COIN | Earnings Review — Q2 2026

Verdict: DECELERATING into multi-quarter contraction — third consecutive dual miss (rev/EPS), third straight double-digit revenue decline YoY, and the first clean S&S guide miss in six quarters. Controllable positives (share ATH 10.3%, opex under mid, Coinbase One ATH, Circle same-terms renewal, prediction markets +106% QoQ) are real but have not cleared Street bars while industry volumes and crypto prices are soft.

Print: Total revenue $1,220M (−18.5% YoY vs $1,497M; −5.6% vs FMP $1,292M). Transaction revenue $599M (−21.6% YoY); consumer $452M (−30.5%); institutional $100M (+64.6%, too small to offset). Subscription & services $555M (−15.3% YoY), short of management’s prior $565–645M band. Adj. EBITDA $208M (−59.4% YoY), margin 17.0% (−1,717 bps YoY). GAAP diluted EPS −$1.36 vs FMP −$0.44.

Guidance: Q3 S&S cut to $500–580M (mid $540M, ~−11% vs prior Q2 mid $605M, ~−28% YoY vs Q3'25 $747M). Transaction QTD through Jul 26 only ~$130M (management cautions against linear extrapolation). FY2026 adjusted expenses mid cut ~$100M to $4.325B ($4.20–4.45B). No total-revenue or EPS guide vs Street Q3 ~$1.21B / ~$0.44 and FY26 ~$5.65B / −$0.53.

Tone: More sober on revenue (first S&S miss, weaker txn QTD, explicit run-rate caution); more assertive on cost execution and product diversification (Everything Exchange “real revenue,” Coinbase One ATH, Circle de-risked). Cost-disciplined builder, not a volume optimist.

Contradictions (material): CLARITY timeline/materiality walk-back; OpEx invest→RIF U-turn; “less correlated” vs still high crypto beta; fee denial vs Coinbase One 0-fee shift. Weight management quality down on promise tracking under stress.

Tier-1 catalysts into Q3 (Oct 29): (1) crypto volume recovery vs Street’s ~$700–800M txn assumptions (July pace implies large shortfall if it sticks); (2) CLARITY Senate path after Polymarket odds collapsed to ~25–32% Yes for 2026 (mgmt still optimistic; failure framed as “business as usual”). Product green shoots are secondary.

Consolidated

| Metric | Q1'24 | Q2'24 | Q3'24 | Q4'24 | Q1'25 | Q2'25 | Q3'25 | Q4'25 | Q1'26 | Q2'26 | |---|---:|---:|---:|---:|---:|---:|---:|---:|---:|---:| | Total revenue ($M) | 1,638 | 1,450 | 1,205 | 2,272 | 2,034 | 1,497 | 1,869 | 1,781 | 1,413 | 1,220 | | Rev YoY % | +112.0 | +104.8 | +78.8 | +138.2 | +24.2 | +3.3 | +55.1 | −21.6 | −30.5 | −18.5 | | Net revenue ($M) | 1,588 | 1,380 | 1,129 | 2,197 | 1,960 | 1,420 | 1,793 | 1,710 | 1,339 | 1,154 | | Gross margin % | 86.7 | 86.8 | 85.7 | 86.0 | 85.1 | 83.6 | 86.4 | 87.7 | 86.1 | 84.4 | | Adj. EBITDA ($M) | 1,014 | 596 | 449 | 1,289 | 930 | 512 | 801 | 566 | 303 | 208 | | Adj. EBITDA YoY % | +257.6 | +207.1 | +147.9 | +322.5 | −8.3 | −14.0 | +78.5 | −56.1 | −67.4 | −59.4 | | Adj. EBITDA margin % | 61.9 | 41.1 | 37.2 | 56.7 | 45.7 | 34.2 | 42.8 | 31.8 | 21.5 | 17.0 | | Op. income ($M) | 760 | 343 | 170 | 1,034 | 706 | (25) | 481 | 274 | (21) | (113) | | GAAP diluted EPS | 4.40 | 0.14 | 0.28 | 4.68 | 0.24 | 5.14 | 1.50 | (2.49) | (1.49) | (1.36) |

Revenue drivers

| Driver ($M) | Q1'24 | Q2'24 | Q3'24 | Q4'24 | Q1'25 | Q2'25 | Q3'25 | Q4'25 | Q1'26 | Q2'26 | |---|---:|---:|---:|---:|---:|---:|---:|---:|---:|---:| | Transaction revenue | 1,077 | 781 | 572 | 1,556 | 1,262 | 764 | 1,046 | 983 | 756 | 599 | | Txn YoY % | +187.3 | +138.7 | +98.4 | +194.0 | +17.2 | −2.1 | +82.8 | −36.8 | −40.1 | −21.6 | | Consumer, net | 935 | 665 | 483 | 1,347 | 1,096 | 650 | 844 | 734 | 567 | 452 | | Consumer YoY % | — | — | — | — | +17.1 | −2.2 | +74.5 | −45.5 | −48.3 | −30.5 | | Institutional | 85 | 64 | 55 | 141 | 99 | 61 | 135 | 185 | 136 | 100 | | Inst. YoY % | +283 | +272 | +292 | +285 | +15.8 | −4.4 | +144 | +30.9 | +37.2 | +64.6 | | Subscription & services | 511 | 599 | 556 | 641 | 698 | 656 | 747 | 727 | 584 | 555 | | S&S YoY % | +41.2 | +78.6 | +66.3 | +70.8 | +36.6 | +9.5 | +34.3 | +13.5 | −16.4 | −15.3 | | Stablecoin revenue | 197 | 240 | 247 | 226 | 298 | 332 | 355 | 364 | 305 | 292 | | Stablecoin YoY % | — | — | +43.2 | +31.6 | +50.8 | +38.3 | +43.7 | +61.2 | +2.7 | −12.1 | | Blockchain rewards | 151 | 185 | 155 | 215 | 197 | 144 | 185 | 152 | 101 | 83 | | Txn % of net rev | 68 | 57 | 51 | 71 | 64 | 54 | 58 | 57 | 56 | 52 | | S&S % of net rev | 32 | 43 | 49 | 29 | 36 | 46 | 42 | 43 | 44 | 48 |

Absolute revenue & Adj. EBITDA ($B)

YoY growth trajectory (%)

What the trends say

1. Revenue is decelerating into multi-quarter contraction. Path: +112% → +105% → +79% → +138% → +24% → +3% → +55% → −22% → −31% → −18.5%. Q3'25 bounce was one-quarter vol; trend since is down. Q2'26 is the first re-acceleration of the down-leg (+1,203 bps) but still deep red.
2. Consumer trading is the hole. Consumer $452M (−30.5%) vs institutional $100M (+64.6%). Consumer still ~75% of transaction revenue.
3. Subscription diversifier is cracking. S&S went from +37–79% YoY in 2024 to −16% / −15% in Q1–Q2'26. Stablecoin flipped from +38–61% to −12.1% — rate/USDC-balance sensitivity showing.
4. Profitability compresses faster than revenue. Adj. EBITDA margin 61.9% → 17.0% (−4,490 bps peak-to-trough). Gross margin stable (+83 bps YoY) → problem is opex leverage and crypto MTM, not COGS.
5. Mix has structurally shifted. S&S rose from ~32% of net revenue (Q1'24) to ~48% in Q2'26 as trading compressed — higher mix is partly denominator effect.

Five-year annual

| Metric | FY2021 | FY2022 | FY2023 | FY2024 | FY2025 | |---|---:|---:|---:|---:|---:| | Total revenue | 7,839 | 3,194 | 3,108 | 6,564 | 7,181 | | Rev YoY % | — | −59.3 | −2.7 | +111.2 | +9.4 | | Transaction revenue | 6,837 | 2,356 | 1,520 | 3,986 | 4,055 | | Subscription & services | 518 | 793 | 1,407 | 2,307 | 2,828 | | Adj. EBITDA | 4,090 | (371) | 964 | 3,348 | 2,808 | | Adj. EBITDA margin % | 52.2 | −11.6 | 31.0 | 51.0 | 39.1 | | Diluted EPS | 14.50 | (11.83) | 0.37 | 9.48 | 4.45 |

Annual read: FY2024 +111% rebound → FY2025 only +9.4%; 1H26 annualizes ~$5.3B, well below FY2025 $7.2B. S&S 5-yr CAGR ~53% remains the durable positive; transaction barely grew in FY2025 (+1.7%).

CLEAR MISS on revenue and EPS — third consecutive dual miss. L4Q beat rate 25% (consistent misser, deteriorating). Management does not guide total revenue or EPS — pure Street vs actual.

This quarter vs consensus

| Metric | Consensus | Actual | Variance | Verdict | |---|---:|---:|---:|:---:| | Total revenue | $1,292M (FMP) | $1,220.1M | −$72M / −5.6% | MISS | | Transaction revenue | ~$628M (CoinDesk) | $599.2M | −$29M / −4.6% | MISS | | Subscription & services | ~$599M Street; mgmt $565–645M | $555.1M | −7.3% vs Street; below mgmt low | MISS | | GAAP diluted EPS | −$0.44 (FMP) | −$1.36 | −$0.92 | LARGE MISS | | Adj. EBITDA | n/a clean Street | $207.8M | — | 14th consecutive +Adj. EBITDA |

Heatmap (last 8 quarters — this quarter highlighted)

Beat rates & pattern

| Window | Revenue beat rate | EPS beat rate | Strict dual beat | |---|---:|---:|---:| | L12Q | 5/12 (42%) | 7/12 (58%) | 5/12 (42%) | | L4Q | 1/4 (25%) | 1/4 (25%) | 1/4 (25%) |

Pattern: Mixed historically → Consistent misser recently (deteriorating). Only Q3'25 beat in L4Q. Surprise magnitude: sign stuck negative; Q2'26 miss size back to mid-single digits (−5.6%). Beats cluster in crypto up-cycles; misses dominate cooling tape.

Management variance explanation

1. Macro / volumes: Spot volumes down >20%; market cap double-digit down; vol at multi-year lows — “market headwinds once again outpaced growth.”
2. S&S shortfall (admitted): (a) later-than-expected on-platform USDC deals; (b) larger crypto price headwinds on staking / blockchain rewards.
3. Controllables: Record 10.3% trading-volume share; S&S ~48% of net rev; 88% net rev non-BTC spot; every expense line below mid; 14th consecutive +Adj. EBITDA; Coinbase One paid subs ATH; prediction markets 2× QoQ.

Headline: Q3 S&S mid cut ~11% to $540M (~−28% YoY) and FY adj. opex mid cut ~$100M to $4.325B. First S&S low-end miss; expense credibility intact; no total-rev/EPS guide vs Street recovery math.

New outlook (as of 2026-07-30)

| Metric | New guide | Notes | |---|---|---| | Transaction rev QTD | ~$130M through Jul 26 | Explicit caution vs linear extrapolation | | S&S revenue | $500–$580M | Mid $540M; “roughly flat with Q2” | | Transaction expenses | Mid-teens % of net rev | ~flat vs Q2 16.4% actual | | Adjusted expenses | $980–$1,080M | Mid $1,030M; new combined range | | SBC | ~$245M | ~flat vs Q2 | | Total revenue / EPS / margins | Not guided | Street owns the gap |

FY2026 adjusted expenses (only formal FY guide)

| | Prior mid (Q1 call) | New mid (Q2 call) | Change | |---|---:|---:|---:| | Low | $4,250M | $4,200M | −$50M | | High | $4,600M | $4,450M | −$150M | | Midpoint | $4,425M | $4,325M | −$100M / −2.3% | | Range width | $350M | $250M | Narrowed $100M |

Waterfall — S&S mid ($M)

Prior guide vs Q2 actual (credibility)

| Metric | Q2 outlook (May) | Q2 actual | Result | |---|---|---|---| | S&S | $565–645M mid $605M | $555.1M | MISS low by ~$10M (−8.2% vs mid) | | T&D + G&A | $820–870M | $830M | In-range, below mid | | S&M | $200–300M | $240M | In-range, below mid | | SBC | ~$240M | $237M | Slightly under | | Txn expenses | Low-to-mid teens % net rev | 16.4% | Slightly above band (prediction-market mix) |

S&S guidance accuracy tracker (6 quarters)

| Actual period | Guide mid | Actual | vs Mid | Class | |---|---:|---:|---:|---| | Q1'25 | 725 | 698 | −3.7% | In-range | | Q2'25 | 640 | 656 | +2.5% | In-range | | Q3'25 | 705 | 747 | +5.9% | Beat (above high) | | Q4'25 | 705 | 727 | +3.2% | In-range | | Q1'26 | 590 | 584 | −1.1% | In-range | | Q2'26 | 605 | 555 | −8.2% | Miss (below low) |

Read: Expense guide is the clean positive (deliver under mid; FY cut & narrowed). The only formal top-line guide was cut and the path worsens YoY (~−28% on Q3 S&S mid). Street Q3 ~$1.21B rev / ~$0.44 EPS and FY26 ~$5.65B require H2 volume recovery management will not underwrite.

Tone: Q1'26 → Q2'26

| Dimension | Q1'26 | Q2'26 | Delta | |---|---|---|---| | Market macro | Volumes/mcap −20%+ | Same severity; multi-year low vol | Unchanged | | Revenue confidence | Hit S&S mid; “durable buffer” | Missed S&S; lower Q3 band | Weaker | | Expense confidence | RIF ~$500M; FY $4.3–4.6B | Every line under mid; FY $4.2–4.45B | Stronger | | Product | Green shoots | “Real revenue”; prediction mkts 2×; CB1 ATH | More confident | | Circle / USDC | Auto-renew language | Conditions met; same terms | Higher confidence | | CLARITY | End-of-summer signed | Optimistic on floor vote; failure path | Slightly more realism |

| Metric | Q3'24 | Q4'24 | Q1'25 | Q2'25 | Q3'25 | Q4'25 | Q1'26 | Q2'26 | |---|---:|---:|---:|---:|---:|---:|---:|---:| | Total revenue ($M) | 1,205 | 2,272 | 2,034 | 1,497 | 1,869 | 1,781 | 1,413 | 1,220 | | Rev YoY % | +78.8 | +138.2 | +24.2 | +3.3 | +55.1 | −21.6 | −30.5 | −18.5 | | Rev accel (bps) | −2,599 | +5,937 | −11,394 | −2,094 | +5,177 | −7,664 | −895 | +1,203 | | GAAP diluted EPS | 0.28 | 4.68 | 0.24 | 5.14 | 1.50 | −2.49 | −1.49 | −1.36 |

Inflection points

| | Quarter | What happened | |---|---|---| | A | Q4'24 | Revenue YoY peaks +138.2% (+5,937 bps) — cycle high / post-election vol | | B | Q1'25 | Hard deceleration (−11,394 bps) as 2024Q1 ETF-launch base hits | | C | Q2'25 | Growth trough at +3.3% (still positive, stall-speed) | | D | Q3'25 | One-quarter re-acceleration to +55.1% | | E | Q4'25 | First negative YoY of the window (−21.6%) | | F | Q1'26 | Deepest decline (−30.5%) | | G | Q2'26 | First re-acceleration of the downturn (+1,203 bps to −18.5%) |

Trajectory verdict: Cycle peak → multi-quarter deceleration → negative YoY → first rate improvement of the down-leg. EPS is a three-quarter GAAP loss streak with modest loss-narrowing (−$1.49 → −$1.36) — do not read EPS YoY “acceleration” as operating turn (easy/impossible comps + crypto MTM). Management frames diversification (S&S, Coinbase One, derivatives, prediction markets, multi-stablecoin) as buffering the trading downcycle even as headline revenue remains negative YoY.

| # | Catalyst | Timing | Consensus / market | Mgmt stance | Stock implication | |---|---|---|---|---|---| | 1 | CLARITY Act | Imminent–H2'26 | Polymarket Yes-on-2026 ~25–32% (was ~80%+ Feb) | “Pretty optimistic” on floor vote; failure = “business as usual” | Highest binary. Odds already discounted; asymmetric upside if surprise pass | | 2 | Crypto volume recovery | Continuous; critical for Q3 print 2026-10-29 | Street embeds material txn recovery (~$700–800M band cited) | Share ATH 10.3% despite soft market; “Bitcoin will come back” | Dominant earnings catalyst. July MTD $130M implies large shortfall if pace sticks | | 3 | Everything Exchange (prediction mkts, perps, equities) | In market now | Optionality in models | “Real revenue”; prediction mkts +106% QoQ; ~$100M+ ann. | Medium-term diversifier | | 4 | Deribit → app options (non-US) | Sep 9, 2026 | Integration through 2026 | Fully integrate 2026; US options later | Dated Q3 narrative catalyst | | 5 | USDC / Circle / multi-stablecoin | Renewal already met | Secular, rates-sensitive | Same-terms renewal; multi-stablecoin (Onyx) additive | Structural de-risk; stablecoin rev still $292M (−12% YoY) | | 6 | Base + agentic (x402 / AiFi) | Ongoing | Long-duration optionality | Early lead; no numeric forecast | Strategic moat narrative, not Q3 P&L | | 7 | Coinbase One scale | Continuous | Quality mix; take-rate debate | ATH paid subs in down market; net accretive claim | Counter-cyclical engagement | | 8 | Q3 print + expense execution | 2026-10-29 | FMP ~$1.21B / ~$0.44 | Control what they can (opex) | Forced reset of volume assumptions |

Prioritization: Tier 1 = volume recovery vs Street txn bar + CLARITY post-recess. Tier 2 = Sep 9 options, prediction markets scale, USDC balances. Tier 3 = Base/agentic, US options/perps, L2 consolidation.

Format: Live X Spaces Q&A (Brian Armstrong / Alesia Haas), 11 scored pairs. Scorecard: 7 Well Answered (64%), 3 Deflected (27%), 1 Partial (9%).

| # | Analyst | Topic | Quality | Takeaway | |---|---|---|---|---| | 1 | Eric Pan | CLARITY downside | Well Answered | Optimistic on floor vote; failure = business as usual; consumers lose more than COIN | | 2 | Owen Lau (Clear Street) | Onyx USD vs Circle | Well Answered | Circle renews same terms; multi-stablecoin strategy; killed renegotiation rumor | | 3 | Brian Jung | Crypto-native “pivot” | Well Answered | Multi-constituency, not a pivot (Cobie / Base app) | | 4 | Kenneth Worthington (JPM) | Senior departures | Deflected | Succession + “strategy unchanged”; never why cluster of exits | | 5 | Austin Hankwitz | Agentic trust vs price | Well Answered | Agents care about reliability/compliance too (AWS analogy) | | 6 | Alexander Markgraff (KeyBanc) | Cross-product / marketing $ | Well Answered | 1-year payback; prediction markets incremental spot; $ allocation not quantified | | 7 | Eric Pan (r2) | Diversification / agentic forecast | Partial | Strong on mix + CB1 ATH; no agentic numbers | | 8 | Owen Lau (r2) | Pre-IPO perps / US timeline | Deflected | Roadmap only; no names, cadence, or US date | | 9 | Brian Jung (r2) | Robinhood L2 competition | Well Answered | Fragmentation→consolidation; Base KPIs (~$32T stablecoin transfer LTM) | | 10 | Kenneth Worthington (r2) | Hyperliquid / USDC economics | Well Answered | Happy to share economics to grow USDC to #1 AUM | | 11 | Austin Hankwitz (r2) | CB1 take-rate math | Deflected | Self-labeled “unsatisfying”; refuses rev per $ traded member vs non-member |

Open quantitative gaps: (a) take rate under Coinbase One migration; (b) agentic monetization; (c) pre-IPO perps US timeline; (d) senior-exit causality. Street left with a clearer strategic map but still cannot tightly model take rate or AiFi if volumes stay soft.

Material narrative contradictions (not pure arithmetic falsehoods). Red-border cards for the high-severity set.

| # | Theme | Severity | Investor implication | |---|---|---|---| | 1 | CLARITY timeline + materiality | High | Treat legislative catalyst as option, not base case | | 2 | OpEx invest → RIF U-turn | High | Cost story is reactive; AI-native framing may mask volume stress | | 3 | “Less correlated” vs crypto beta | High | Platform breadth real; earnings beta still high | | 4 | Fee denial vs CB1 0-fee | Medium–High | Model trading take-rate compression explicitly | | 5 | USDC primacy vs multi-stablecoin / shared economics | Medium | Gross Circle share ≠ net retained economics | | 6 | “Nothing changing” amid RIF + senior exits | Medium | Bundle of actions is operating-model change | | 7 | “+Adj. EBITDA any market” vs GAAP losses | Medium (framing) | Parse non-GAAP carefully; durability signal can mislead |

Consistent (not contradictions): Circle auto-renew/same terms; Everything Exchange roadmap direction; share gains in down markets; Base agentic share claims.

Macro

| Theme | Signal | Read | |---|---|---| | Crypto trading | Volumes down; low vol multi-year | Soft risk-asset consumer; pure txn-fee models pressured | | Membership / engagement | Coinbase One ATH in down market | Sticky platform ARPU; quality users stay, casual volume fades | | Policy | CLARITY live binary; mgmt ~optimistic vs ~30% market odds | Highest-signal “macro” is Washington, not Fed/CPI (absent on Q2 call) | | BTC mix | ~12% of business (was >50%) | Intentional decoupling from pure BTC beta; cycles still matter for absolute $ | | Institutions | G-SIBs building on infrastructure; trading/hedging soft | Flow revenue soft; custody/tokenization/rails firm |

Company / entity map

| Name | Type | What COIN signaled | Implication | |---|---|---|---| | Circle | Core partner | Contract conditions met; renews same terms | Near-term USDC economics de-risked for both | | Tether (USDT) | Competing stablecoin | USDC #2 on AUM; will share economics to win #1 | Competitive pressure on Tether AUM narrative | | Onyx USD (JPM ecosystem) | Prospective stablecoin | Multi-stablecoin platform; FX/revenue opportunity | Better for COIN as rails; mixed for Circle exclusivity | | Robinhood (HOOD) | Competitor (crypto + L2) | Fragmentation then consolidation; Base 2-yr head start, ~$32T LTM stablecoin transfer | Near-term retail mindshare fight; structural network-effects defense of Base | | Stripe | Coopetition (own chain + x402) | Named among chain launchers | Fragmentation risk to Base payments; coopetition on agent standards | | Hyperliquid | Partner / large USDC holder | Happy to share economics for network effects | Better for Hyperliquid; mixed for exclusive COIN USDC rewards claim | | SpaceX | Pre-IPO perps underlying | Early traction encouraging; US on roadmap | Product narrative; private-markets demand signal | | Ethereum / Solana | L1 / L2 base | Base largest ETH L2; Solana acknowledged | ETH L2 fee/security demand; competitive for agent activity | | Bitcoin | Asset / mix | Cycles; will “come back”; mix down to ~12% | Soft BTC hurts pure BTC-beta names more than diversified COIN | | SEC / CFTC | Regulators | Agency rules as CLARITY contingency | Operating floor even without legislation | | CME / CBOE (implied) | Derivatives peers | Perps, options path, Hyperliquid ecosystem | Contest for institutional crypto derivatives share | | Kraken / Gemini / offshore (implied) | Exchange peers | Trust + product breadth + share gains in soft markets | Flight-to-quality favors COIN when agents/humans optimize trust |

Bottom line for peers

• Down crypto trading + sticky Coinbase One — bearish near-term for pure transaction-fee crypto brokers until vol returns; bullish for subscription stickiness comps.
• CLARITY is the live policy binary for regulated U.S. venues (COIN, HOOD crypto, CME crypto products).
• Circle auto-renews while COIN multi-homes stablecoins and discounts HOOD/Stripe L2 threats via Base/USDC network-effects thesis — watch whether economics-sharing for USDC #1 AUM dilutes net retained share.

1. Trajectory, not the dual miss alone, is the problem. Three quarters of double-digit revenue contraction, S&S now negative YoY, Adj. EBITDA margin at 17%. One +1,203 bps rate improvement does not make a turn.
2. Respect opex execution; do not treat commentary as a volume inflection. FY opex mid cut ~$100M with high hit-rate under mid is real self-help.
3. S&S is no longer guidance-proof. First low-end miss + Q3 mid ~−28% YoY undercuts the “durable buffer” absolute-dollar story even as mix % rises.
4. Estimate risk skewed to the downside into Oct 29 if July txn pace persists and Street keeps ~$1.21B Q3 / recovery FY26 math without a company total-rev backstop.
5. Stock-relevant catalysts remain exogenous (volumes + CLARITY). Product milestones (Sep 9 options, prediction markets, CB1) support the multi-year Everything Exchange thesis if clouds partially clear.
6. Management quality score should reflect narrative non-stationarity on CLARITY, correlation, fees, and OpEx posture — even where individual statements are locally defensible.

HOLD on fundamentals: market-leading franchise with genuine product diversification and cost discipline, still in a high-beta crypto earnings trough with a deteriorating beat/miss record and an S&S guide that is still sliding.