Built by Grok 4.6 High

NET | Earnings Review — 2026 Q2

Verdict: ACCELERATING on the top line; GM stabilizing (not yet expanding YoY); EPS still growing high-thirties. Cloudflare printed $696.1M revenue (+35.9% YoY — fastest in the 10-quarter window), non-GAAP diluted EPS $0.29 (+38% YoY), non-GAAP op margin 13.8%, and FCF $56.4M (8.1% margin). DBNR returned to 120%; $100K+ customers hit 4,698 (+26.6% YoY), 73% of revenue.

Clean double beat vs Street (FMP: ~$664.7M / $0.268) and vs company guide high ($665M / $0.27). Revenue cleared guide high by +$31.1M / +4.7% — the widest % Street overshoot in L12Q. L12Q beat rates: revenue 100%, EPS 91.7% (sole miss Q1'25 −2.3%). Pattern: Consistent beater with improving revenue-beat magnitude (L4 avg ~3.7% vs prior-4 ~1.9%).

Non-GAAP GM 73.1%: first +30 bps sequential lift in eight quarters after a 72.8% trough, still −320 bps YoY on AI infra. US revenue $354.4M (+40.7% YoY) is the swing region.

Beat-and-raise. FY26 revenue raised to $2.864–2.870B mid $2,867M (+$58M / +2.1% vs prior; +32% YoY); non-GAAP EPS to $1.25–1.26 (+$0.06). Q3 guide $736–737M (+31% YoY) / EPS $0.34 sits in-line with Street — classic Cloudflare sandbagging with more confident “stellar / up-and-to-the-right” tone.

Tone & Q&A: Management sold the Agentic Internet vision hard (7.4M+ developers, ~2M adds in Q2; Agents Week rails) but ~41% of Street asks on product revenue mix, NRR path, and PoF headwinds were deflected. Four material contradictions: restructuring $140–150M→$165M, “not cost-cutting” vs ~20% RIF, 75–77% GM target vs stabilize ~73%, and Q1→Q2 GM outlook flip.

Next catalysts: Q3 print (~late Oct) vs $736–737M / $0.34; GM hold ~73%; DBNR / $100K+ durability; FCF vs “unchanged” outlook despite higher cash restructure.

Headline: Revenue re-accelerated from a +26.5% trough (2025Q1) to +35.9% in 2026Q2; US +40.7% and DBNR 120% drive the mix; non-GAAP GM first sequential uptick after eight quarters of AI-infra pressure.

Revenue drivers — geography & enterprise

Consolidated profitability

Revenue absolute levels + YoY trajectory

Trajectory verdict: Accelerating on the top line (~+940 bps from trough); GM compressing YoY but sequential bottom may be in; op margin roughly flat YoY near 13.8%. Trajectory favors growth over near-term margin expansion.

Double BEAT — rev +4.7%, EPS +8.0% vs Street; also above company guide high by +4.7% on revenue. Pattern: Consistent beater (L12 rev 100%, EPS 91.7%). Revenue beat magnitude improving.

This quarter vs consensus

| Metric | Consensus | Actual | Variance | Call | |---|---|---|---|---| | Revenue | $664.7M (FMP) | $696.1M | +$31.4M / +4.7% | BEAT | | Non-GAAP diluted EPS | $0.268 (FMP) | $0.29 | +$0.022 / +8.0% | BEAT | | vs company guide high | $665M / $0.27 | same actuals | +$31.1M / +$0.02 | Above guide | | Non-GAAP GM | ~73% context | 73.1% | +30 bps QoQ; −320 bps YoY | Seq. BEAT | | Non-GAAP op margin | mid-teens path | 13.8% | −30 bps YoY | Solid | | Free cash flow | n/a | $56.4M | +$23.1M YoY | YoY BEAT |

Management framing: Re-acceleration / Agentic Internet demand, large-customer land-and-expand (DBNR 120%, $100K+ +27% YoY), current RPO +35% YoY, and explicit prudent guidance cushion — not one-time accounting. GAAP distorted by $150.7M restructuring (excluded from non-GAAP).

Beat/miss heatmap (last 8 quarters)

| Window | Rev beat rate | EPS beat rate | Avg rev beat % | |---|---|---|---| | L12 quarters | 12/12 = 100% | 11/12 = 91.7% | ~2.6% | | L4 quarters | 4/4 = 100% | 4/4 = 100% | ~3.7% (widening) | | Prior 4 | 4/4 | 3/4 | ~1.9% | | This quarter | BEAT | BEAT | +4.7% peak L12 |

Watch: Street may re-anchor higher after two quarters of 3%+ rev beats; future “easy beat” cushion could compress even if fundamentals stay strong.

Material FY26 raise on a blowout Q2 beat — rev mid +$58M / +2.1%, non-GAAP EPS mid +$0.06 / +5%. Q3 guide essentially in-line with Street. Tone shifted from Q1 restructuring solemnity to “stellar / up and to the right,” while still advertising “prudent” guidance — classic sandbagging.

New guidance snapshot

| Metric | Period | Low | High | Mid | YoY at mid | |---|---|---:|---:|---:|---| | Revenue | Q3 2026 | $736M | $737M | $736.5M | +31% vs $562M | | nGAAP op. income | Q3 2026 | $129M | $130M | $129.5M | ~17.6% margin | | nGAAP EPS | Q3 2026 | — | — | $0.34 | +26% vs $0.27 | | Revenue | FY 2026 | $2,864M | $2,870M | $2,867M | +32% vs $2,168M | | nGAAP op. income | FY 2026 | $443M | $445M | $444M | ~15.5% margin | | nGAAP EPS | FY 2026 | $1.25 | $1.26 | $1.255 | +35% vs $0.93 |

Other: Network CapEx still 14–15% of rev; restructuring charges raised to up to $165M / cash up to $130M; FCF outlook unchanged.

Waterfall — FY26 revenue guide mid ($M)

Guidance vs Street / prior

| Metric | Prior mid | New mid | FMP consensus | vs Prior | vs Street | |---|---:|---:|---:|---:|---:| | FY26 Revenue ($M) | 2,809 | 2,867 | ~2,810 | +$58 / +2.1% | +$57 / +2.0% | | FY26 nGAAP EPS ($) | 1.195 | 1.255 | ~1.197 | +$0.06 / +5.0% | +$0.058 / +4.8% | | Q3'26 Revenue ($M) | n/a | 736.5 | ~735.8 | n/a | +$0.7 / in-line | | Q3'26 nGAAP EPS ($) | n/a | 0.34 | ~0.340 | n/a | in-line |

Implied 2H residual

| Build | $M | YoY | |---|---:|---:| | H1 2026 actual | ~1,335.8 | +34.7% | | FY26 guide mid | 2,867 | +32.2% | | Implied H2 | 1,531 | +30.1% | | Q3 guide mid | 736.5 | +31.0% | | Implied Q4 residual | ~795 | ~+29% |

Q3 guide steps down ~5pp from Q2’s +36% print — multi-year practice of resetting sequential expectations after a beat, not a demand cliff. Historical guide accuracy: 6/6 quarterly rev beats of next-quarter guide; avg ~+3.4%; Q2 +4.8% widest.

Tone: Q1 → Q2

| Dimension | Q1 2026 | Q2 2026 | Δ | |---|---|---|---| | Opening | Restructuring / “hard day”; ~1,100+ roles | “Stellar second quarter”; record customer/developer growth | More bullish | | Demand | AI biggest tailwind; lean in | Agentic Internet rewrite; trajectory “up into the right” | Higher confidence | | Guidance | “Thoughtful and prudent” | “Continue to be prudent” but trajectory clear | Same prudence, weaker risk hedge | | Margins | GM may continue down; unit economics | GM 73.1% first seq. up; stabilize around current | Relief | | Cash/restructure | Charges $140–150M | Up to $165M; FCF unchanged | Cost true-up without FCF cut |

Headline: Revenue YoY reaccelerated from +26.5% (2025Q1 trough) to +35.9% in 2026Q2 (+233 bps QoQ in the growth rate), while non-GAAP EPS YoY cooled from +56% to +38% on a tougher prior-year base.

8-quarter YoY series

| Metric | 24Q3 | 24Q4 | 25Q1 | 25Q2 | 25Q3 | 25Q4 | 26Q1 | 26Q2 | |---|---:|---:|---:|---:|---:|---:|---:|---:| | Rev YoY % | +28.2% | +26.9% | +26.5% | +27.8% | +30.7% | +33.6% | +33.5% | +35.9% | | Rev accel (bps) | −183 | −126 | −35 | +122 | +292 | +293 | −7 | +233 | | EPS YoY % | +25% | +27% | 0% | +5% | +35% | +47% | +56% | +38% | | Revenue ($M) | 430 | 460 | 479 | 512 | 562 | 615 | 640 | 696 | | nGAAP EPS ($) | 0.20 | 0.19 | 0.16 | 0.21 | 0.27 | 0.28 | 0.25 | 0.29 |

Inflection map

1. 2024Q3–2025Q1 — orderly deceleration into trough. Rev YoY slid to +26.5%; EPS YoY collapsed to 0% on seasonal opex.
2. 2025Q2 — reacceleration starts (+122 bps). Growth turned to +27.8%; EPS still soft at +5%.
3. 2025Q3–Q4 — decisive multi-quarter reacceleration. +292 then +293 bps; $100K+ mix and DBNR 120%; Seifert: “accelerating for the third consecutive quarter to 34%.”
4. 2026Q1 — growth held (−7 bps); EPS YoY peaked +56%. Easy prior-year base; agentic AI restructuring announced.
5. 2026Q2 — new 8Q high +35.9% (+233 bps); EPS normalizes to +38%. Base-math cooling, not operational deterioration. First sequential GM improve in 8Q.

Plain English: Classic trough-to-reacceleration arc powered by enterprise mix and AI/Workers demand. Top line still accelerating into the just-reported quarter; treat large EPS YoY % swings as base effects.

Multi-year annual arc

| Metric | 2021 | 2022 | 2023 | 2024 | 2025 | |---|---:|---:|---:|---:|---:| | Revenue | $656M | $975M (+49%) | $1,297M (+33%) | $1,670M (+29%) | $2,168M (+30%) | | nGAAP op. margin | −1.1% | 3.7% | 9.4% | 13.8% | 14.0% | | nGAAP EPS | −$0.05 | $0.13 | $0.49 | $0.75 | $0.93 | | FCF | −$43M | −$40M | $120M | $167M | $261M |

1H’26 run-rate (~+35% YoY) implies further acceleration into FY26 guide +32%.

| # | Catalyst | Timing | Consensus / watch | Implication | |---|---|---|---|---| | 1 | Beat-and-raise / Q3–FY bar | Realized Aug 6; next test ~late Oct Q3 | Q3 $736–737M / $0.34; FY mid $2.867B | Highest near-term stock catalyst; raise > Q2 beat size | | 2 | Agentic Internet productization | Agents Week Aug 4–7; multi-year | Monetization Gateway / Wallets / AI control plane still qualitative $ | Multi-year re-rating; near-term via security/Workers/ZT | | 3 | Developer platform scale | Ongoing; record Q2 adds | 7.4M+ developers; ~2M adds in Q2 alone | Leading indicator of land-and-expand; GM held seq. despite step-up | | 4 | GM stabilization | Inflection claimed Q2; H2 “around current” | 73.1% first seq. +30 bps in 8Q | Credibility for Rule-of-40 / op leverage | | 5 | Enterprise / DBNR | Every print | 4,698 $100K+ (+27%); DBNR 120% | Core fundamental; sustains 30%+ without pure new logos | | 6 | Restructuring harvest | Cash peaked Q2; benefits H2’26–27 | Charges up to $165M; FCF guide held | Op leverage catalyst with execution risk | | 7 | VoidZero (Vite) / toolchain | Closed Jun 2026; integrate H2+ | Strategic developer lock-in, not 2026 rev | Complements Workers growth | | 8 | SASE share vs ZS/PANW/NETSK | Continuous | NET at 36% growth vs peers mid-20s | Relative growth supports premium | | 9 | Network CapEx discipline | FY26 14–15% of rev held | Invest-behind-demand vs hyperscaler build | FCF margin durability is the KPI |

Risk catalysts: valuation already prices AI leadership; SASE competition; developer-mix GM drag; CapEx elevated; GAAP path to profitability by 2028 vs large one-time restructuring.

Bottom line: (1) 36% growth re-acceleration with beat-and-raise to ~$2.87B / $1.25–1.26; (2) developer + agentic product stack underwriting the multiple; (3) GM seq. turn + DBNR 120% + $100K+ +27% as proof the enterprise machine compounds. Next checkpoint: Q3 vs $736–737M / $0.34.

9 analysts, ~17 discrete asks. Management was expansive on Agentic Internet vision and structurally light on product-level modeling inputs. ~59% well answered / ~41% deflected.

Scoreboard

| # | Analyst | Topic | Badge | |---|---|---|---| | 1a | Saket Kalia — Barclays | Non-human traffic monetization |

Key Q&A pairs

Hardest deflection — Boolani (Citi): Workers + Workers AI % of revenue? Prince: “I will start out not fully answering… We haven’t broken out by product area.” Critical for underwriting Act 3 contribution to +36% YoY.

Highest-value financial — Ho (WB): GAAP profitability by 2028? Pacing ahead; ex-restructure Q2 GAAP loss ~$18M — “almost striking distance.”

PoF headwinds — Singh (MS): Never answered the binary past/not-past; instead “higher standard deviation… continue to be prudent on guidance.”

NRR path — McWilliams: “Strength across the board” — no range, no sequential path despite DBNR just at 120%.

CapEx philosophy — Colville: Refuse commodity GPU rental; sell work done; path to ~10× utilization vs box rental. Aligns with CapEx 14–15% of FY26 rev.

Dog that didn’t bark: $150.7M Q2 restructuring / ~20% RIF barely probed in Q&A.

Street open items

1. Workers / developer platform revenue mix (still undisclosed)
2. NRR trajectory beyond the 120% print
3. Pool-of-funds headwind status
4. Cyber demand timing (Q2 vs 2H)
5. Act 4 monetization timing
6. Restructuring ROI / FCF bridge

Four material management-statement tensions — no fully irreconcilable numerical fraud signal.

Lower severity: CapEx range narrowed 12–15% → 14–15% (directionally consistent); FCF “unchanged” while cash restructure raised (unsupported joint claim without bridge); GAAP vs non-GAAP optics (presentation, not logical conflict).

No contradiction found on: rev vs prior guide (clean beat), FY guide path (only up), DBNR path, $100K+ counts, cash balance ~$4.2B.

Q2 call was light on classic macro (rates/CPI/consumer). Strongest signals are structural: hyperscaler/neocloud GPU CapEx arms race, egress-tax losses, and AI-native platforms defaulting to Cloudflare.

Macro

| Theme | What NET said | Read-through | |---|---|---| | Demand / IT spend | +36% YoY; trajectory “up into the right”; still “prudent” guide | Positive for enterprise security / edge / developer platforms — AI-agent traffic more than discretionary consumer IT | | Macro uncertainty (10-Q) | “Could persist through 2026”; may lengthen sales cycles | Cautious on deal timing, not end-demand quality | | Inflation | “Do not believe inflation has had a material effect” | Neutral / modestly constructive | | Tariffs | Filing risk language; no call traffic-crystal-ball | Relative positive for cloud-native vs appliance shippers | | AI CapEx arms race | Explicitly opts out; hyperscalers/neoclouds signing massive GPU contracts; claims superior utilization | Confirms elevated CapEx intensity at hyperscalers/neoclouds; not demand weakness |

Companies / competitive

| Name / group | Role | Implication | |---|---|---| | Hyperscalers (AWS/Azure/GCP) | Competitors + customer primary infra | Egress tax “would break the economics”; multi-cloud wins; CapEx arms race narrative negative for FCF optics if util stays low | | Neoclouds (GPU pure-plays) | Alternative capacity models | Elevated CapEx intensity confirmed; NET capital-efficiency differentiation | | OpenAI | Customer / content-ecosystem pilot | Constructive bridge AI labs ↔ publishers (pay-per-crawl / content rails) | | Lovable, Replit, Wix/Base44 | Vibe-coding ecosystem | AI-generated apps often default to Cloudflare — share pressure on traditional serverless hosts | | Visa | Scale benchmark for agent payments | Agent rails need ~3 orders of magnitude more TPS than Visa peak — new layer, not a bear call on cards | | Point-solution security/CDN | Being consolidated away | Customer retiring 5–7 point solutions — pressure on niche WAAP/bot/CDN/ZTNA | | AKAM, FSLY, ZS, PANW, FTNT, F5 | Deck competitive map | Platform RFPs favor multi-act suites; structural heat on pure-plays |

Synthesis for the book

1. Hyperscaler CapEx & utilization — cross-check AMZN/MSFT/GOOGL and neocloud capacity narratives.
2. Egress economics — watch pricing concessions / multi-cloud object storage share (R2 continuity).
3. SASE / security platform war — retirement of point solutions keeps heat on ZS/PANW/AKAM/FSLY.
4. AI labs + publishers — OpenAI pilot constructive for content monetization rails.
5. Vibe-coding defaults — share risk for traditional serverless on greenfield agent apps.
6. Do not over-read as macro barometer — demand driver is AI traffic + security consolidation.

Cloudflare delivered a classic beat-and-raise re-acceleration print: revenue $696.1M (+35.9% YoY, +4.7% vs Street), non-GAAP EPS $0.29 (+8% vs Street), FY26 raised to $2.864–2.870B / $1.25–1.26, DBNR 120%, and the first sequential GM improve in eight quarters. Guidance remains a floor (serial sandbagger; 6/6 next-quarter rev beats). The open debate is quality of growth — product mix still undisclosed, long-term 75–77% GM target soft-retired at ~73%, and restructuring costs higher than first guided — against a management team that continues to over-deliver on the top line.

Next test: Q3 vs $736–737M / $0.34; GM hold; FCF vs “unchanged.”