Bid Cap
Company library Exchanges & Financial Data

Company report

FactSet Research Systems Inc. FDS

Three-pass checked

The bet you're really making is that money managers keep paying every year for FactSet, the screen and data feeds their analysts use all day to research investments and build portfolios. You're betting they can't easily quit, because the data is wired into how they work, and that new AI tools make FactSet more useful instead of replacing it. Right now it is mixed: revenue keeps climbing, up about 6% over the year, but profit is shrinking, down almost 15% last quarter as costs rose faster than sales. You pay about 20 times last year's earnings, and by the measure that tracks this company best over twelve years, the price is lower than it has been since 2014.

Key data

Price$301.96
52-week range$185.00 – $377.07
P/E, trailing / FY2028E20x / 14x
EV/EBITDA11.6x

FDS · price with moving averages

Daily · 6MWeekly · 3Y
$171$257$343$429$514 Oct '23May '24Dec '24Jul '25Feb '26Oct '26 Bid Cap
EMAs82140

Source: market data.

The business

FactSet sells the workstation on the analyst's desk: the screen, data feeds and analytics that money managers, banks, hedge funds, and private-equity and venture teams use to research securities, build portfolios and report to clients. Revenue comes almost entirely from subscriptions, billed as Annual Subscription Value, the forward-looking twelve-month revenue from services currently supplied. The filing quantifies ASV and its organic growth but never states recurring revenue as a percentage, so the subscription share is best described plainly, dominant and contractual, rather than pinned to a number the company does not publish. The moat is switching cost stacked on a data network: once FactSet is wired into a firm's risk system and daily workflow, ripping it out means retraining analysts and rebuilding pipes. Competition is Bloomberg, LSEG's Refinitiv, S&P Global and Morningstar, all fighting for the same seats and the same enterprise data budgets.

The numbers

Two things move at once: the top line grinds higher while GAAP profit has turned down.

Quarter endRevenueNet incomeDiluted EPS
May 2025$585.5M$148.5M$3.87
Aug 2025$596.9M$153.6M$4.03
Nov 2025$607.6M$152.6M$4.06
Feb 2026$611.0M$133.1M$3.59
May 2026$622.9M$126.7M$3.50

Revenue rose every quarter, about 6% year over year. But net income peaked in the August 2025 quarter at $153.6M and fell in the three since, to $126.7M, down almost 15% from a year earlier, and diluted EPS followed, 4.06 to 3.50. The small beats you may have seen, 4.53 against 4.45 last quarter, are adjusted EPS; GAAP diluted was 3.50. Adjusted numbers edged past estimates by a point or two even as GAAP earnings fell.

Fiscal yearRevenueNet incomeDiluted EPS
FY2021$1.6B$399.6M$10.36
FY2022$1.8B$396.9M$10.25
FY2023$2.1B$468.2M$12.04
FY2024$2.2B$537.1M$13.91
FY2025$2.3B$597.0M$15.55
2026, 9M to May$1.8B$412.4M$11.15

Step back and the record is a steady compounder: revenue grew about 10% a year from fiscal 2021 to 2025, EPS about 11%, helped by buybacks. The nine months of fiscal 2026 break the pattern, revenue up about 7% but net income down 7% against the same stretch a year earlier. The whole question is whether that divergence is temporary. What this memo believes the market is missing: the fiscal-2026 profit drop looks more like spending and integration than lost pricing, and the print that settles it is the operating margin over the next two quarters.

PeriodRevenueOperating incomeOp margin
FY2024$2.2B$701.3M31.8%
FY2025$2.3B$748.3M32.2%
Nov 2025$607.6M$192.1M31.6%
Feb 2026$611.0M$185.0M30.3%
May 2026$622.9M$166.3M26.7%

There is the crack in one column. Operating margin held near 32% through fiscal 2025, then slid three quarters running to 26.7%. Revenue never stopped growing, so cost is the culprit, not demand. If margin steadies near 30%, the compounding story survives at a twelve-year-low multiple. If it keeps sliding, the de-rating to 11.6 times EBITDA (EV/EBITDA), below peers at 14.6, is the market being right early.

Management

Insiders were net sellers over the past year, about $1.5M sold against $0.3M bought, led by roughly $760K across two January sales and a $588K sale by another insider; plan status is not disclosed, so planned and discretionary can't be separated, though the amounts are small against a $10.7B company. Capital return is the real signal: buybacks rose to $300M in fiscal 2025 from $235M the year before, and debt that spiked to nearly $2.0B in fiscal 2022 to fund an acquisition has been paid down to $1.39B, with interest covered nine times over. This is a company returning cash while it invests, not stretching the balance sheet.

How it fails or surprises you

Margins keep sliding. Operating margin fell from 31.6% to 26.7% across fiscal 2026's three reported quarters while revenue rose. If that is structural, higher data and AI infrastructure cost or real pricing pressure rather than a one-time investment year, the 11% earnings compounding breaks. The print is operating margin over the next two quarters; another leg below 27% is the tell.

The terminal loses its grip. New AI tools let clients query and analyze data without a FactSet seat. If organic ASV growth drifts below about 4%, that is seats and enterprise budgets leaking to cheaper or in-house tools. Watch net new ASV and client and user counts; a stall would confirm the market's fear that the workstation moat is thinning.

The de-rating reverses (right tail). At an EV/EBITDA of 11.6x, the cheapest in twelve years and under peers at 14.6, a return to even 15 times on stable growth is a large move. If FactSet's own embedded AI reaccelerates ASV rather than cannibalizing seats, and margin steadies near 30%, the multiple and the earnings rise together. The print: organic ASV holding mid-single digits with margin stable.

Closing thoughts

The next two quarters of operating margin and organic ASV growth convert the story one way or the other: margin steadying near 30% with ASV holding mid-single digits says the fiscal-2026 dip was an investment year and the decade-low multiple is a gift; another margin leg down with decelerating ASV says the market de-rated for a reason. The left tail, a structural margin and growth decline dressed up as AI disruption, is the one to avoid; the right tail, a re-rating from 11.6x on a business still growing, is worth more than the downside costs because the balance sheet and subscription base buy time to be wrong. On judgment the fatter tail is the recovery, though not by much, and an ambiguous print keeps you waiting rather than committing.

The bet is still that money managers keep paying every year for FactSet and can't easily walk away, and that AI helps the workstation more than it hurts it. What breaks it is simple to watch: operating margin below 27% and organic ASV growth slipping toward 4% in the same quarter would say the earnings power is eroding, not pausing. Until those two turn down together, the falling GAAP profit reads as a spending year bought at a decade-cheap price, not a broken business.

Methodology

Anchored to the Form 10-Q for the quarter ended May 31, 2026, filed July 1, 2026 on EDGAR; revenue, operating income, net income and diluted EPS taken as filed from SEC XBRL company facts.

FactSet's fiscal year ends August 31, so fiscal quarters do not align to calendar quarters; fiscal Q4 2025 (Aug 2025) was derived as fiscal 2025 full-year results less the first nine months; the year-to-date row covers the first nine months of fiscal 2026.

GAAP diluted EPS ($3.50 last quarter) differs from the adjusted EPS on which consensus beats are measured ($4.53 vs $4.45 estimate); Annual Subscription Value and organic growth are company-published measures referenced from the filing, not vendor fields.

Trailing P/E, EV/EBITDA and moving averages are vendor market data as of Sep 5, 2026; EV/EBITDA position uses the pack's 2014-2025 series (current 11.6x, below its 15.5x-29.7x range). Forward P/E uses FY2028 consensus from 13 revenue and 8 EPS estimates and is a thin out-year figure.

Insider values and counts are from the 12-month vendor window; 10b5-1 plan status is not disclosed in that feed. Documentation prepared with AI assistance. Not investment advice.

Fact check: all figures reconciled to filed XBRL or vendor data; growth rates and margins derived from filed quarterly results. No corrections needed. Final analysis verified as of Sep 5, 2026.

Bid Cap

Daily ideas, a 390-name database, and a model long/short book from an investor who mostly covers financials. $70 a month or $700 a year.

Subscribe on Substack