ODD · ODDITY Tech Ltd. · Stock research

Current Thesis

Nothing dated has happened since the last note, and the tape drifted anyway: $14.12 on 2026-08-07 to $13.22 on 2026-08-21, with the most recent company announcement still the 2026-07-22 press release scheduling Q2 2026 results for 2026-09-09 before the US open. The compounder story that priced the $63.68 52-week high is finished — Q1 2026 revenue $197.9M (-26% YoY), first orders -50%, gross margin 69.7% versus 74.9%, all reported 2026-06-02 — and the replacement story, normalization of the IL MAKIAGE paid-acquisition engine, still has no evidence attached to it until the September print. What changed in the last two weeks is the valuation gap: at $14.12 the stock traded above four of five published post-Q1 targets; at $13.22 it sits between Truist's $12 and Evercore ISI's $15, so the de-rating toward the Street has largely happened without any new disclosure. The remaining question is binary and dated.

The narrative is dead. The evidence is the structure, not the mood — 79.2% below the $63.68 52-week high, consensus target cut from $17.33 to $10.09 after 2026-06-02, Goldman Sachs to Sell and KeyBanc to Sector Weight in the same window, and a lower-high sequence from ~$17.88 on 2026-07-03 through $14.12 and $13.22. A dead narrative can still produce a violent print reaction; it has not produced a base.

Bullish and bearish views on ODDITY Tech Ltd.

The model's bull view on ODDITY Tech Ltd. (ODD), in brief: Q1 2026 revenue $197.9M, -26% YoY (reported 2026-06-02) landed inside the pre-announced ~-30% cut — the single datapoint the normalization camp anchors on. The bear view: Q2 2026 was guided to -25% to -30% YoY revenue at the Q1 report. Both cases follow in full.

Bull Case

• Q1 2026 revenue $197.9M, -26% YoY (reported 2026-06-02) landed inside the pre-announced ~-30% cut — the single datapoint the normalization camp anchors on.
• Repeat orders -15% against first orders -50% (Q1 2026 disclosure). The installed base held; management attributed the shortfall to one advertising partner's auction dynamics rather than to demand.
• $667.4M in cash, equivalents and investments plus a $350M undrawn credit facility at Q1 2026. Solvency is not the debate at a mid-hundreds-of-millions market capitalisation.
• The issuer is a standing bidder in its own stock: ~6.1M Class A shares repurchased for $82.3M in Q1 2026, with $167.3M of authorization remaining as of 2026-06-02.
• 2026-06-12: $50M face of 0% exchangeable notes due 2030 retired for $35M — roughly 70 cents on the dollar, and it removes part of the 2030 dilution overhang.
• FY2026 adjusted EBITDA is still guided positive despite Q1 adjusted EBITDA of -$7.0M, which requires an H2 swing management has not yet withdrawn.
• Prior-cycle economics were real: FY2025 adjusted EBITDA $163M, Q4 2025 revenue +24% YoY (reported February 2026).
• Methodiq, launched 2025-11-18, was framed on the Q1 2026 call as a ~$25M revenue contributor in 2026 per transcript coverage — small against a $197.9M quarter, but it is the only growth line management has quantified.

Bear Case

• Q2 2026 was guided to -25% to -30% YoY revenue at the Q1 report. The 2026-09-09 print is not scheduled to show growth; only the shape of the decline is contested.
• The -50% first-order hole compounds forward. DTC repeat revenue is a function of prior new-customer cohorts, so a lost H1 2026 acquisition year pressures 2027 repeat revenue even if customer-acquisition cost normalizes immediately.
• Margin went with the cohorts: 69.7% gross margin and a $21.4M net loss in Q1 2026, against 74.9% a year earlier.
• The Street has not underwritten a recovery. Post-Q1 targets: Barclays $8, Morgan Stanley $10, Jefferies $10.25, Truist $12, Evercore ISI $15. A consensus screen retrieved 2026-08-22 showed a 7-analyst average of $11.07, high $16, low $8 — still below the 2026-08-21 close.
• No new information for a month. The last dated company item is 2026-07-22. Price fell from $14.12 to $13.22 across 2026-08-07 to 2026-08-21 on no disclosure at all, which is supply meeting a thin bid.
• The reporting calendar slipped twice in 2026 — Q1 to 2026-06-02 against a prior early-May cadence, Q2 to 2026-09-09 against 2025-08-04 for Q2 2025 — without a disclosed reason.
• The advertising partner has never been named, so the central causal claim in the bull case cannot be verified outside management commentary.

Setup & Price Structure

Reference points as of the 2026-08-21 close of $13.22: 79.2% below the $63.68 52-week high, a three-month price change of +7.3%, RSI(14) at 38.2. The three-month window now opens after the initial de-rating, so that +7.3% describes churn inside a roughly $9.25–$17.88 range rather than recovery. The July sequence is unambiguous: ~$17.88 on 2026-07-03, $14.12 on 2026-08-07, $13.22 on 2026-08-21 — three lower observations into an event.

The $13 area is the last visible shelf between the current close and the $9.25 bounce origin from the post-Q1 washout. RSI at 38.2 is mid-range: no oversold flush, no trend. There is no base here to fail; there is a drift into a pre-market print.

Crowding and positioning observables, stated as filed:

• Earnings gap risk is 18 days out as of 2026-08-22 — 2026-09-09 before the US open, call at 8:30 a.m. ET (announced 2026-07-22). Pre-market release means the reaction is a gap, not an intraday process.
• Each filing states the sale covers tax-withholding obligations, so the pattern is mechanical vesting rather than discretionary distribution — the observable is the recurrence and the RSU vest date, not a sentiment signal.
• The issuer's own bid is the identifiable non-discretionary buyer: $167.3M of buyback authorization remained at 2026-06-02 after $82.3M was spent in Q1 into a quarter with -$7.0M adjusted EBITDA.
• Retail-sentiment coverage has thinned rather than clustered. No ODD headline appeared on the tape between 2026-07-22 and 2026-08-21 in the sources checked; a name 79% off its high with no news flow has no fresh bid to crowd.

Catalyst Calendar (next 30 days)

• 2026-09-09 — Q2 2026 results, released before US market open, conference call 8:30 a.m. ET (date announced 2026-07-22). Resolves whether Q2 revenue landed inside the -25% to -30% guide, the first-order trajectory against Q1's -50%, whether repeat orders decelerated past -15%, the direction of customer-acquisition cost, and whether the FY2026 positive-adjusted-EBITDA guide survives.
• 2026-09-09 — updated cash and buyback disclosure in the same release. Q1 showed $667.4M in cash, equivalents and investments and $167.3M of remaining authorization; the Q2 figures show whether the repurchase pace continued through a loss-making half.
• No other dated company event falls between 2026-08-22 and 2026-09-21 on the investor-relations calendar as of 2026-08-22.
• Outside the window: Q3 2026 results, ~2026-11 (est.), given a calendar that has already moved twice this year.

What Would Change Our Mind

The structure that has to give way first is the $13 shelf — the last horizontal reference between the 2026-08-21 close of $13.22 and the $9.25 area the July bounce started from. A weekly close below $13 abandons it and re-opens that lower zone; a print-driven gap through it on 2026-09-09 would be the same event with a cause attached.

On the other side, the datapoints that would force a rebuild of the frame are all in the 2026-09-09 release: Q2 revenue declining less than the guided -25%, first orders improving to better than -25% YoY from Q1's -50%, customer-acquisition cost explicitly down sequentially, and the FY2026 positive-adjusted-EBITDA guide reaffirmed with an H2 bridge. A weekly close back above the 2026-07-03 high near $17.88 would mark the market pricing that recovery ahead of the evidence.

The condition that would make the name uninteresting rather than wrong: 2026-09-09 comes and goes with revenue inside the guide, no first-order improvement, and the FY guide quietly softened — a slow bleed with no dated event until roughly November.

Correlation Notes

• The stated cause of the collapse is an ad-auction dislocation, not beauty demand. Read-throughs therefore come from large advertising platforms' pricing commentary before they come from beauty peers; a DTC comparable's strong quarter does not falsify the ODD story, and vice versa.
• Foreign private issuer, Israel-domiciled. Reports on 6-K/20-F rather than 10-Q, so quarterly disclosure is thinner than a US domestic filer's and there is no quarterly cash-flow statement of the usual granularity to check the buyback against.
• Dual-class structure: listed Class A versus founder-held Class B super-voting shares. Buybacks shrink the Class A float while leaving voting control unchanged, so float compression is not the same as control change.
• Consumer-discretionary beta is present but secondary. The 2026 drawdown from $63.68 to $13.22 is company-specific — it dates to the pre-announcement and the 2026-06-02 print, not to a sector move — so the name should be expected to trade on its own release schedule rather than with a rotation.