Phase Space AI

Valuation

Intuit [INTU]

Intuit [INTU] — Valuation

As of 2026-07-29. Spot $333.13. Two outputs, per references/valuation.md: a 12-month target and the five-year implied-path test.

---

0. Company state — declared first

STATE A — mature and structurally stable. Confirmed by test.

Evidence. Profitable for 20+ consecutive years. GAAP operating margin 20.2% → 21.9% → 22.3% → 26.1% → 27.5% TTM across five fiscal years — a 7.3pp range, monotonically improving. No transformative acquisition (nothing closed in FY2026). No business-model transition: the DIY-to-assisted shift is a mix change inside an existing product. The August-2025 segment recast is a reporting change with prior periods recast to conform, so there is no accounting-basis break.

The seasonality question, answered. Tax season is a within-year seasonal, not a multi-year exogenous cycle, so Intuit is not State B. State B exists for names whose revenue or margin is set by fab utilisation, memory pricing or a commodity input, where a trailing peak margin paired with a cycle-average growth forecast is internally inconsistent. The 47.0% margin printed in the April quarter is not a cycle peak; it is April. The correct handling — used throughout — is matched-fiscal-period comparison and a TTM window spanning one full seasonal cycle.

Consequence: the reverse DCF is the primary long-horizon instrument; the terminal margin is built from Intuit's own normalised economics plus a forward bridge, with industry data as a sanity band only. Evidence grade A — 81 quarterly periods of normalized statements plus ten quarters of transcripts.

---

1. Inputs, verified

AV defect found and rejected: AV reports shortTermDebt of $833m against the filed $750m — an $83m overstatement that propagates into net cash. The filed figure is used. AV's sellingGeneralAndAdministrative again carries G&A only ($1,656m TTM against $7,901m of S&M + G&A — 4.77x); S&M is derived as operatingExpenses − R&D − G&A. See INTU_Financial_Model_Notes.md.

---

2. Terminal margin — derived, bridged, and justified

The starting point: the actual bridge, TTM to 2026-04-30

Prior-year matched TTM operating margin: 24.4%. Three years ago (TTM to 2023-04-30): 21.7%. Intuit has expanded its GAAP operating margin 5.8pp in three years on matched periods.

The terminal bridge, FY2031

The four checks

1. m_EBIT,T ≤ m_gross,T: 33.0% ≤ 80.5%. Satisfied with 47.5pp of headroom.
2. Full expense bridge: 80.5 − 26.0 − 15.0 − 6.5 = 33.0. Reconciles exactly.
3. Is it below the company's own trailing actual? No — it is 5.5pp above (33.0% vs 27.47%). And the company has already delivered 5.8pp of expansion in three years on matched periods, with a sized, dated 17% headcount reduction still to land. The forward assumption is slower than the demonstrated run-rate. No instance of the recorded defect (21 of 84 names carried terminal margins below their own trailing actual).
4. Non-GAAP cross-check: 33.0% GAAP ≈ 44% non-GAAP − ~9.5% SBC − ~1.5% amortisation. TTM SBC is approximately $2.05bn, 9.8% of revenue — roughly 6.6pp lower than ServiceNow's 16.4% and already near the terminal assumption, so the SBC leg here requires almost nothing. FY2025 non-GAAP operating margin was ~39%; Q3 FY26 non-GAAP operating income was $4.7bn on $8.558bn. Reaching ~44% by FY2031 is 5pp of expansion over five years, against 5.8pp already delivered in three.

Prose justification. Intuit's terminal operating margin is 33.0% because the company has demonstrated 5.8pp of GAAP operating-margin expansion over three matched-period years, is executing a sized and dated 17% workforce reduction whose savings are partly redeployed and partly retained, and carries SBC at 9.8% of revenue — already close to steady state, so unlike ServiceNow there is no large SBC assumption doing hidden work. The gross-margin leg is set BELOW the trailing actual and against the three-year trend, because the growth engine (assisted tax, 53% of TurboTax revenue and rising) is human-delivered and lands in cost of revenue. That single conservatism is the honest core of this build: Intuit is getting more profitable at the operating line while its revenue mix gets less profitable at the gross line, and the terminal margin reflects both.

Contrast with the NOW build, deliberately. NOW's 23.0% requires SBC to fall from 16.4% to under 10% — a company-published commitment, but a 6.9pp swing doing most of the work. Intuit's 33.0% requires SBC to move 0.3pp. Intuit's terminal margin has far less parameter risk in it.

---

3. Exit multiple — identity first, traded multiple as sanity check

EV_T / EBIT_T  =  (1−t)(1−g/ROIC) / (WACC−g)

ROIC is Intuit's actual measured 22.0% ($4,369m NOPAT ÷ $19,835m invested capital) — not an estimate and not an organic proxy, because Intuit's invested capital is not distorted by recent M&A.

Base: 12.5x — the identity at (WACC 9%, g 4%, ROIC 22%). The same base is used on NOW, so the two names are compared on identical terms.

Growth-matched anchoring check. Intuit's growth at the exit year is ~9–11% (decelerating from 14.0% nine-month / 10.4% latest matched quarter). The mature-software comparator set from the enterprise-software cluster memo spans 7.0%–20.6% latest-quarter growth and brackets it: ADBE 11.6x · ORCL 14.9x · INTU 15.9x · MSFT 18.8x · CRM 20.8x · VEEV 27.7x, median 17.5x. Admissible, so UNIDENTIFIED is not declared.

The important observation. Intuit trades at 15.9x today. The identity at a 9% WACC gives 12.4x; at an 8.5% WACC and 4.5% terminal growth it gives 15.1x. Intuit's current multiple is roughly what the reinvestment identity warrants at a modest cost of capital — the market is not paying a premium for this business. That is the opposite of NOW, which trades at 27.2x on a non-GAAP basis against the same identity.

Implied compression at the 12.5x base: from 15.9x today, −21.4%.

---

4. The implied-path test — the Valuation Criteria

Solved for: revenue CAGR. Held fixed: terminal margin 33.0%, exit multiple 12.5x EV/EBIT, WACC 10.0%, 5 years, revenue base $20,925m, net cash $794m, 276.3m diluted shares.

Demonstrated

Demonstrated is taken as 14.2% (three-year TTM). Note the five-year figure is inflated by the post-COVID Mailchimp and Credit Karma acquisitions and by the tax-season timing distortions of FY2021, and the most recent matched quarter is 10.4% — so 14.2% is the generous-but-defensible middle, and the deceleration is a real risk to it.

Result

Required revenue CAGR: 11.2%. Demonstrated: 14.2%. Margin = demonstrated − required = +3.0pp. Valuation Criteria: PASS.

Not "PASS WITH ARGUMENT" — the price requires less than the business has already demonstrated, so no forward argument is needed. Note the margin is only +3.0pp, which sits at the edge of the ±3pp noise band the cluster memo established. Read it as a genuine but narrow pass.

Sensitivity over the exit multiple

Required 5-year revenue CAGR. PASS where the cell is ≤ 14.2%.

Intuit PASSes at every exit multiple from 11.0x upward at the base terminal margin, and at 13.0x upward even with ZERO margin expansion from today. It fails only below ~10x exit. That is a materially more robust result than NOW's, which fails everywhere below 20.8x.

Flip points

Both flip points sit inside what the company has already done. That is what a clean PASS looks like.

Reconciling with the screen's +15.1pp

The screen recorded required −1.1% against demonstrated 14.0%, a +15.1pp margin. From the grid, a required CAGR near zero occurs at roughly a 21x exit multiple. The screen's +15.1pp was manufactured by an exit multiple around 21x — 8.5x above the identity's 12.5x.

And a negative required CAGR is itself the tell. The brief flags negative required growth rates as a symptom of the wrong instrument or the wrong parameter, not of a 15pp-cheap name. Required growth of −1.1% says the price is covered even if Intuit shrinks — which is arithmetically true at 21x and economically not a claim anyone should underwrite.

My +3.0pp is the defensible figure: a PASS, 12.1pp narrower than the screen's, on the same direction.

---

5. The 12-month target

NTM base, on matched fiscal periods. NTM from 2026-07-29 is essentially fiscal 2027 (Aug 2026 – Jul 2027).

Multiple, anchored on Intuit's own history with the percentile stated.

Quarterly EV/EBIT computed from filed TTM EBIT, the filed balance sheet, and the price ~45 days after each period end. Window: the 18 quarters from 2022-04-30 — the post-ZIRP rate regime. The FY2021 observations (50x–72x) are excluded as a different cost-of-capital regime.

$6,960m × 20.0x = EV $139,200m + net cash $794m = equity $139,994m ÷ 276.3m = $507/share.

12-month target: $507. +52.2% to spot.

Why 20.0x and not the 47.7x median. A reversion to the median gives $1,207 (+262%), which is not a forecast — it is an arithmetic restatement of how far the stock has fallen. Intuit's EV/EBIT went 48.9x (2025-04-30) → 37.6x → 35.9x → 24.3x → 15.9x in fifteen months while operating margin rose from 24.4% to 27.5%. That is a pure multiple event — rates plus the priced risk that general-purpose AI agents disintermediate DIY tax and SMB bookkeeping — and there is no evidence it fully reverses inside twelve months. 20.0x is a partial recovery, sitting only 4.1x above spot and still 4.4pp below the identity-warranted 15.1x-at-8.5%-WACC upper case expressed on the same basis, justified by three named, dated items: the 2026 Plan's $300–340m charge clearing in Q4 FY2026 with the savings accruing in FY2027; TurboTax Live at 53% of TurboTax revenue growing 36% into the FY2027 season; and Credit Karma at +21.5% with a documented 54% increase in filings originating in Credit Karma.

Sensitivity. At today's 15.9x the target is $403 (+21.0%); at 25.0x it is $633 (+90.0%).

The caveat that matters. Spot rose 17.1% in the five sessions to 2026-07-29 ($284.47 → $333.13), and the 252-day range is $255.07 – $807.39 — spot is 58.7% below the high. The 6th-percentile reading is measured at the top of a violent bounce off a multi-year low; a week earlier the same calculation gave the 3rd percentile. I do not have a sourced cause for the five-session move and will not invent one.

No external professional target was obtained inside the time box, so the mandated sanity-band comparison is not performed — stated rather than substituted for.

---

6. Criteria summary