eat the margin during that doubling. In 2026Q1, GAAP gross margin already fell from 75.9% a year earlier to 71.2%. So the answer here is: “a revenue doubling is almost not the suspense; the suspense is the profit quality of that doubling.” That belongs in the unit-economics and valuation questions.

  Jun 11, 2026
• QFive years from now, what will take over as the next growth engine? Does this “second curve” exist today?5/10

  A

  The second curve already exists, and it is an extension growing out of the same network, not a separate bet built from scratch. That is the biggest advantage of Cloudflare's platform model. Its handoff engines are mainly three, all already visible in the revenue base, though still small as a share:

  • AI inference and application infrastructure (Workers AI / AI Gateway / edge inference). This is the second curve on which the market places the highest hopes. The logic is to turn “the global network closest to users” into the carrying layer for AI inference and agent traffic. The report includes “AI application-related infrastructure” in the company's capability map and explicitly says the company is continuing to expand into “AI, developer platforms, and enterprise networks.” A contemporaneous signal is that in 2026 Q1 the company announced a shift to an “agentic AI-first operating model,” laying off about 20% of employees (about 1,100 people) and recording $140 million-$150 million of restructuring charges. That is a forceful organizational bet on the AI second curve. Note: this occurred after the report's May 17, 2026 cutoff and was not covered in the report.
  • Developer platform (Workers/storage/databases and related services). The report repeatedly stresses that “Workers, storage, and AI-related services stacked onto the same platform” raise switching costs. The value of this curve is not only revenue; it turns customers from “buying a CDN” into “building applications on Cloudflare,” combining moat and second curve.
  • Zero Trust / SASE / enterprise network replacement. Replacing enterprise-built SD-WAN, VPN, and firewall stacks with Cloudflare's global network maps to large-customer penetration. Large customers rose to 4,416 in 2026Q1, up 25% year over year (about 3,527 a year earlier, according to the report), which is exactly the trace of this curve starting to convert.

  From the Baillie Gifford lens, this is a positive: the truly dangerous company is one where the first curve is topping out and the second curve is still only on slides. Cloudflare is not that. Its second curve shares the same network, sales motion, and customer relationships as the first curve, so marginal acquisition costs are lower and synergies are strong. But two limitations need to be marked honestly. First, the company does not separately disclose the actual revenue pull from the AI curve; it remains more “imagination plus early stage” than realized cash flow. Second, AI inference is a high-infrastructure-cost business. It is both the second curve and one source of the cost pressure that drove gross margin from 75.9% down to 71.2%. The stronger the second curve, the greater the possible short-term margin pressure. So the second curve “exists and is not weak,” but it is a high-imagination, high-investment, profit-still-to-be-proven type, not a mature engine already printing money.

  Jun 11, 2026
• QWhat is its core competitive advantage? Will this moat widen or narrow over the next three to five years?5/10

  A

  The core advantage is an architectural moat built from “one global network + unified platform breadth + gradually accumulating switching costs.” Over the next three to five years, I would call it stable to slightly wider, but nowhere near deepening into a fortified moat. The report gives its moat strength only 3/5, and I agree with that moderately neutral setting. This is not a brand-premium business (consumer mindshare is weak), not a physical monopoly (no license/concession barrier), and not a strong two-sided network-effect business (weaker than social, exchanges, or payment networks). The report identifies the essence accurately: it is more like a “hard-to-replicate integrated engineering system” and a “data flywheel + product flywheel,” not a legally protected exclusive right.

  Reasons it can widen, the side Baillie Gifford would care about:

  • Platform breadth is still expanding, and enterprise penetration is deepening. The more a single customer stacks DNS, WAF, DDoS, Zero Trust, Workers, storage, and AI services, the harder migration becomes and the more procurement tilts toward “the whole platform.” Large customers (>$100,000) rose to 4,416 in 2026Q1 (up +25% year over year), and dollar-based net retention rose to 118% (about 3,527 and 111% a year earlier, respectively, according to the report). That is direct evidence of switching costs thickening.
  • Replication costs are high and cycles are long. A rival needs not only capital to build the network, but also long-term investment in product integration, the control plane, distribution channels, and customer mindshare.

  Reasons it can narrow or be offset, which must be stated honestly and explain why I do not score it higher:

  • Large rivals are accelerating platform integration, and the profit pool is being shared. AWS/Azure/Google Cloud, Akamai, Palo Alto (next-generation security ARR up +33% to $6.3 billion), and Zscaler (FY26Q2 revenue up +26%) are all doing the same thing: converging point capabilities into platforms. The report is direct: competitors “include both on-premises hardware vendors and various point-solution providers and traditional public clouds,” so the moat “cannot be overstated.”
  • Pricing power is not solidly proven. One mark of a truly strong moat is margin expansion alongside scale. Yet 2026Q1 gross margin fell from 75.9% to 71.2%, which the company attributed to third-party technology services, bandwidth, and server depreciation. The report therefore marks both “pricing power” and “durable moat” as “uncertain” in its checklist.

  Overall judgment: this is a real, hard-to-replicate, but non-exclusive moat. Directionally it is “slightly wider,” not “clearly deeper.” Its strength lies in engineering integration plus switching costs. Its weakness lies in the absence of network-effect lock-in, unproven pricing power, and many peer-level rivals. Over three to five years, it is more likely to slowly thicken as breadth expands than to become a gulf rivals cannot cross. That still falls short of the Baillie Gifford preference for moats that keep deepening and create winner-take-most outcomes.

  Jun 11, 2026
• QIf the core business is disrupted, does it have the genetic capacity to reinvent itself? How does it handle mistakes and bad news?5/10

  A

  Reinvention DNA: fairly strong. Cloudflare's product history is itself a history of continually adding new categories on the same network and actively redefining its own boundaries. That is exactly the “can reinvent itself if the core is disrupted” quality Baillie Gifford values. But its shareholder-friendliness in handling mistakes and bad news is only medium.

  Start with the reinvention DNA. The report repeatedly emphasizes that Cloudflare is “not assembling multiple networks for different products, but trying to keep adding new functions on one global network.” From the original CDN to WAF/DDoS, DNS, Zero Trust/SASE, the Workers developer platform, and then AI inference infrastructure, the company's expansion pattern is itself a form of continuous reinvention. The latest and most aggressive example occurred after the report's cutoff date (2026-05-17): in 2026 Q1 the company announced a shift to an “agentic AI-first operating model,” laying off about 20% of employees (about 1,100 people), recording $140 million-$150 million of restructuring charges, and expecting the process to be largely complete by the end of Q3. Whether that move proves right or wrong, it shows management is willing to cut into its own organization when AI may disrupt “labor-intensive software operations.” That is a strong signal of reinvention DNA, not passive waiting. Founder control, discussed below, also gives enough strategic continuity to support multi-year directional bets, which is positive for whether the company can rebuild itself if the core is disrupted.

  Now, how it handles mistakes and bad news requires an honest discount. The positive signal is reasonably transparent disclosure. The report itself cites negative items the company disclosed, such as 2026Q1 gross margin falling from 75.9% to 71.2% with a direct attribution to cost structure, and the annual report's acknowledgment of content/copyright/platform liability risks (the report notes that “in October 2025 a Japanese court held the company liable for damages in an infringing-content case involving a website using its services, and the company has appealed”). But the negative signals are also clear. First, the dual-class share structure (founders together holding about 50% voting power) means outside shareholders have almost no check on whether management admits mistakes or corrects course. Correction depends on the founders' own discipline. Second, the incremental cost of founders' personal use of company-leased aircraft is not small (about $1.349 million for Prince and about $1.250 million for Zatlyn in 2025). The report judges these signals “not the most restrained shareholder-culture signals,” and I agree.

  Overall: reinvention capability, at the product and organizational levels, is this company's real long board. But “handling bad news” has to be split in two: information disclosure is acceptable, while governance correction lacks external checks because of the control structure. In essence, investors are betting that the founders will correct themselves. For long-term holders, that means you are buying the founders' judgment and honesty, not a system that can force management to admit mistakes.

  Jun 11, 2026
• QDoes management, especially the founders, have a long-term orientation and interests deeply tied to the company? Are they willing to sacrifice current profit for five to ten years out?6/10

  A

  Long-term orientation and willingness to sacrifice current profit for five to ten years out: clearly present. This is exactly the side of Cloudflare that most resembles a Baillie Gifford ideal. But “interests deeply tied to outside shareholders” deserves a discount, because the binding mechanism is super-voting control, not the economic interest of being in the same boat as you.

  The evidence for long-termism and sacrificing current profit is hard. From 2020 to 2025, the company grew revenue from $431.1 million to $2.1679 billion (five-year CAGR about 38%), at the cost of persistent GAAP operating losses: 2025 operating loss was about $207 million, with an operating margin of about -10%. In other words, management has for years chosen to reinvest in R&D, sales, and the global network rather than dress up short-term profit. That is exactly the “willingness to sacrifice the present for the long term” that Baillie Gifford likes. The latest example is the 2026 Q1 announcement of a shift to an agentic AI-first model, about 20% layoffs, and $140 million-$150 million of restructuring charges, clearly a bet on cost structure and AI positioning years out, even at the cost of a current one-time hit. This occurred after the report's 2026-05-17 cutoff.

  Interest alignment has to be split between “control alignment” and “economic alignment.” This is the key reason both I and the report score management only 3/5:

  • Control alignment is extremely strong, and double-edged. Under the dual-class structure, Class B shares carry 10 votes per share. As of year-end 2025, the two co-founders together held about 50.5% of total voting power, and all directors, executives, and holders above 5% together held about 71.0% (the report records Prince at about 41.7%, Zatlyn at about 10.5%, and all executives and directors together at about 53.1%, a similar basis). This is a plus for strategic continuity, but a clear constraint on governance checks for outside shareholders. On key issues, your vote carries almost no weight.
  • Economic alignment is not a pure positive for ordinary shareholders. Founder wealth is mainly tied to super-voting shares and the long-term share price, so the direction is aligned with you. But the “cost of alignment” is borne by all shareholders: 2025 stock-based compensation and related payroll taxes totaled about $489.9 million (the report's pure SBC figure is about $451.5 million), and weighted-average shares rose over five years from 299.8 million to about 348.4 million shares (about +16%, or about 3% annual dilution). Add the incremental cost of founders' personal use of company-leased aircraft (about $1.349 million for Prince and about $1.250 million for Zatlyn in 2025), and I agree with the report's view that these signals “do not belong to the most restrained shareholder-culture signals.”

  Overall: management's long-term view, strategic persistence, and willingness to sacrifice current profit for the long term are real strengths, meeting Baillie Gifford's preference for founder-led long-duration companies willing to tolerate ugly short-term optics. But “deeply tied to the company” is more accurately “deeply tied to control of the company,” not “deeply tied to outside common shareholders' intrinsic value per share.” Persistent high SBC and super-voting rights make the alignment directionally consistent for minority shareholders, but asymmetrical in cost.

  Jun 11, 2026
• QIf it disappeared tomorrow, how much would customers miss it? Is its growth sustainable without relying on social harm or regulatory arbitrage?6/10

  A

  Indispensability: quite high. If Cloudflare disappeared tomorrow, many websites, applications, and enterprise networks would immediately go down, slow down, or be exposed to attacks, and customers would miss it badly. Its growth model is also broadly “legitimate infrastructure subscriptions,” not growth built on social harm or regulatory red lines. Together, these form a positive answer set in the Baillie Gifford framework.

  Start with “how much would customers miss it.” Cloudflare sells critical-path capabilities such as connectivity, protection, acceleration, and Zero Trust access. It sits on the necessary path between users and applications. The report notes that its revenue “almost entirely comes from subscriptions and support,” with “continuous billing, continuous renewal, and continuous expansion,” and that demand is highly diversified: in 2023, 2024, and 2025, no single customer accounted for more than 10% of revenue. That means it is “individually indispensable” to thousands of customers, not tied to one or two giants. The best quantification of “missed” is stickiness: 2026Q1 dollar-based net retention rose to 118%, and large customers reached 4,416 (up +25% year over year) (about 111% and 3,527 a year earlier, respectively, according to the report). Customers are not only staying; they are buying more on the same platform. Migrating away means replacing DNS, WAF, DDoS protection, Zero Trust access, and more at the same time. The pain is high. So the “missed” factor is real and strong.

  The honest discount is that its indispensability is mostly at the enterprise IT and developer level, not emotional dependence at the consumer level. It also operates in a market with real substitutes (Akamai, Fastly, and the major public clouds can all absorb parts of the functionality): “difficult and painful, but not irreplaceable.”

  Now to whether growth is sustainable and whether it depends on harming society or regulators, the implicit double premise in this question. I would answer in two layers:

  • Social/legitimacy layer: basically healthy. The business model software-izes network infrastructure and charges subscriptions. It does not rely on regulatory arbitrage, addictive design, or monetizing privacy invasion. Capabilities such as DDoS protection and secure access carry positive externalities. The sources of growth, enterprise penetration, usage expansion, and new modules, are legitimate B2B subscription expansion.
  • Regulatory layer: real but manageable tail risk that needs monitoring. The report explicitly notes that content, copyright, platform liability, data privacy, and cybersecurity incident disclosure laws could lead to litigation and higher costs, and specifically discloses that in October 2025 a Japanese court held the company liable for damages in an infringing-content case involving a website using its services, and the company has appealed. The dispute over whether platforms should bear responsibility for hosted or transmitted content is a structural issue for neutral infrastructure providers like Cloudflare. It usually will not destroy the company, but it can raise compliance costs and operational complexity.

  Overall: indispensability is high, the growth model is legitimate, and growth is not built on social harm. This passes Baillie Gifford's “social sustainability” test. The only item requiring long-term monitoring is the tail risk around platform content liability, which is a cost and complexity issue rather than a fundamental problem of growth being built on harm.

  Jun 11, 2026
• QHow are the unit economics of this business (gross margin, incremental returns)? Do they improve or worsen with scale? Where does the money earned get spent?5/10

  A

  Unit economics: the absolute gross margin level is good (70%+), but the SaaS textbook rule that “larger scale leads to better unit economics” is now being questioned in Cloudflare's case. This is the report's sharpest and most important caution, and I fully agree with its prudence. This is the key distinction between Cloudflare and a “pure software, scale equals profit” asset, and it is also one area where Cloudflare is not especially strong in the Baillie Gifford framework.

  Gross-margin level and trend: long-term gross margin has stayed around 75%-78% (75% for full-year 2025, with gross profit of $1.6154 billion), a good base for a subscription model. But the inflection signal is clear: in 2026 Q1, GAAP gross margin fell from 75.9% a year earlier to 71.2%, which the company directly attributed to third-party technology service costs, network and bandwidth costs, and higher depreciation from server investment. The report's summary is apt: its cost structure is “not light.” It is not only R&D and sales headcount, but also bandwidth, racks, network facilities, and server depreciation, all hard costs. Scale expansion may not automatically bring linearly higher margins, especially as the company expands into AI, developer platforms, and enterprise networks, where hard costs can directly compress profit. This means its incremental returns sit between “pure software” and “infrastructure/compute”: lighter than asset-heavy CDN businesses such as Akamai, but heavier than pure SaaS with no marginal cost.

  Incremental returns, meaning whether incremental investment can produce excess return: operating leverage has improved on the face of the accounts. Operating margin narrowed from -25% in 2020 to about -10% in 2025, and operating cash flow reached $603.1 million in 2025. But the report cuts through the “cash-profit quality”: 2025 operating cash flow added back about $490 million of stock-based compensation and related payroll taxes (pure SBC about $451.5 million), and that year also benefited from substantial interest income that flattered net profit (according to the report, 2026Q1 operating loss was about $61.99 million, but interest income of about $40.17 million reduced pretax loss to about $21.40 million). After treating SBC as a real shareholder cost, the report estimates “conservative owner earnings close to 0.” That means current incremental returns are “improving in accounting terms, but not yet realized as hard cash for ordinary shareholders.”

  Where the money goes: three destinations. First, R&D and sales, which drive volume growth and enterprise penetration. Second, continuing reinvestment in the global network, servers, and other infrastructure, a necessary expense to maintain competitiveness; the report estimates maintenance capex at about $130 million-$170 million per year. Third, about $490 million of stock-based compensation in 2025. This nominally “does not spend cash,” but is paid by all shareholders through dilution per share (share count up about +16% over five years). It is worth noting that the 2026 announcement of an agentic AI-first transition and about 20% layoffs (after the report cutoff) is exactly an active attempt to improve unit economics and lower the human-cost structure. The direction is right, but the effect remains to be proven.

  Overall: the gross-margin base is good and incremental returns are moving upward, but the Baillie Gifford-favored flywheel of “larger scale, better unit economics” has not yet been proven at Cloudflare, and even reversed in 2026Q1. Add SBC's erosion of per-share value, and its unit economics are “adequate, slightly neutral,” not a top-tier “gets sweeter with scale” profile.

  Jun 11, 2026
• QWhat conditions must all hold for it to rise 5x in 10 years? Are those conditions realistic? What expectations are embedded in today's share price?3/10

  A

  For Cloudflare to rise 5x in 10 years, let alone the higher multiples in Baillie Gifford's ideal, several conditions have to hold at the same time. None is absurd on its own, but getting all of them right is not easy. Today's share price has already prepaid a substantial portion of those good outcomes, leaving a negative margin of safety. This question is also the core landing point of the report's “Watch, not cheap” conclusion.

  First, separate the conditions needed for “5x in 10 years” (where the Baillie Gifford blue-sky firepower should be concentrated in years 3-10):

  1. Revenue must continue compounding at a high rate. If the 10-year 5x outcome is mainly driven by fundamentals, revenue likely needs to compound in the high-teens to around 20% range for many years. The anchors are full-year 2025 growth of +30% and 2026Q1 growth of +34%. The starting point is good, but preventing growth from falling to the low- to mid-teens over 10 years is a high bar.
  2. Gross margin and operating margin must materially turn positive and climb. This is the key shift from “revenue story” to “cash story.” But 2026Q1 gross margin fell from 75.9% to 71.2%, and full-year 2025 GAAP operating margin was still about -10%. This condition therefore requires both a directional reversal and a long-term upward path.
  3. SBC as a share of revenue must decline meaningfully, and dilution must converge. 2025 SBC and related payroll taxes were about $490 million, and the share count rose about +16% over five years. If dilution does not converge, even cash generated by the company will have a chunk eaten away before it reaches intrinsic value per share.
  4. The AI/developer-platform second curve must truly scale and monetize, not merely contribute imagination. This is the incremental source that can pull growth back up from natural decay.
  5. The moat must not be diluted in crowded competition (rivals such as Palo Alto, with next-generation security ARR up +33% to $6.3 billion, and Zscaler, with FY26Q2 revenue up +26%, are competing for the same pool), and the valuation multiple must not compress sharply.

  Are these conditions realistic? Each one is “possible,” but having 1-5 all hold for a long period is a near-perfect execution scenario. The report captures it in one sentence: “Cloudflare may be a very good company, but the current share price requires it to become an almost great company.” Conditions 2 and 3 in particular (margin turning positive plus dilution convergence) are currently moving against the thesis, and 2026Q1 data moved in the wrong direction.

  What expectations are embedded in today's share price? It embeds near-certain delivery of “high growth plus high margin expansion,” with a negative margin of safety, and the setup is even more extreme than at the report cutoff. The report anchored on the May 15, 2026 share price of $197.56 and market cap of about $69.66 billion (per the report), and already judged it “clearly overvalued (above $190).” Its DCF estimated intrinsic value at $80-$110 in the conservative case, $120-$160 in the reasonable case, and $180-$230 in the optimistic case, meaning the then-current price could only barely be justified under the optimistic case. By early June 2026 the share price had risen to roughly $234-$248, touched an intraday all-time high of about $276.82 on June 4, and market cap had climbed to about $83 billion-$87 billion. It had already stood above the report's optimistic intrinsic-value upper bound ($230), and even broke above it intraday. In other words, the market is not only treating the optimistic case as the default; it is adding a further premium on top. Even using the broadest lens, applying 2024 audited free cash flow of about $167 million implies P/FCF far above 400x. That is not a “reasonable doubt” price; it is a “high-confidence delivery price plus extra premium.”

  Conclusion: the conditions needed for a 10-year 5x are a high-difficulty bundle where “everything has to go right,” and two of the conditions must first reverse before improving. They are not fantasy, but today's share price already prices them as facts. The current price (about $234-$248, having touched $277) has moved above the report's optimistic intrinsic-value upper bound. For Baillie Gifford-style growth investors willing to wait 10 years, the asset deserves long-term tracking, but at current levels the upside has been heavily prepaid and the downside exposure is real. This is not an entry point.

  Jun 11, 2026
• QWhy has the market not realized all this yet? Is it because the market does not understand, dismisses it, or cannot look far enough? What will become the “narrative inflection point”?3/10

  A

  For Cloudflare, this Baillie Gifford question has to be used in reverse. The issue is not “why has the market not realized how good it is,” but “has the market already realized too much and prepaid the good news into the price.” For this stock, the direction of the perception gap is more likely negative: the market understands it clearly, and may even be too optimistic. That is the honest answer, without forcing a “misunderstood growth story” angle just to fit the template.

  Compare it item by item against Baillie Gifford's “does not understand / dismisses / cannot look far enough” frame:

  • Does not understand? Basically no. Cloudflare is one of the best-covered high-growth software stocks on Wall Street, with dense institutional ownership and heavy sell-side research. Its platform logic, network scale, growth rate, and customer metrics are repeatedly analyzed. 2026Q1 revenue growth of +34%, dollar-based net retention of 118%, and 4,416 large customers are familiar parts of the market narrative. There is no “nobody understands it” information gap.
  • Dismisses it? The opposite. The market is giving it a premium, not a discount. When the report anchored on the May 15, 2026 price of $197.56 and market cap of about $69.66 billion (per the report), it already judged valuation “extremely high,” with a simple price-to-sales ratio around 30x and P/FCF above 400x. By early June 2026, the share price had risen to roughly $234-$248 and touched an intraday all-time high of about $276.82 on June 4. The market is not looking down on Cloudflare; it is treating it as a top-tier platform asset.
  • Cannot look far enough? This is the only possible perception gap, but the direction is uncertain. The bulls' “long view” underwrites the 10-year scale of the AI second curve (Workers AI, agentic traffic carriage). The bears' “long view” focuses on what the report identifies: conservative owner earnings close to 0 after SBC, 2026Q1 gross margin falling rather than rising, down to 71.2%, and profit maturity lagging far behind the valuation. In other words, the long term is the point of disagreement, not a blind spot. It is not “the market has not seen it”; it is “the market has chosen to believe the optimistic version of the long term.”

  What will become the narrative inflection point, the implied landing point of this question? The inflection will not be “the market suddenly discovers it is good,” but “the market is forced to reprice the pace of profit realization,” in either direction:

  • Downward inflection, the one to watch more carefully: net retention falling back below 110% and staying there, large-customer growth clearly decelerating, gross margin remaining in the low 70%s for an extended period (already down from 75.9% a year earlier) or lower without recovery, SBC as a share of revenue staying high, or treatment of the 2026 convertible debt causing more dilution than expected. If any one of these is confirmed, it would shake the core bull premise that “a larger platform produces better unit economics” and trigger a valuation reset (the report warns that from current levels, 50%-70% downside is “not unimaginable”). The 2026 announced agentic AI-first transition plus about 20% layoffs (after the report cutoff) is itself a double-edged sword: if execution is smooth, it can be narrated as a “margin inflection”; if execution stumbles, it confirms that growth is getting more expensive.
  • Upward inflection, realization-driven rather than discovery-driven: several consecutive quarters in which growth stays around 20% while GAAP operating losses and SBC ratio both converge and gross margin returns to the mid- to high-70%s. That would falsify the “high valuation, low cash return” concern and turn the story into “high growth with real cash.” The report is explicit that if this set of things happens together, valuation would be revised meaningfully higher.

  Overall: for Cloudflare, Baillie Gifford's “why has the market not realized it yet” question offers almost no buried-discount dividend. This is not a dusty cheap quality asset. It is a fully recognized, optimistically priced quality asset. The real variable is the race between profit realization and valuation expectations. Narrative inflection will come from quarterly confirmation or disconfirmation of margins and dilution, not from a sudden market epiphany.

  Jun 11, 2026