Initiated 2026-09-20 · Price $144.29 (as of 2026-09-18 close, stockanalysis.com) · Mkt cap $11.83bn · Technology / property-and-casualty insurance software · Model: verified (generic, full tier)

Rating: Hold — Conviction: Medium

Guidewire is the core system of record for the P&C insurance industry, and the cloud transition that consumed five years of operating losses is now visibly paying: 23% revenue growth, GAAP operating income up from $41mm to $150mm, a 26% operating cash flow margin, and the lowest gross ARR attrition the company has measured. The stock has nearly halved from its 52-week high because FY2027 ARR guidance of 18% constant currency is one point slower than FY2026, and because application software generally is being re-rated on AI-disruption fear. That has moved the name from obviously expensive to roughly fair with a hostile skew: the model's Base still lands 28% below the last close, but the Bull case now clears it, which was not true a quarter ago. Hold rather than Sell because the asset is durable and unlevered; Hold rather than Buy because you have to underwrite the bull driver set to justify the price.

Model value range vs price
Bear $44.21Base $104.60Bull $167.11Price $144.29

Business overview

Guidewire sells InsuranceSuite — the policy administration, billing and claims systems a property-and-casualty carrier runs its book on — plus data, analytics and digital products layered on top, now delivered as Guidewire Cloud. FY2026 revenue of $1,475mm split roughly 66% subscription and support ($971mm, +33%), 16% term license ($235mm, −7% as customers convert to cloud) and 18% services ($270mm, +23%). Earnings turn on three things: net new ARR ($1,242mm at 2026-07-31, +19% constant currency), cloud subscription gross margin as multi-tenant scale arrives (74% GAAP in Q4, +6pp year over year), and stock-based compensation, which at $182mm was 12.3% of revenue — the entire distance between a 24% non-GAAP operating margin and a 10% GAAP one.

Bull case

  1. A quarter of the growth is already contracted. Fully ramped ARR was $1,578mm against reported ARR of $1,242mm — $336mm, 27% of the base, signed and stepping up on schedule. That makes FY2027's guided 18% closer to a floor than a hope. Model: rev_growth
  2. Cloud gross margin has further to run. Subscription and support gross margin hit 74% GAAP in Q4 FY2026, up six points year over year; total gross margin 64%. FY2027 guidance implies a 24.1% non-GAAP operating margin at the midpoint, and incremental margin on a multi-tenant platform that is done being built is very high. Model: ebitda_margin
  3. Retention is close to annuity-grade. Gross ARR attrition ran below 1.5% overall and below 1% in core systems — the lowest since the company began measuring it. Replacing a policy administration system is a multi-year, board-level program; the switching cost is the moat. Model: rev_growth, nwc_pct_rev
  4. The balance sheet lets it buy its own de-rating. Net cash: $753mm of cash and short-term investments against $678mm carrying value of 1.25% converts struck at $244.65, not due until November 2029. $606mm repurchased in FY2026; FY2027 operating cash flow guided to $445–465mm. Model: FY0 net cash (market.cash, credit.debt) enters value directly; the repurchases themselves (share_change, capital_return_pct) move EPS only

Bear case

  1. The base case does not reach the price. Base midpoint $104.60 against a $144.29 close, −28%. Only the Bull driver set — 15–18% revenue growth for five years and GAAP EBITDA margin reaching 28% — values the equity above the current price, and it clears it by just 16%. Model: exit_ev_ebitda, wacc
  2. Growth is decelerating and the flattering tailwind is being withdrawn. FY2027 ARR guided to 18% constant currency from 19%, with management saying attrition normalizes from unusually low FY2026 levels. The one-point step is small; losing a record-low-churn tailwind while guiding down is the part that matters. Model: rev_growth
  3. The exposure is the multiple, not the revenue line. At ~9.5x ARR and 66x FY2026 GAAP EBITDA the valuation does nearly all the work; Verisk trades near 16x EV/EBITDA and Sapiens 13x. Shares went $255.89 to $144.29 with the revenue trajectory intact — the definition of multiple risk. Model: exit_ev_ebitda, terminal_growth
  4. Margin leverage now has to do the work the multiple used to do. Base needs GAAP EBITDA margin to double, 12.1% to 23.5%, over five years — mostly by taking SBC from 12.3% of revenue toward 8%. If growth slows, retention grants rise rather than fall and that path stalls; the Bear case, which holds SBC and sales intensity near current levels, is worth $44.21. Model: ebitda_margin

Valuation & balance sheet

Metric (definition) Current Own history Peers Source, as-of
EV / ARR (EV $11.75bn ÷ FY2026 ending ARR $1,242mm) 9.5x ~16.8x at the 52-week high of $255.89 (est., same ARR) n/a (unverified — Duck Creek private, Sapiens does not disclose ARR) q4fy26 + quote, 2026-09-18
EV / forward revenue (FY2027 guidance midpoint $1,717mm) 6.8x ~12.2x at the 52-week high (est.) n/a (unverified) q4fy26 + quote, 2026-09-18
EV / EBITDA (GAAP operating income + D&A; FY2026 $177.8mm) 66.1x not meaningful before FY2025 — EBITDA was negative through FY2024 VRSK ~16.0x · SPNS ~13.2x (aggregator definitions) q4fy26; gurufocus, 2026-09-19
Rule of 40 (guided FY2027 cc ARR growth + guided non-GAAP operating margin) 42 (18 + 24) 42 in FY2026 on the same construction (19 + 23) n/a (unverified) q4fy26 + slides, 2026-09-03

Model-implied value range (from model-summary.json; generic module, DCF-Gordon and DCF-exit-multiple, midpoints): Bear $44.21 · Base $104.60 · Bull $167.11 per share, i.e. implied returns of −69% / −28% / +16% vs $144.29. These ranges show how the bull and bear drivers translate into value; they are not price targets. Nothing floors at zero — net cash — so the range and the scenario table are fully informative. The price sits between Base and Bull and nearer Bull: the market is paying close to the bull driver set, while Base already embeds a 14% five-year revenue CAGR, a doubling of GAAP EBITDA margin and a 20x exit multiple against Verisk's 16x. That is why this is not a Buy despite a 44% drawdown. The Base return of −28% is more bearish than the Hold: I stop short of Sell because the bear branch is a stall in contracted, sub-1.5%-churn revenue rather than an operating break, because net cash and a live buyback mean nothing can be forced, and because a five-year DCF with a fixed exit multiple under-rewards exactly this kind of retention. That tension is why conviction is Medium, not High.

Balance sheet: net cash $75.0mm; net leverage −0.42x FY2026 EBITDA, gross 3.81x, coverage 21.0x; liquidity $753.1mm of cash and short-term investments, no disclosed revolver. The only debt is the 1.25% converts due 2029-11-01 (carrying $678.1mm; the October 2024 offering priced at $600mm upsized, and the carrying value implies ~$690mm principal after the over-allotment). Conversion price ~$244.65, 70% above the last close, so these behave as straight, almost-free debt that has to be refinanced at market rates in 2029 if the stock does not recover. No agency rating was retrieved in this session.

Model note: Built, full tier, generic module, 0 failing error checks, no scenario CHECKs. verification.status is verified as of 2026-09-20 (LibreOffice's independent recalculation matches the Python values on every formula cell; it was not_run at initiation because the environment's LibreOffice install was missing its Calc component). No unverified inputs, no assumptions without a basis. Four conventions. (1) ebitda is GAAP operating income plus D&A, so SBC is expensed — not the company's non-GAAP figures, which is why the EV/EBITDA above dwarfs the headline multiples quoted elsewhere. (2) Because SBC is expensed, share_change carries only the net diluted-count change (−0.6% in FY2026 despite $606mm repurchased); charging gross dilution too would double-count. (3) cash includes short-term investments and cost_of_debt acts as a net-cash yield; at −$75mm net debt the model books slightly less interest income than the company earns, understating each case a little. (4) Historicals are three fiscal years: FY2023 and earlier predate the cloud-margin inflection and would mislead the base rates. Peer multiples come from a screening service, not filings. Generator note (2026-09-24): in this module share_change, capital_return_pct and cost_of_debt move EPS and the net-cash path only, and peer_pe feeds a cross-check only; none changes the value range, so bull/bear tags no longer cite them. Scenario deltas were rebuilt after multiple_pct shocks were extended to the Gordon method (previously they moved only the exit-multiple half), and EPS now uses average diluted shares.

Scenario stress test

Reasoned from the bull/bear drivers above. The model column comes from the scenario overlays (Base case + shock).

Scenario Effect Mechanism Magnitude Model Δ value vs Base ($/sh)
S1 Fast equity crash − Long-duration, high-multiple asset de-rates in a liquidity event with contracted ARR untouched Medium −$13.12
S2 Slow bear / recession − P&C premium is inelastic and replacement programs are contracted, so revenue holds better than most software; new logos and expansion slip Medium −$12.05
S3 Rapid rate shock − Almost pure duration — no debt to reprice — partly offset by insurers' investment income funding IT budgets High −$16.46
S4 Slow rate grind − Multi-quarter terminal-multiple compression across long-duration software, with no single repricing event Medium −$12.07
S5 Soft-landing cuts + Best regime for the asset: discount rate falls while carrier IT spending stays intact High +$14.13
S6 Recession-driven cuts ± Two-sided and, in the model, mildly positive: the discount-rate relief outweighs two years of slower new business, because the revenue is contracted Low +$1.59
S7a Credit liquidity shock − No balance-sheet channel at all — net cash, 1.25% converts not due until Nov 2029 — so this is pure risk appetite Medium −$8.75
S7b Slow default cycle − Customers are statutorily capitalized regulated insurers, not a leveraged-default cohort; sentiment rather than credit loss Low −$3.50
S8 Stagflation ± A hard market lifts direct written premium, to which cloud pricing is partly indexed, but wage inflation hits services and R&D cost and the discount rate turns hostile Medium −$5.15
S9a Dollar spike − Non-US ARR translated lower; the company reports ARR growth constant-currency precisely because this moves Low −$2.71
S9b Dollar slide + Mirror-image translation tailwind Low +$2.28
S10 Melt-up + Narrow breadth and low volatility push growth-software multiples back up on no change in fundamentals Medium +$13.12
S11 Energy supply shock 0 No material transmission channel: higher loss costs feed a harder market and more premium, roughly offsetting budget pressure — $0.00
S12 Mega-cap/AI derating − The row that already happened. Application software re-rated on fear that AI-native entrants compress vertical-software economics; it took the stock from $255.89 toward $144.29 with the revenue line intact, and can go further High −$14.70

The ranking is the finding: the four worst rows (S3, S4, S12, S2) are all discount-rate or multiple driven, and S6 — recession with cuts — is net positive. Exposure here is to the cost of capital and to sentiment on vertical software, not to operations. The deltas are also small against the $123 spread between Bear and Bull: what you underwrite is the driver set, not the macro overlay. Currently active/on watch per the playbook: S3 partially active; S8, S10, S11 on watch. S3 is the single largest modeled headwind for this name; S10 is where it would run hardest.

Model value change vs Base, by scenario
S3 Rapid rate shock−$16.46S12 Mega-cap/AI derating−$14.70S1 Fast equity crash−$13.12S4 Slow rate grind−$12.07S2 Slow bear / recession−$12.05S7a Credit liquidity shock−$8.75S8 Stagflation−$5.15S7b Slow default cycle−$3.50S9a Dollar spike−$2.71S11 Energy supply shock$0.00S6 Recession-driven cuts+$1.59S9b Dollar slide+$2.28S10 Melt-up+$13.12S5 Soft-landing cuts+$14.13

What would change the call

Upgrades if: FY2027 ARR growth is guided or reported above the 18% constant-currency bar; or gross attrition holds below 1.5% through the "normalization" management flagged while subscription gross margin clears the high-70s — the bull driver set becoming the base one. A further de-rating with the ARR trajectory intact would do it arithmetically. Downgrades if: FY2027 ARR guidance is cut at any quarter; or gross attrition rises above roughly 3%; or subscription gross margin expansion stalls below ~75%; or an AI-native competitor is confirmed to have displaced Guidewire at a tier-1 carrier's core system.

Watch items

Sources