Built by Grok 4.6 High

MDB | Earnings Preview

FQ2 FY2027 (three months ended July 31) has not printed. MongoDB reports Tuesday, September 1, 2026, after the close; the call is 5:00 p.m. ET. Last-reported baseline is FQ1 FY2027 (May 28): revenue $687.6 million (+25.2% YoY vs $549.0 million), Atlas $512.5 million (+29.4% / +29%), non-GAAP operating income $123.2 million / 18% (+200 bps), non-GAAP EPS $1.32 (+32.0%). That print cleared the then-guide high of $664 million by +$23.6 million / +3.6% and raised FY2027. The raise is still in force. Do not treat this quarter as reported.

Growth trajectory — re-accelerated, then guided to slow. Reported revenue YoY trough-ed at +18.7% in FQ3 FY2026, re-accelerated to +26.7% / +25.2%, and the live Q2 guide itself decelerates back toward last year’s Q2 rate. Atlas is where the deceleration is explicit.

Four straight Atlas prints at ≥+29%. The Q2 Atlas pin of ~26% is a −300 bp deceleration vs that streak and vs last year’s Q2 +29%. Company-level +23–24% is Atlas slowing, EA having a timing quarter (~20% on “line of sight” into multiyear deals), not a platform re-acceleration.

Watch items into Tuesday

• Guidance / the real bar. Live Q2 band is $729–$734 million (+23–24% YoY), Atlas ~26%, EA ~20%, non-GAAP operating income $152–$156 million (~21% at the high end), EPS $1.58–$1.61. FMP $735.2 million / $1.61 sits $1.2 million above the revenue high and on the EPS high. A 12-for-12 beat of Street and of the company high, with last-two revenue beats compressed to ~+4%, puts the historical-pattern whisper near $761 million. Berry told Goldman not to expect large intra-quarter swings. An in-line-to-guide print is a beat on the scorecard and a miss versus the 12/12 habit. The FY action (hold vs lift the $2.92–$2.96 billion book) and the Q3 guide (Street already needs $745.3 million / +18.6% YoY) are the second event.
• Tariffs. Not a live item. MongoDB is a consumption-cloud / software name with no goods-cost stack. The analog risk is Atlas consumption, not duty rates. Nothing in the May–August IR trail revises the “environment remains relatively stable” assumption.
• Atlas vs the 26% pin. ≥29% again is the old sandbag and the only fundamental that funds a FY raise. 24–26% is the honest guide they asked the Street to believe. ≤23% is the first Atlas miss of the Desai/Berry era and puts the FY 23–25% Atlas range at risk.
• AI still “early.” Voyage productization (EA Search GA; Atlas embeddings + MCP into coding agents) is confirmatory, not an estimate-mover. Core enterprise workloads still drive the P&L. A quantified Voyage / AI-mix number is the unmodelled positive; another quarter of “still early” is what the AI-premium tape will fade.

Peer tape into 9/1 is already printed and fast: DDOG +36%, NET +36%, TEAM +28%, NOW +24.5%. There is no demand-side excuse if Atlas comes in light of ~26%. CRWD (8/26) and ESTC (8/27) can still change the setup; SNOW prints the next afternoon (9/2). Investor Day is New York, September 29.

Issued May 28 on the FQ1 FY2027 call / 8-K EX-99.1. Never revised. MongoDB guides next quarter + full year (revenue, non-GAAP operating income, non-GAAP EPS). It does not guide gross margin, EBITDA, NRR, customers, or RPO. Classification: CONSERVATIVE vs Street and vs the 12/12 beat record; less conservative than last year’s Q2 growth-rate guide (+23–24% this year vs a +15% guide that then printed +23.7%).

Verbatim (CFO Michael Berry, FQ1 FY2027 call, May 28): “For Q2, we expect revenue of $729 million to $734 million which equates to 23% to 24% year-over-year growth. … We expect Atlas revenue growth of approximately 26%. … For EA and other, we have line of sight into a very strong Q2 and expect to see revenue growth of approximately 20%. … we would encourage you to not expect large swings versus guidance for the current quarter as changes in consumption intra quarter only have a modest impact on revenue within the period.”

Same-quarter last year. Last year’s Q2 guide was $548–$553 million (implied +15% YoY) and printed $591.4 million (+7.4% vs mid / +23.7% YoY). This year’s growth-rate pin is ~800–850 bps more aggressive and essentially matches last year’s actual. They are no longer pretending Q2 is a mid-teens quarter. They are still guiding Atlas to decelerate and telling the Street not to extrapolate the historical 4–7% revenue beat.

Remaining-year arithmetic (not company guidance). Q1 actual $687.6 million + Q2 guide mid $731.5 million = $1,419.1 million in H1. Against the FY mid of $2,940 million, that leaves $1,520.9 million for H2 vs H2 FY26 $628.3 million + $695.1 million = $1,323.4 million → H2 +14.9%. Street H2 (FQ3 $745.3 million + FQ4 $797.2 million = $1,542.5 million) is $22 million above that implied H2 — the FY raise is already in FMP. A Q2 print that only hits $731–$734 million and a Q3 guide at ~$737 million would reopen that gap.

Annual trajectory (fiscal years ended January 31). Revenue growth trough-ed at +19.2% in FY2025, re-accelerated +360 bps in FY2026, and is being guided back down ~350 bps in FY2027. Operating margin is the other half of the tape: −100 bps in FY2025, then +400 bps in FY2026, guided +80 to +150 bps more. This is a margin-expansion / growth-deceleration year unless Atlas refuses to slow to 23–25%.

YoY only. MongoDB is a consumption-based multi-cloud database platform — the driver stack sits above consolidated revenue (Atlas, then Enterprise Advanced and other, then a thin services stub). Internal estimates are n/a.

(a) Current quarter — FQ2 FY2027 (upcoming)

FMP /analyst-estimates has no 2026-07-31 quarterly row; the $735.2 million / $1.61 pair is print-level /earnings (epsActual / revenueActual still null as of 2026-08-17). Segment-level street is unavailable (no VA / Bloomberg). The 26% / 20% mix plus a Q1-like services run-rate foots to the high end of the $729–$734 band, not the midpoint. Clarity (~$10 million services annually, ~breakeven) is already in the raised guide.

Last-reported mix (FQ1 FY2027) — the baseline this print will be read against

Q1 Atlas printed 29.4% against a ~26% guide — the same ~300 bp overshoot now being guided back out of Q2. EA +13% was the under-appreciated beat; Q2’s ~20% EA pin is the one line management is leaning in, not sandbagging.

(b) FQ+1 — FQ3 FY2027 (expected; will be issued Sep 1)

The FQ3 FY2027 guide does not exist. Italic = expected from the conservative-guide policy and the standing FY envelope, not issued. MongoDB has not skipped a next-quarter revenue / nGAAP OI / nGAAP EPS guide.

Implied FQ3 YoY: $745.3 million vs $628.3 million = +18.6% — a ~510 bp deceleration from the Q2 guide’s +23.7%. That is the H2 EA “approximately flat” math showing up in street numbers, not a new Atlas break. Why ~$737.5 million ±~$2.5 million is the expected guide, not $745.3 million: the last seven next-quarter bands have been $4–$5 million wide and set 3.6–6.9% below the eventual print. A $735–$740 million guide against $745.3 million would read as a miss on the outlook even if Q2 clears $735 million.

(c) FY+1 — FY2028 (unguided) and the standing FY2027 book

MongoDB does not issue FY+1 (FY2028) guidance. Investor Day (September 29) is the first place a FY2028 framework could appear.

FY2028 vs FY2027 street: revenue +17.9%, EPS +19.6% — a deceleration from FY2027’s +19.3% / +23.1% versus FY2026 actuals. Arithmetic check: Q1 actual $687.6 million + Q2 street $735.2 million + Q3 $745.3 million + Q4 $797.2 million = $2,965.3 million, which foots to the $2.965 billion annual average. Street is above the company’s high on revenue.

Prior FY2027 guide, issued on the FQ4 FY26 call (March 2): revenue $2.86–$2.90 billion; nGAAP OI $545–$565 million; nGAAP EPS $5.75–$5.93. The May 28 raise was +$60 million / +200 bps of growth at both ends, +$26 million on nGAAP OI, +$0.20 / +$0.21 on EPS, and Atlas FY growth +200 bps to 23–25%.

Tone into the quarter

Management walks into September 1 having raised the FY2027 book on May 28 and having not updated it since. No 8-K, no pre-announce. Conviction on Atlas durability and Rule-of-40 at the FY high end is up versus the conservative March 2 initiation; conviction that Q2 will crush the pin is what they just took away.

Conviction read: The Q2 revenue pin is a floor they want treated as the forecast, not a sandbag they want the Street to fade by another 3–4%. Atlas ~26% is the number they will be judged on. EA ~20% is a timing print. The FY book is a beat-and-raise that still leaves 2H consumption open — that is where a September 1 tone shift would show up.

Assumptions still live

• Atlas consumption stays “strong and consistent”; Q2 growth ~26%. Core enterprise workloads (FSI, tech, media) remain the engine; AI is a smaller, accelerating add-on, not the print.
• In-quarter Atlas consumption no longer moves recognized revenue much. “As Atlas has gotten larger, it has become more predictable and less sensitive to movements from individual customers or cohorts.”
• EA ~20% in Q2 is deal timing, not a new run-rate. Several large multiyear deals with existing customers; they “largely fell where we expected.” Not a pull-forward out of H2.
• H2 EA-and-other ~flat YoY because FQ4 FY2026 ($170.5 million, $90 million tech + $100 million+ FI) is the compare, not because the book emptied.
• Macro “relatively stable,” similar to FY2026. Unfalsified. No post-May 8-K changed it.
• AI is real, growing, and not yet material. Voyage customers doubled QoQ on the Q1 call; vector search “far outpacing” company growth; Frontier Labs = plural; core workloads still drive the P&L. Davidson (June 11): material enterprise AI deployments “12–18 months.”
• No GTM disruption in FY2027. Territories and comp already out; CRO Ryan McVay will not re-cut plans mid-year.
• Clarity (~$10 million services, ~breakeven) is already in the raised book. FedRAMP High is a FY item, not a Q2 P&L swing.
• Non-GAAP tax 20%; Q2 / FY diluted shares 86.3 / 86.7 million. High-confidence plugs.

Tone arc — adjectives up on the year, down on the Q2 beat

Net shift Q4 → Q1: one-notch up on the annual number (Atlas +200 bps, Rule of 40 named), one-notch down on implied Q2 beat convexity. The bull case they want believed is “Atlas is a durable mid-20s compounder and EA is not dying.” The thing they do not want believed is “every quarter beats by $25 million.”

Post-guidance updates (May 28 → Aug 17)

No formal guidance revision. The 4 Aug IR release only set the print date. Silence plus a product-heavy Build Fest is consistent with a company that wants FQ2 judged on the printed guide, not on a whisper.

What Tuesday has to do

1. Atlas YoY vs the ~26% pin. ≥29% = they sandbagged the thing they said they wouldn’t. 24–26% = they meant it. ≤23% = the four-quarter ≥29% streak was the peak.
2. Clear $735.2 million / $1.61 without needing the old +5–7% analog. A $735–$740 million “beat” of street that undershoots the 4% historical overshoot is the soft print — and it is the print Berry pre-conditioned the Street to accept.
3. Give a Q3 number that does not force the Street to fade H2. A sub-$740 million Q3 guide is in character and would be read as a miss on outlook. EA multiyear-deal timing is the swing factor.
4. Hold, raise, or “roll the beat” the FY book. A raise after a merely in-line Q2 is the bull case their Q1 language set up. A hold is the base case they telegraphed. A cut would be a break with four straight beat-and-raise prints.
5. Keep AI language honest. “Early but ready” for a fifth straight quarter is consistent. Any claim that agentic is now moving the needle, without Atlas accelerating through 26%, is a narrative/P&L split.
6. Tee Investor Day. 29 Sep is 28 days later. Listen for whether they want 1 Sep to be a quiet in-line or a new FY pin.

Ranked by information value on September 1. “Consensus” = FMP / published sell-side. Bloomberg and Visible Alpha are not connected.

Scorecard on the 1st

Where management and the Street actually disagree (investing-principles: the only contrarian setup is a credible team repeating something bullish the Street will not underwrite):

1. EA is durable, not in runoff. Four quarters of positive EA growth, Q2 guided +20%, ARR +~11%. Street models treat EA as a declining residual.
2. AI is a 2027–28 dollar story, not a 2026 one. Desai stacks architectural arguments (JSON/BSON, agent memory, Voyage retrieval) while Berry keeps repeating that core enterprise consumption is the P&L. The Street paid up for an AI re-rate after FQ1; an FQ2 that is “just” 26% Atlas with AI still “early” is the gap.
3. The 29% Atlas run is not the run-rate they will underwrite. Berry is asking the Street to accept 26% as honest. FMP is at the high end of a guide that already decelerates Atlas by 300 bps. The print resolves whether the Street was right to fade 29% into 26%, or whether management is still sandbagging a consumption business that has not rolled over.

Window: 28 May 2026 (Q1 print) through 17 Aug 2026. Company and brand news excluding the Q1 print itself and the 4 Aug call notice. Post-print news is Voyage productization — confirmatory, not an estimate-mover.

What this does / does not change

• Does not change Q2 FY27 guide math. Company already guided $729–$734 million with Atlas ~26%. Nothing above is a Q2 shipment beat in hand. Build Fest shipped after quarter-end.
• Does confirm the Q1 AI-retrieval thesis is being executed. Bengaluru extends Voyage into on-prem EA; Build Fest extends it into agentic coding tools. Both are consistent with “leaders remain leaders” in document/operational databases.
• Does not invert sentiment. Street enthusiasm after the Q1 beat-and-raise is already robust (FMP grades: Buy 36 / Hold 6 / Sell 2). This is confirmatory, not a worse→better inversion.
• Governance is clean aside from the Cochran withhold. No C-suite turnover, no litigation, no demand shock, no competitor print that changes share.

Explicitly excluded: Q1 print and 4 Aug call notice; sell-side Buy reiterates / PT cluster; MongoDB.local London (7 May, pre-window); Ireland expansion / CPO appointments (April, pre-window); Voyage acquisition (Feb 2025 — only the productization is in-window).

MDB is a Consistent beater. Versus FMP street it is 12 / 12 on revenue and 12 / 12 on non-GAAP EPS over the last 12 printed quarters, and 4 / 4 on both lines over the last 4. Versus the high end of its own quarterly box it is also 12 / 12. Beat magnitude is deteriorating. Revenue surprises have compressed from +8.6% (FQ2 FY2024) to +3.5% (FQ1 FY2027). EPS surprises have collapsed from +50–100% through FY2026 Q3 to +11.5% / +10.9% in the last two prints as Street finally sat on the high end of the company range. Guidance is a floor, not a forecast — but it is migrating toward informative.

Versus own guide (the more informative bar). Offset: guide issued in Q(N) applies to Q(N+1). Last four:

OI/EPS beats collapsed at the same inflection. Through mid-FY2026, nGAAP OI beats vs mid ran +50–100%. FQ4 FY26 / FQ1 FY27 compressed to +12.6% / +15.1% — the first two quarters after Berry started pinning Atlas and telling the Street the guide “reflects the true strength of the underlying business.”

What they are telling the Street not to do into FQ2. Berry, answering Goldman on the FQ1 call: “we think it is important that our guidance reflects the true strength of the underlying business… As Atlas has gotten bigger, it has become more predictable… we would encourage you to not expect large swings versus guidance for the current quarter.” That is management trying to talk the Street out of extrapolating the old +6% revenue / +50% EPS analog. EA philosophy is unchanged: still conservative (only closed / high-probability deals).

Bar into Tuesday

Two whispers, one print. The historical-pattern whisper ($761 / $1.80) is what two years of 100% beats trained the tape to expect. The new-philosophy whisper is Street $735 / $1.61 plus a token $5–10 million. FQ2 is the test of whether anyone believed him. Directionally, another beat is still the higher-probability outcome. The magnitude the stock has been trained on is the wrong analog.

Last-print +1-day reaction (AMC convention, Daloopa OHLCV): +3.0% ($325.68 on 2026-05-28 → $335.55 on 2026-05-29) on 3×+ normal volume. Do not call the Thursday +10.6% the earnings reaction — that session closed before the AMC release. OptionsLam weekly implied move into this print is ±17.2% (expires Sep 4); monthly ±19.4% (expires Sep 18). The options market is pricing a much larger move than the last realized close-to-close.

Do not treat FY2027Q2 as printed.