MDB · Forward model · Bull case

The Bull case, 20 quarters out

Model as of 2026-08-29

Each segment is projected from its own operating driver, rolled up into consolidated cash flow, and discounted back to a fair value per share. The assumptions below are editable — change them and every number on this page moves with them.

MongoDB reports a single operating segment but publishes, in every 10-Q and 10-K, an exact three-line ASC 606 revenue disaggregation: Atlas-related, MongoDB Enterprise Advanced and other, and Services. That is the split modelled here, and it is the only one used - nothing is split by geography, by product inside Atlas, or by AI workload, because none of those is disclosed. The three verticals foot to reported consolidated revenue with a difference of exactly zero in all 25 quarters from fiscal 2021 Q1 to fiscal 2027 Q1, against our stored revenue series. Fiscal labelling matters more here than for most names: MongoDB's year ends 31 January, so '2027 Q1' is the February-April 2026 quarter and seasonal index 0 is that February-April quarter. The whole model is built on a NON-GAAP basis, because that is the basis on which MongoDB guides operating income and earnings per share, and each figure is labelled. Stock-based compensation of $137.8 million in the basis quarter - 20.0% of revenue - is excluded throughout, and GAAP operating income was negative at -$24.8 million against non-GAAP +$123.2 million for the same quarter. That is the single largest caveat on every margin and multiple in this model: MongoDB's own fiscal 2027 guidance adds back $634.6 million of stock-based compensation to bridge GAAP to non-GAAP operating income, more than the $571-591 million of non-GAAP operating income it is guiding to. The exit is taken on revenue rather than EBITDA for exactly this reason. Two things in the model are assumed rather than reported and must not be read as disclosure. First, the Atlas and Enterprise Advanced gross margins of 73.5% and 89.0%: MongoDB publishes no cost of revenue split between them, only a blended non-GAAP subscription margin of 77.06%, which this pair reproduces to within 0.02 points along with the disclosed 74.49% company non-GAAP gross margin. Second, corporate overhead at 54.73% of revenue, which is what the full-year guide implies (74.49% gross margin less the 19.76% non-GAAP operating margin at the guide midpoints) rather than the 56.58% the basis quarter actually ran at - the guide already assumes operating leverage inside the year, and because the engine's overhead rate is flat with no glide, the further leverage MongoDB describes is carried by the vertical margin ladders instead. What was deliberately rejected: Atlas customer counts as a driver. MongoDB publishes total customers, Atlas customers and customers over $100,000 of ARR, but the first two are floors rounded to the nearest hundred ('66,400+'), not a series, and the evidence says they are the wrong quantity anyway - Atlas accounts grew 19.0% year over year while Atlas revenue grew 29.4%. There is no disclosed subscriber-times-price history for Atlas and none was invented; consumption revenue is driven by workloads running, and the only honest driver is a deseasonalised trend growth rate. Remaining performance obligations were likewise not used as a revenue driver, because the 10-Q states recognition is 'inherently variable at the customers' discretion'. No sell-side consensus was used anywhere: Nasdaq's reported EPS for the basis quarter is neither MongoDB's GAAP $0.05 nor its non-GAAP $1.32, so its estimates cannot be paired with either basis. Seasonality was tested on all three verticals by ratio to a centred four-quarter moving average over 25 quarters, and accepted on Atlas alone: 0.9728 / 1.0013 / 1.0023 / 1.0237, a 5.1-point signal against a worst within-quarter spread of 2.9 points, with a mechanical explanation (a 89-day February-April quarter against 92 days elsewhere, plus a 1.6% November-January workload lift). Enterprise Advanced and Services were left aseasonal because their apparent shapes - 5.4 and 6.6 points - are swamped by within-quarter spreads of 17.6 and 15.5 points. The model's central claim is arithmetic rather than editorial. Base settings reproduce the guided fiscal 2027 Q2 revenue at $733.8 million, inside the company's $729-734 million range, but carry the full year to $2,994 million - 1.2% above the top of the $2.92-2.96 billion guide. No defensible trend rate reproduces that guide, because it embeds a second half decelerating to 13.4-16.4% year-over-year growth three months after the company printed 25.2% and raised the year.

Where each case comes from

From cash flow to fair value

The projected path

The published model's 20 quarters. The horizon and sliders above do not change this table.

Every row is projected. A year-over-year change is shown only where the quarter it compares against exists — an em dash means there is no comparable quarter, not a flat year.

Model revisions

Assumptions are marked to reality as each quarter prints. Every change is appended here, with the fair value the model produced at the time, so the model's own history stays visible.