Phase Space AI

02 Financial Model Notes

Cooper Companies [COO]

The Cooper Companies [COO] — Financial Model Notes

Workbook: COO_Financial_Model.xlsx · 7 tabs · live formulas throughout · verified by opening in Excel and reading computed cells back via AppleScript (the workbook was closed with close every workbook saving no before each re-open, so no stale copy was read — that gotcha bit once during this build and produced a false "CFO tie-out fails" reading that had already been fixed on disk).

Tabs: Assumptions · Income Statement · Balance Sheet · Cash Flow · DCF · Scenarios · Comps. Blue = hardcoded input; black = formula. Per standing convention, the top line of each statement block carries the $ sign and every row beneath it is a plain number; per-share, price and target cells always carry $.

1. Tie-out — the check that actually matters

A zero balance check verifies internal consistency, not input accuracy. Every actual-year line is therefore tied back to the filed statement, and the tie-out block lives in the workbook. All seventeen checks return 0.0:

Check Model Filed Δ
Revenue FY2025A 4,092.4 4,092.4 0.0
Gross profit FY2025A 2,682.1 2,682.1 0.0
Operating income FY2025A 682.9 682.9 0.0
Net income FY2025A 374.9 374.9 0.0
Revenue FY2024A 3,895.4 3,895.4 0.0
Operating income FY2024A 705.7 705.7 0.0
Revenue FY2023A 3,593.2 3,593.2 0.0
Operating income FY2023A 533.1 533.1 0.0
Total assets FY2025A 12,394.8 12,394.8 0.0
Total liabilities FY2025A 4,155.7 4,155.7 0.0
Total current assets FY2025A 2,106.4 2,106.4 0.0
Total current liabilities FY2025A 1,112.8 1,112.8 0.0
Total assets FY2024A 12,315.2 12,315.2 0.0
Total assets FY2023A 11,658.9 11,658.9 0.0
Cash from operations FY2025A 796.1 796.1 0.0
Capex FY2025A (362.4) (362.4) 0.0
Cash from operations FY2024A 709.3 709.3 0.0

Balance check: 0.2 across all nine years — a disclosed rounding/restatement-vintage plug carried as an explicit named line, not absorbed silently. Cooper's own filed totals (Assets 12,394.8 vs the sum of its named components 12,392.7) do not reconcile to the penny across filing vintages; per references/edgar-pipeline.md this is disclosed rather than chased.

Every opex line also ties by construction: FY2025A gross profit 2,682.1 − SG&A 1,627.8 − R&D 172.2 − amortisation 199.2 = operating income 682.9, exactly as filed; 682.9 − interest 100.0 − other 16.4 = pre-tax 566.5; 566.5 − tax 191.6 = net income 374.9.

Non-GAAP tie: the modelled non-GAAP EPS bridge reproduces the company's own reported figures — FY2023A $3.24, FY2024A $3.69, FY2025A $4.13 (company-reported, FY25Q4 release: "Fiscal 2025 non-GAAP diluted EPS of $4.13, up 12%"). FY2026E lands at $4.62, the midpoint of the company's $4.58–4.66 guidance.

2. Data sources, line by line

Item Source
FY2023A–FY2025A income statement, balance sheet, cash flow SEC EDGAR XBRL companyfacts, CIK 0000711404, grouped by each fact's own end date with 350–380-day duration filtering (never by fy), latest filed wins on restatement
H1 FY2026 actuals Filed 10-Q, six months ended 30 April 2026 (accession 0001628280-26-041305)
Segment and category revenue FY2025 10-K, Business Segment Information (R-files R80/R81/R82) — company-facts strips dimensional data, so R-files were used
Quarterly revenue, category, geography, guidance history 20 consecutive 8-K Item 2.02 Exhibit 99.1 earnings releases, FY21Q3–FY26Q2, downloaded to data/pr/
FY2026E revenue and EPS Company guidance midpoint, 4 June 2026
Consensus Alpha Vantage EARNINGS_ESTIMATES, one call, cached to data/av_earnings_estimates_COO.json
Prices, beta, volatility, ADV Alpaca SIP daily bars to 2026-07-28
Options Alpaca v2/options/contracts + v1beta1/options/snapshots (top-level path with explicit symbol list)
Cash taxes, SBC, deferred tax, buyback SEC XBRL IncomeTaxesPaid, ShareBasedCompensation, DeferredIncomeTaxExpenseBenefit, PaymentsForRepurchaseOfCommonStock

3. Key assumptions, and which are sourced vs assumed

SOURCED: - FY2026E revenue $4,303.5m and non-GAAP EPS $4.62 — company guidance midpoints. - FY2026E CVI $2,895.5m / CSI $1,408.0m — company segment guidance midpoints. - Cash tax rate anchor 17.7% — FY2025 cash taxes paid $100.0m / pre-tax $566.5m. - Beta 0.558 raw — 252 daily log returns vs SPY. - Cost of debt 4.02% — FY2025 interest expense over average debt.

ASSUMED (flagged as such in the workbook's source-notes block): - FY2027E–FY2031E revenue growth 4.5% fading to 3.5%. Anchored on management's own statement that it expects contact-lens market growth at "the low end of the historical 4% to 6% range." - Gross margin 67.5% → 68.0%. FY2025A was 65.5% but was depressed by Q4 reorganisation costs in COGS (Q4 GAAP GM 61% vs non-GAAP 66%); H1 FY2026 actual is 67.9%. - SG&A 37.6% → 36.5% of revenue ex one-offs. H1 FY2026 actual ex-litigation is 37.75%. - Capex 8.3% of revenue in FY2026E falling to 5.8% by FY2031E. The peak was 10.8% in FY2024A; H1 FY2026 is 8.96%. This assumption carries a large share of the FCF inflection and is the model's most consequential unsourced input after the exit multiple. - Buyback $300m FY2026E then $450m/yr. $860.8m authorisation remains; FY2025A was $290.1m. - Amortisation of acquired intangibles runs down 190.6 → 125.0 as the acquisition cohort ages. - Working capital: forecast balance-sheet items are driven off revenue at FY2025A ratios (AR 20.26%, inventory 20.67%, other current assets 7.84%, AP 7.34%, accrued/other 19.85%), and the cash-flow working-capital line is derived from the balance sheet rather than assumed independently — which is what makes the balance check hold to 0.2 in every forecast year.

4. Base-rate check on every scenario (required)

Reference class: US-listed medical device companies with revenue above $3bn. Chan, Karceski & Lakonishok (2003) is the anchor: growth persistence beyond chance is close to non-existent, and the modal modelling error is extrapolating the recent rate.

Scenario FY2027–31E revenue CAGR Base-rate verdict
Bear 1.8% Below the 4–6% historical contact-lens market range. Requires share loss or persistent Asia-Pac weakness. This is not a tail outcome — it is roughly a median outcome for a company whose own organic guidance has been cut from 6–8% to 3.5–4.5% in six quarters
Base 4.0% In line with management's own stated market growth and assumes Cooper holds share. The modal outcome, not an optimistic one
Bull 5.8% Requires ~180bp of share gain above Cooper's own view of the market, sustained for five years, at $4.3bn scale. Roughly a top-quartile outcome. Named explicitly as above the base rate and weighted 25% accordingly

Margin base rate — the assumption doing the most work. The Base case embeds +200bp of adjusted operating margin over five years (25.6% FY2025A → 27.0% FY2026E → 29.4% FY2031E). Sustained margin expansion at scale is less common than revenue growth persistence, not more. It is supported by one strong observed datapoint — H1 FY2026 non-GAAP operating margin +220bp year on year — but that datapoint is a reorganisation benefit that laps in Q4 FY2026. The model therefore takes +140bp in FY2026E and then only ~+60bp cumulatively over the following five years, which is deliberately conservative relative to extrapolating H1.

No regime-change / two-tranche modelling is required here. references/regime-change-test.md applies where a contracted tranche runs for a defined term and must not be extrapolated beyond it. Cooper has no contracted revenue tranche, no backlog disclosure, no RPO, and no long-term supply agreements of that character. The one duration-sensitive item — the reorganisation margin benefit — is handled by the lapping treatment above rather than by a two-tranche split.

5. Known model limitations

  1. MiSight is not modelled as a separate line, because Cooper does not disclose it in any filing. It is inside "Toric and multifocal". The TAM work in 01_Company_Research.md §4 is built outside the model from a $32m quarterly datapoint given verbally on one call.
  2. CooperSurgical separation is not in the base model. It is handled as a scenario input via the exit multiple and separately as a sum-of-the-parts in 03_Valuation_Analysis.md §3. Modelling it inside the three statements would require assuming a price, a structure and a date, none of which is knowable.
  3. The FY2026 litigation accrual and its FY2027 cash settlement are carried in accrued liabilities and flow through the derived working-capital line. They net to approximately zero across the DCF horizon, but they distort FY2026E and FY2027E free cash flow individually (+$204m / −$204m) and the reader should not read either year's FCF in isolation.
  4. Tariffs are not modelled explicitly — they sit inside the gross-margin assumption. Cooper has not quantified them.
  5. FX is not modelled explicitly. Roughly half of revenue is non-US. FY26Q2 reported growth was 8% with organic at 4% — a 4pp FX tailwind. The forecast is effectively a constant-currency forecast, which means reported revenue will differ from the model whenever the dollar moves. This is stated rather than dressed up with a currency assumption that would be a guess.
  6. The terminal value is 76.7% of DCF enterprise value. On a 4%-growth business at a 7.46% WACC this is arithmetically unavoidable and is the reason the DCF is presented as one input among several rather than as the answer.