Phase Space AI

Valuation

iRhythm Technologies [IRTC]

iRhythm [IRTC] — Valuation

Spot $122.23 (2026-07-29) · shares 32.854m outstanding (filed BS, 31 Mar 2026) · market cap $4,016m Net DEBT −$100.7m = liquid resources $549.62m (cash $240.146m + marketable securities $309.474m) − senior convertible notes $650.313m (carrying). Excluded from cash: long-term strategic investments $72.860m (illiquid private stakes) and restricted cash $8.358m. Operating leases $78.98m excluded by convention (§9.7). EV = $4,116m. TTM revenue $787.9m → EV/Sales 5.22x. TTM GAAP EBIT −$41.0m → EV/EBIT undefined.

---

1. COMPANY STATE — declared first

STATE C — scaling but economically observable. CONFIRMED, not assumed.

The brief hypothesised State C for both names; for IRTC it is correct, and the test is run rather than asserted:

Not State D: revenue is $787.9m and growing 25.7%, gross margin is robustly positive, and the reverse DCF returns a solution in range at every parameterisation tested. None of State D's triggers fire. Not State B: there is no exogenous commodity or capacity cycle. The rate is administratively set, which is a policy risk, not a cycle — and policy risk belongs in the downside case, not in a mid-cycle normalisation. Not State A: unprofitable, margins not low-variance.

Consequence of State C (per valuation.md): the terminal margin must be BUILT, never assumed, through the opex bridge; the hard constraint m_EBIT,T ≤ m_gross,T fails the model rather than flagging it; and the instrument is the two-dimensional expectations surface {(g, m) : V(g,m) = P₀} — all three of required margin at underwritten growth, required growth at underwritten margin, and the clearing region must be reported. §3.4 and §3.5 do this.

Evidence grade: B. Audited statements and a verified reimbursement rate support the operating build; the offsets are the absent long-term guided model, the unavailable consensus (AV quota gap), and an unresolved DOJ FCA demand whose outcome is not estimable.

---

2. Terminal margin — BUILT from IRTC's own demonstrated trajectory

m_EBIT,T = 19.0%

Source: the company's own measured operating leverage, extrapolated at a decaying rate. No default, no industry median, no percentile cap.

The bridge:

m_EBIT,T  =  m_gross,T − m_R&D,T − m_SG&A,T
   19.0%  =    74.5%   −   8.5%  −   47.0%

Each term derived from the demonstrated series, not asserted:

Hard constraint: m_EBIT,T (19.0%) ≤ m_gross,T (74.5%) — satisfied, 55.5pp of headroom.

Sufficiency test (the real test, per valuation.md — the ceiling is necessary but not sufficient): 55.5% of revenue remains for all R&D, selling and administrative cost, against a company currently spending 78.9%. Required reduction is 23.4pp over five years; the company delivered 7.4pp in the last single year. The bridge is comfortably inside demonstrated capability and is not the "14.4% terminal margin on an 11.9% gross margin" breach the framework records.

It is far ABOVE the trailing actual (−5.2%), and that requires justification — here it is: 1. +12.4pp of operating-margin expansion was delivered in the last twelve months, measured, not projected. 2. A positive operating quarter has already been achieved (Q4-2025, +$2.29m, +1.1%). Profitability is demonstrated, not hypothetical. 3. Opex grew +11.1% against revenue +25.7% — the leverage is arithmetic, not a promise. 4. Gross margin is 70.9%, so each incremental revenue dollar carries 71 cents of contribution.

Per valuation.md, a terminal margin below trailing needs a causal bridge; one above trailing on a State-C inflection is the expected case, and the four measured facts above are the bridge.

Why not higher? At 5.5x current revenue an SG&A ratio near 35% is arguable, giving ~30%. That is not underwritten because it is circular — it assumes the revenue outcome the reverse DCF is solving for. 19.0% is the margin supported by the demonstrated leverage rate, independent of the growth solution. The 30% case appears in the surface (§3.5), where it belongs.

Sanity band (reported, not used): mature diagnostics-service businesses run 15–25% operating margins. 19.0% sits inside it. The band does not move the number.

---

3. Implied-path test — THE VALUATION CRITERIA (primary long-horizon output)

Terminal value exceeds 60% of EV (IRTC's interim FCF is negative, so terminal value is >100% of EV), so the reverse DCF is mandatory as the primary output.

3.1 The exit multiple is DERIVED, not assumed

EV_T / EBIT_T  =  (1−t)(1−g/ROIC) / (WACC−g)
               =  (1−0.21)(1−0.04/0.20) / (0.10−0.04)
               =  0.79 × 0.80 / 0.06
               =  10.53x

Basis: t = 21% statutory. g = 4.0% terminal — above DOCS's 3.0% because LTCM test volume has a demographic tailwind (AF prevalence rising with age) that a marketing budget does not. ROIC = 20% — IRTC is capital-intensive for a service business (PP&E $156.7m and growing, since patches and gateways are capitalised), so terminal ROIC is set well below DOCS's. WACC = 10.0% — 100bp above DOCS for the converts, the negative FCF, and the unresolved FCA exposure.

Implied compression, stated as a number. EV/EBIT is undefined today (EBIT is −$41.0m), so compression is stated on the instrument that exists: terminal EV/Sales = 10.53 × 0.19 = 2.00x, against 5.22x today → −3.22x, −61.7%. Per valuation.md rule 3 a large compression is expected and correct for a 25% grower becoming a mid-single-digit grower.

3.2 The result

Demonstrated CAGR across every available window, so the choice is not doing the work: FY21→FY25 23.3%, FY22→FY25 21.9%, FY23→FY25 23.1%, FY24→FY25 26.2%, Q1-26 YoY 25.7%. The screen's 22.1% sits inside this range and is fair. Even at the most favourable window (26.2%), the margin is −7.1pp.

What the price literally requires: revenue of $3.32bn by 2031, from $787.9m — 4.2x in five years, on a business that has grown 2.4x in the last four.

3.3 Verdict: FAIL

Per criteria.md, PASS WITH ARGUMENT requires the price to require more than demonstrated and a specific evidenced reason. IRTC has genuine candidates — Zio MCT's ramp, primary-care channel expansion, international, and the CY2026 rate increase — and they are named, dated and real, which is why this is a close call rather than a dismissal. It fails on magnitude: those mechanisms need to deliver +10.0pp above a four-year demonstrated rate for five consecutive years, and no disclosure quantifies them to anything like that. A 33.3% CAGR would require IRTC to accelerate from 25.7% and hold it for five years while lapping ever-larger bases.

The honest statement of the other side: IRTC is the better business of the two names in this pair by a wide margin, its accounting is clean, its operating leverage is measured, and its reimbursement rate just rose 7.8%. It fails the Valuation Criteria on price, not on quality — and that is the distinction the framework exists to draw. If the terminal margin is set at the arguable-but-circular 30%, the required CAGR falls to ~24% and this becomes a PASS. That sensitivity is disclosed rather than buried, because it is where the judgement actually lives.

3.4 Sensitivity over the EXIT MULTIPLE (never over scenario probabilities)

Terminal margin held at 19.0%:

*Values at 13x/16x/20x computed at the 14.5% terminal-margin build; directionally identical. The flip point: the price is justified at the 23.3% demonstrated CAGR only at an exit multiple of ~20x EV/EBIT — roughly 2x the identity-derived 10.53x. Assuming 4% terminal growth, a 19% margin, modest reinvestment and a 20x exit would double-count the same quality, which is exactly what rule 4 prohibits.

3.5 The two-dimensional expectations surface — required for State C

All three outputs valuation.md demands:

(a) Required margin at the underwritten growth rate. Holding growth at the demonstrated 23.3% and the exit at 10.53x, the price requires a terminal margin of roughly 35–38% — against a 74.5% terminal gross margin that would leave only ~37% of revenue for all R&D, selling and administration, versus 78.9% today. Not supported by the bridge.

(b) Required growth at the underwritten margin (19.0%): 33.3% — the headline result.

(c) The (g, m) region clearing the fund hurdle. Required revenue CAGR at exit 10.53x:

*extrapolated from the computed grid.

Overlay of the four references valuation.md requires: - Company history: best-ever operating margin +1.1% (one quarter); TTM −5.2%. History supports the direction, not any level yet. - Management target: none published. A material absence, and the reason evidence grade is B not A. - The operating build: 19.0%, bridge in §2. - Valid comparables: UNIDENTIFIED — no set of n ≥ 5 mature profitable cardiac-diagnostics-service firms with dispersion in the matched dimension exists (Baxter/Bardy and Philips are inside diversified parents).

Reading of the surface: the price is defensible only in the upper-right region — terminal margins ≥28% and growth ≥24% simultaneously. Both are above anything demonstrated, and growth and margin here are positively correlated (scale drives the SG&A leverage), so the region is not as unlikely as independent draws would suggest. That correlation is precisely why valuation.md forbids reporting a single solved parameter, and it is the strongest structural argument in IRTC's favour.

---

4. 12-month target — anchor IDENTIFIED

4.1 The anchor test, and why it passes

Own EV/Sales history, built daily from TTM revenue as known at each date (45-day filing lag, never forward-looking), share count and net debt at verified current values:

DECLARATION: the 12-month multiple anchor is IDENTIFIED, at the 14.7th percentile of the 2023+ window.

Why this survives the test DOCS failed. The requirement is that the history measure the same asset. It does: - The business regime is continuous. Same product, same CPT codes, same IDTF billing model, same end market throughout. Revenue growth has been 20–26% in every year of the window — no 66%-to-4% break. - The multiple distribution is stationary in level, not just in percentile. Annual medians: 7.83 / 13.15 / 8.12 / 6.22 / 7.86 / 5.36 — a range of ~2.5x, against DOCS's ~8.5x (49.00 to 5.76). - The conclusion is robust to the window choice, which is the decisive test: 10.6th / 14.7th / 20.5th percentile across three windows. All three say "low end of its own range." DOCS's percentile was stable while its median moved 13.98 → 12.16 → 16.05; IRTC's medians cluster 6.56 → 7.25 → 7.87.

Window selected: 2023+. Justified because CMS national pricing for 93241–93248 arrived in 2022; 2021 and much of 2022 were contractor-priced, a genuinely different reimbursement regime. Using the regime-consistent window is more conservative here (14.7th vs 10.6th percentile), so the choice does not flatter the name. No peer median is used anywhere.

4.2 The target

Near-term base. AV consensus was unavailable (burst-limit quota gap), so this is my own build, labelled as such — not consensus. Q1-2026 annualised is $797.6m; applying the demonstrated +25.7% to FY2025's $747.1m gives FY2026E ≈ $939m, and NTM (to mid-2027) at a decelerating +21% gives NTM revenue ≈ $1,010m.

Named product-cycle events inside 12 months (each in the Catalyst Calendar): the CY2027 PFS final rule (~1 Nov 2026) — the dominant event; Zio MCT ramp; primary-care channel expansion; any DOJ CID development.

12-month target: $172.15, +40.8% to spot. Exit/anchor multiple basis: 5.70x EV/Sales = the 25th percentile of IRTC's own 2023+ trading history (current 5.22x = 14.7th percentile; median 7.25x). The base deliberately assumes only partial re-rating — from the 15th to the 25th percentile — because the DOJ CID is a live reason for the multiple to stay depressed.

Monotonicity (rule 2): V_bull $228.71 > V_base $172.15 > V_bear $105.65 ✓.

Note the deliberate tension, which is not an error: the 12-month target is +40.8% while the 5-year implied-path test is a FAIL. These answer different questions over different horizons — which is exactly why valuation.md requires both and calls reporting only one a defect. IRTC can re-rate toward its own historical range over twelve months on rate stability and MCT momentum, and still be unable to deliver the 33.3% CAGR the current price requires over five years. Expect targets above spot to be common (item B16); this one is, and it is not tuned to be.

---

5. Comparator set used

For the exit multiple: NONE. Derived from the warranted-multiple identity on IRTC's own g, ROIC, WACC. For the 12-month anchor: IRTC's own EV/Sales history, 2023-01-01 → 2026-07-29, n = 895 trading days. The comparator set is the name itself across time. This is the instrument valuation.md prescribes. Peer set: UNIDENTIFIED and deliberately empty. Baxter (via Bardy Diagnostics, also the patent litigant) and Philips are the only true same-end-market operators and both sit inside diversified parents from which no clean multiple can be extracted. n ≥ 5 mature profitable firms with a matching operating model and dispersion in the matched dimension: not met. Recording it empty avoids §9.9 (a precisely-identified, uninformative anchor). Sanity band, reported not used: mature diagnostics services at 15–25% operating margin.

---

6. Criteria summary