Skip to content

Dossier · SOPH · Dormant

SOPH · SOPHiA GENETICS SA · Stock research

Last analysed ·

Current thesis

Cloud-genomics platform narrative inflected on the 2026-08-04 Q2 print: net dollar retention 117% vs 107% YoY, FY26 guide raised to $94–96M, and a two-program AstraZeneca companion-diagnostics deal. Price has already run 77.3% in three months into a 52-week high at $8.65 with RSI 82.3, and the next scheduled disclosure is ~10 weeks out.

Invalidation trigger

A weekly close below $6.84 (the 2026-08-04 post-print session close) hands back the entire guidance-raise gap; secondary condition is a Q3 print near 2026-11-03 (est.) that leaves the $94–96M FY2026 guide unchanged or lower, or net dollar retention back under 110%.

Thesis status

Open commitment scored if the trigger above fires How this is scored →

Latest analysis and events for SOPH —

As of 2026-08-23, orbyd's latest analysis for SOPHiA GENETICS SA (SOPH): Cloud-genomics platform narrative inflected on the 2026-08-04 Q2 print: net dollar retention 117% vs 107% YoY, FY26 guide raised to $94–96M, and a two-program AstraZeneca companion-diagnostics deal. Price has already run 77.3% in three months into a 52-week high at $8.65 with RSI 82.3, and the next scheduled disclosure is ~10 weeks out.

Invalidation trigger: A weekly close below $6.84 (the 2026-08-04 post-print session close) hands back the entire guidance-raise gap; secondary condition is a Q3 print near 2026-11-03 (est.) that leaves the $94–96M FY2026 guide unchanged or lower, or net dollar retention back under 110%.

.# SOPH — SOPHiA GENETICS SA

Current Thesis

The leg being bought is that SOPHiA's cloud genomics platform has crossed from "grinding land-and-expand" into visible operating leverage, and that pharma is now paying for it. Two datapoints from the 2026-08-04 Q2 print carry the story: net dollar retention of 117% against 107% a year earlier, and a global collaboration with AstraZeneca covering two companion diagnostic programs (a decentralized solid-tumor CDx and a hematological oncology test). Revenue of $23.3M came in +27% reported / +25% constant currency against a $21.73M consensus, and full-year 2026 guidance moved up to $94–96M from $92–94M (22–24% growth). The market has repriced fast: the shares are up 77.3% over three months and the 2026-08-21 close of $8.65 is the 52-week high, with RSI(14) at 82.3.

The tension is that the same print missed on the bottom line — EPS of -$0.30 against a -$0.20 estimate — and the equity that funded the balance sheet was sold two months ago at $4.75.

Bullish and bearish views on SOPHiA GENETICS SA

The model's bull view on SOPHiA GENETICS SA (SOPH), in brief: Retention inflected, not just grew. Net dollar retention 117% in Q2 2026 versus 107% in Q2 2025 (2026-08-04 release). Existing accounts are expanding volume, which is what turns a customer-count business into a compounding one. Installed base still adding. 542 core genomics… The bear view: The beat was on revenue only. EPS of -$0.30 missed the -$0.20 estimate (2026-08-04). Net loss was $22.4M, roughly flat YoY, on $23.3M of revenue — the company still loses close to a dollar for every dollar it books. Unit economics went the wrong way. Adjusted gross margin of… Both cases follow in full.

Bull Case

  • Retention inflected, not just grew. Net dollar retention 117% in Q2 2026 versus 107% in Q2 2025 (2026-08-04 release). Existing accounts are expanding volume, which is what turns a customer-count business into a compounding one.
  • Installed base still adding. 542 core genomics customers versus 490 a year earlier, with 24 new customers signed in Q2 2026; SOPHiA DDM analyses of 115,000 in the quarter, +22% YoY (2026-08-04).
  • Guidance moved up, not merely reaffirmed. FY2026 revenue guide raised to $94–96M from $92–94M against a $93.403M consensus (2026-08-04) — a raise off a beat rather than a reallocation of a prior range.
  • Pharma revenue is arriving, with a named counterparty. The AstraZeneca CDx collaboration announced 2026-08-04, alongside Kartos and other biopharma projects management cited as beginning to generate revenue, moves mix toward higher-value milestone work.
  • Loss narrowing on the adjusted line. Adjusted EBITDA loss of $8.8M, a 27% YoY improvement, with management reiterating it expects to approach adjusted EBITDA breakeven by end-2026 and cross to positive in H2 2027 (2026-08-04).
  • Funded through the stated crossover. Cash and cash equivalents of $107.7M at 2026-06-30 against an FY2026 adjusted EBITDA loss guide of $29–32M.
  • Sell-side moved the same week. HC Wainwright raised its target to $14 and BTIG to $9, both Buy-rated, both dated 2026-08-05.

Bear Case

  • The beat was on revenue only. EPS of -$0.30 missed the -$0.20 estimate (2026-08-04). Net loss was $22.4M, roughly flat YoY, on $23.3M of revenue — the company still loses close to a dollar for every dollar it books.
  • Unit economics went the wrong way. Adjusted gross margin of 72.1% versus 74.4% a year earlier, attributed to cloud costs, services revenue mix and early-stage project ramps. Scaling volume compressed margin rather than expanded it.
  • The adjusted line flatters. The $8.8M adjusted EBITDA loss and the $22.4M net loss differ by more than $13M of items excluded from the adjusted measure, including share-based compensation.
  • the Q2 release describes approximately $57.5M gross, indicating the option was taken up. That stock was distributed at roughly half the 2026-08-21 close of $8.65.
  • FX helped. Reported growth of 27% against 25% in constant currency means about two points of the headline came from currency.
  • The raise was small relative to the repricing. Guidance moved $2M at the midpoint. The share price moved 77.3% in three months.
  • Price has met one published target. The 2026-08-21 close of $8.65 sits just under BTIG's $9, leaving HC Wainwright's $14 as the only materially higher published anchor.

Setup & Price Structure

Narrative life-cycle: ACCELERATING, and the dates support it. The 2026-08-04 print produced a guidance raise, the first AstraZeneca CDx headline, two same-week target raises on 2026-08-05, and a post-print session close Simply Wall St put at $6.84 (+20.2%). Price then continued to a new 52-week high at the 2026-08-21 close of $8.65, roughly two and a half weeks after the catalyst — continuation well past the event day rather than a fade. Coverage remains thin (HC Wainwright and BTIG are the visible marks), which argues against a SATURATED read for now.

Structure is the problem, not the story. At the 2026-08-21 close the shares are 0.0% from the 52-week high with RSI(14) at 82.3, and there is no intermediate consolidation between the 2026-08-04 post-print close of $6.84 and $8.65 — the leg has not built a shelf it can retest. The reference levels that exist are event-defined: $6.84 as the gap floor the entire post-print advance stands on, and $4.75 as the price at which the last block of new supply was created on 2026-06-16.

Crowding and positioning observables, stated as observables: 77.3% appreciation over three months into a 52-week high; RSI(14) 82.3; a completed equity offering 2026-06-16 that expanded the share count immediately before the run; two price-target raises inside 24 hours of the print; and a share price now trading within about 4% of the lower of the two published targets. No insider transaction filings appear in the material reviewed for this note, so nothing is asserted about insider activity either way.

Catalyst Calendar (next 30 days)

  • No company-confirmed event falls inside 2026-08-23 → 2026-09-22. The next scheduled disclosure is the Q3 2026 report, roughly ten weeks out. Any September appearance at a healthcare investor conference would be an unscheduled addition as of this writing.
  • ~2026-11-03 (est.) — Q3 2026 results. Tests whether 117% net dollar retention holds, whether the $94–96M FY2026 range moves again, and whether adjusted gross margin recovers from 72.1%.
  • 2026-12-31 — management's stated marker for "approaching" adjusted EBITDA breakeven, set against the FY2026 adjusted EBITDA loss guide of $29–32M.

Elapsed catalysts

  • Undated — AstraZeneca CDx program progression. The 2026-08-04 announcement disclosed two programs without public milestone dates; any development or validation milestone would arrive unscheduled. (passed 19d ago)

What Would Change Our Mind

The structure that would break first is the 2026-08-04 gap. The whole advance sits on the post-print session close of $6.84 with nothing beneath it until the June offering price of $4.75, so a weekly close below $6.84 would mean the market has handed back everything the guidance raise and the AstraZeneca headline bought, and the ACCELERATING label would no longer be supported by price.

On fundamentals, three observations would change the read independently of price. First, net dollar retention printing back under 110% at the Q3 report would recast the 117% as a one-quarter mix artifact rather than an inflection. Second, adjusted gross margin below 72.1% again would say the cloud-cost and services-mix drag is structural, not a ramp effect. Third, another equity raise inside twelve months of the 2026-06-16 pricing — disclosed via a prospectus supplement or an ATM — would argue the $107.7M cash line does not in fact carry the company to the stated H2 2027 crossover.

The reverse also applies: a Q3 print that raises FY2026 guidance a second time while holding NDR above 115% would extend the leg and pull the sell-side band above HC Wainwright's $14.

Correlation Notes

  • Trades with the small-cap genomics-tools and precision-oncology complex — Tempus AI, Natera, Guardant Health, 10x Genomics, Pacific Biosciences — and inherits their sensitivity to biotech risk appetite; XBI and IWM drawdowns tend to hit this cohort harder than large-cap medtech.
  • Biopharma-partnered diagnostics revenue is milestone-driven and lumpy quarter to quarter, so quarterly revenue lines here correlate less cleanly with volume growth than a pure per-test model would.
  • A Swiss issuer reporting in USD: reported growth (27%) and constant-currency growth (25%) diverged in Q2 2026, so a dollar move shows up directly in the headline versus guidance comparison.
  • Nasdaq-listed ordinary shares of a foreign private issuer — disclosure arrives via 6-K and 20-F rather than 10-Q, which means less interim detail than US-domiciled diagnostics comparables provide.

Notes

  • Foreign private issuer: discloses via Form 6-K and 20-F, not 10-Q/10-K, so interim detail is lighter than US-domiciled diagnostics comparables.
  • Swiss issuer reporting in USD; Q2 2026 reported revenue growth (27%) exceeded constant currency (25%), so FX moves the headline versus guidance.
  • Adjusted EBITDA excludes share-based compensation: Q2 2026 adjusted EBITDA loss was $8.8M against a $22.4M net loss.
  • further shelf issuance remains possible.

Related · shared themes

DELL

Dell Technologies Inc.

Q2 FY27 binary moved forward: a 2026-08-18 company release sets the print for 2026-09-01, four sessions after Nvidia on 08-26. Price closed $434.78 on 2026-08-20, a fourth session under the $440.97 pre-breakout shelf and 12.1% off the 08-13 closing high, while Evercore's $550 (08-19) marked the third above-consensus raise into the break. The margin half stays untested.

LOW

HPE

Hewlett Packard Enterprise Company

Five-session give-back stopped at five: 2026-08-21 closed $53.45 (+1.06%), the first higher close since the 2026-08-13 high close of $59.82. Nothing before the 2026-09-02 Q3 print resolves the $11.5–12.1B / $0.88–0.93 guide, and a screen retrieved 2026-08-21 models $0.94, above its top. Narrative SATURATED — no institutional input dated after 2026-08-14.

LOW

LSTR

Landstar System, Inc.

Supply-driven truckload rate leg still unconfirmed on the weekly tape: FTR's published overview as of 2026-08-23 is still Week 32 (dry van -7c, fourth straight decline, demand index weakest since week three), and ATA's 2026-08-18 release put July tonnage at 113.5, -1.0% m/m and -0.5% y/y. Only contract carries it (DAT van +13c/mi in July). Week 33 print ~2026-08-24 is the next read.

LOW

MRVL

Marvell Technology, Inc.

Custom-silicon ASIC royalty story intact, but the AI-semi theme is mid-correction — chips posted their worst month vs software on record (7/17) with money rotating into energy. Buying a falling sector with no confirmed higher low is the trap; wait for the $250-255 May shelf to hold before re-engaging.

LOW

See also · stocks to watch