ServiceTitan [TTAN] — Financial Model Notes

As of 2026-07-29 · all figures $000 unless stated · fiscal year ends January 31

Every number in this file is traced to a primary filing. No third-party financial data was used. FY2026 = the year ended 2026-01-31. FY2027 = the year ending 2027-01-31 (current year).

1. Quarterly revenue build — the four periods that make TTM

The screen's TTM of $1,014,097 is correct and reproduces exactly. Two of the four quarters are not separately tagged in XBRL and must be derived by differencing cumulative year-to-date periods — the step that produces "a TTM that skipped Q4" when done wrong.

Prior-year TTM (to 2025-04-30) for the YoY comparison, same method: 192,994 + 199,275 + (771,878 − 562,597 = 209,281) + 215,692 = 817,242 → TTM growth +24.1%.

Modelling note. ServiceTitan is seasonal: "demand for our customers' services tends to increase during the second quarter of our fiscal year, as hot weather in the summer months typically results in higher demand" (10-K). Sequential revenue growth is strongest in fiscal Q2 (May–July). Any quarterly model must carry that seasonality on the usage line especially, since it is directly GTV-linked. Q1→Q2 sequential: FY26 +12.4% (215.7 → 242.1); FY27 guided +5.7% (268.8 → 285.0 mid), which on the demonstrated beat pattern is ~+10.9% (268.8 → 298.0).

2. Income statement — annual, as filed

Margins

Two things a modeller must not miss. (1) R&D as a percentage of revenue went UP in Q1 FY27 (31.5% → 32.7%) while every other line levered — the agentic-AI build. On a cash basis (ex the SBC inside R&D) it grew +18.1% against revenue +24.6%, so leverage is intact, but the GAAP line is the one that hits the DCF. (2) The FY2025 G&A jump (+57.7% on FY2024) is IPO-related and is not a run-rate; using FY2025 as the base for opex growth overstates the FY2026 improvement. The TTM measure in §5 avoids that.

3. Revenue disaggregation — the mix

Growth: subscription +28.1% / +25.9% / +24.2%; usage +23.9% / +22.6% / +29.3%; total +25.6% / +24.5% / +24.6% (FY25 / FY26 / Q1 FY27).

Usage revenue is recognised NET of interchange (10-Q revenue-recognition note). This is the single most important accounting fact for the margin model — it means the ~22% payments mix carries software-like gross margin, and is why platform gross margin is 78.7% rather than the 55–65% a gross-basis payments hybrid would show. Do not model an interchange cost line; it does not exist in this P&L.

GTV and take rate — the driver layer

Model GTV first, then apply the take rate, then add the subscription build. That ordering matches how the business actually works and makes the exogenous nature of the growth explicit rather than hiding it inside a revenue CAGR assumption.

4. Balance sheet and cash flow

Debt. The ~$107.0m term loan was voluntarily repaid in full in January 2026, immediately before the Second Amendment to the Wells Fargo credit agreement, which raised the revolver from $140m to $250m and extended it to 2031-01-30. $1.5m of unamortised discount/issuance cost was written off to loss on extinguishment. "As of April 30, 2026, no loans were outstanding under the Amended Credit Agreement." Only $0.4m of unsecured letters of credit. Any model carrying debt for TTAN is using a stale balance — this is the second half of the screen's net-cash error.

Cash flow

Q1 is seasonally cash-negative (annual bonus and payroll-tax timing); the FY figure is the meaningful one. FY26 operating cash flow of +$110.1m against a GAAP net loss of $(159.9)m is a $270m gap, of which SBC is $197.1m — the accrual quality is favourable, not stretched.

Working-capital quality

Annual DSO is flat. The Q1-on-Q1 +3.8 days is quarter-end timing, not a trend — the annual measure did not move. The allowance fell from $11,963 to $9,585 (17.6% → 13.1% of gross AR) while AR rose; noted as a mild quality drag, immaterial at this scale, and still conservative in absolute terms.

5. Stock-based compensation — modelled separately because it has two distinct components

Q1 FY27 SBC by line: platform COGS $1,620; prof. services COGS $1,473; S&M $6,276; R&D $19,373 (vs $11,010 PY, +75.9%); G&A $25,832 (of which $13,074 is Co-Founder RSUs).

The Co-Founder award, modelled correctly. 6,483,088 performance RSUs granted October 2024 (3,241,544 to each Co-Founder), vesting on VWAP hurdles from $140 to $440, expiring 2034-10-21. Under ASC 718 the market-condition expense is recognised over the derived service period and is not reversed if the hurdle is never met. At $78.40 the lowest hurdle is 78.6% away.

Correct treatment in a model: - P&L: carry the ~$52m/yr charge through the forecast (it is a real GAAP expense and it is running). - Share count: do not include the 6.48m shares below $140. - Above $140: add all 6,483,088 shares. This makes the diluted share count discontinuous at $140 — ~95.4m below, ~101.9m above (before organic dilution). Any target above $140 computed on 95.4m shares is overstated by ~6.4%.

Organic dilution runs ~5%/yr: 90,048,337 shares (2025-01-31) → 94,601,151 (2026-01-31) = +5.06%. Use 5% for the forecast; it is directly observed, not assumed.

6. Non-GAAP bridge, as the company defines it

Q1 FY27 non-GAAP operating income $40.8m (15.2% margin) vs $16.2m (7.5%) PY. FY27 guide: non-GAAP operating income $142–147m on revenue $1,130–1,140m = 12.6% margin, below the 15.2% already achieved in Q1 — consistent with either reinvestment or the same sandbagging visible on revenue.

Judgement on the add-backs. SBC (both components) is a real economic cost and is included in the terminal margin used in the valuation — I do not adopt the company's non-GAAP margin as a terminal input. Amortisation of acquired intangibles ($45.2m, from the FY2023 Aspire deal) is non-cash and will run off; lease losses ($11.0m, down from $39.1m) are genuinely non-recurring abandoned-office costs. None of the add-backs is aggressive by software-industry standards, and the Co-Founder RSU line is disclosed separately rather than buried in total SBC, which is better practice than most.

7. Guidance record — reproduced in full, because it is the estimate base

Quarterly beat: 6 of 6, mean +4.57%, range +3.67% to +5.73%. Non-GAAP operating income is sandbagged far harder: FY26 was guided at $48–53m initially and printed $94.058m (+86% on the initial midpoint). Do not extrapolate that magnitude — it reflects first-year-public conservatism — but do not model to the guide either.

House FY27 estimate: $1,190m (+23.8%), from three convergent methods (Valuation §3.2). No consensus estimate was retrievable for comparison; Alpha Vantage returned its 25/day quota-exhaustion message.

8. Key metrics as disclosed — with their disclosure limits marked

9. Valuation inputs — the corrected set, for reuse

spot                78.40      (2026-07-28 close; latest print 2026-07-29 intraday 81.47)
shares_outstanding  95,397,579 (82,746,425 Class A + 12,651,154 Class B; 10-Q cover, 2026-05-31)
shares_diluted_wavg 95,003,992 (Q1 FY27 basic & diluted — identical, loss-making)
market_cap          7,479,170,194
cash                421,531,000
debt                0
net_cash            421,531,000
enterprise_value    7,057,639,194
revenue_ttm         1,014,097,000   (to 2026-04-30)
ev_sales_ttm        6.96x
gross_margin_ttm    70.9%
op_margin_ttm       -14.3%
op_margin_delta_yoy +13.3pp
growth_ttm_yoy      +24.1%
vol_252d            54.1%
adv_252d_usd        105,715,590
momentum_12_1       -33.6%

Do not use: CommonStockSharesOutstanding from XBRL (pre-IPO, 35,397,085, last tagged 2025-01-14) or any debt balance dated before 2026-01-31.