FIG — Lean bullish · edge 5.5/10

Thesis

The market is pricing a growth deceleration (FY27E +24%) that has not yet appeared in any reported metric (Q2 +48%, NDR 136%, raised FY26 guide). The 2025-26 drawdown was driven by IPO-pop normalization, lockup expiry (Jan 27, 2026), and peak AI-disruption fear — all of which have either lapsed or been partially refuted by AI-credit monetization data. If Figma tracks its raised FY26 guide and shows AI credits are durable consumption revenue, FY27 consensus revises upward; honest math says the price upside comes roughly half from that revision and half from a modest multiple re-rating off a still-de-rated ~6x FY27E sales.

Bull case

Figma tracks or beats its raised FY26 guide, Q3 shows AI-credit revenue is recurring consumption, and consensus FY27 growth gets revised from +24% toward ~+30% (~$1.9B revenue). Corrected arithmetic: at 8-9x FY27E sales on the current $1.82B consensus, equity value is ~$16.2-18.0B, or ~$33-37/share (+30-46%); if revenue is also revised to ~$1.9B, 8x yields ~$35-39. Note this DOES embed a re-rating from ~5.9x to 8x — roughly 35% multiple expansion — so the honest framing is that upside comes from revisions and re-rating together, not revisions alone. A revision-only path at a constant multiple caps out in the high-$20s. GAAP profitability inflection as SBC normalizes supports the re-rating leg, and the Feb 2026 guidance-driven jump demonstrates the stock's sensitivity to exactly this catalyst.

Bear case

The +24% FY27 consensus is foresight, not conservatism: AI-native tools (Lovable, v0, Cursor) plus Canva/Adobe compress seat growth faster than credits can offset, and the 450bps gross-margin compression is the leading edge of structurally lower AI-era margins. AI-credit revenue proves to be a one-time enforcement catch-up that laps in H2'26, making the raised guide a high-water mark. Post-lockup float expansion plus ongoing 10b5-1 selling caps rallies, and at ~107% realized vol any SaaS de-rating leg takes FIG back toward the high-teens (-25-30%). Note the bear case is symmetric with the bull case in magnitude — roughly -25-30% downside vs. +30-46% upside — which is why this is a lean, not a strong, conviction.

What would change our mind

Q3 2026 revenue comes in below the $373M guide low end, or FY26 guidance is cut/held after the August raise
NDR drops below ~125% — would signal the seat-erosion/AI-disruption thesis is arriving in the numbers
Management discloses AI-credit revenue is non-recurring or consumption is declining QoQ
Gross margin falls below ~82% with commentary attributing it to structural AI costs rather than investment timing
FY27 consensus revenue estimates get CUT from $1.82B — that would mean the deceleration fear is being confirmed, not priced

**FIG remains deeply down despite recent recovery.** Over the past year, FIG’s steep drawdown versus SPY supports the thesis that the stock’s de-rating far exceeds the recent estimate upgrades and improving fundamentals.

Since publication vs SPY · closed horizons are frozen

1 weekfinal

+8.0%

+6.6% raw · hit

1 monthfinal

-7.2%

-8.7% raw · miss

3 monthsrunning

-14.9%

-16.1% raw

6 monthsrunning

-14.9%

-16.1% raw

The full research everything the view rests on

Direct answer to the question asked

FIG's ~69% one-year collapse was a mechanical reset from an absurd IPO-pop starting point ($115+ vs. $33 IPO price) compounded by lockup-supply overhang — not fundamental decay: Q2 2026 revenue grew 48% with 136% NDR, and FY26 guidance was raised. Consensus estimates were revised UP over the past three months even as the stock remains ~69% below its year-ago level and ~79% below its 52-week high. At ~$25 the market prices FY27 growth decelerating to +24% versus the +48% just reported; corrected math shows ~30-46% upside if FY27 estimates revise up AND the multiple re-rates modestly — revision alone at a constant multiple delivers only low-20s upside. That is a decent but not asymmetric payoff, keeping this lean-bullish at moderate conviction.

Where the edge would come from

What the price assumes

At $25.42 / ~$12.4B market cap (~$10.7B EV on $1.67B cash — note: share count is implied, not filing-verified), the stock trades at roughly 7x FY26E and 6x FY27E consensus sales. Consensus explicitly models growth halving from +48% to +24% in FY27, treats AI-credit revenue as low-quality, and applies a SaaS-de-rating multiple despite 85% gross margins and 14% FCF margins. The price assumes the AI threat to design seats is real and merely delayed.

Why the market may be wrong

Three specific places: (1) The FY27 +24% deceleration is a modeling convention, not an observed trend — NDR of 136% and a $40M guidance raise argue the slowdown is being over-extrapolated. (2) AI-credit data (>75% of over-limit enterprise users still consuming, >95% retained) suggests consumption revenue is sticky, and management attributed part of the raise to it — the market assigns this little value. (3) The FY25 GAAP loss (-$1.25B) that scared investors was one-time IPO-vesting SBC; H1'26 SBC is only ~$60M with no unrecognized option SBC remaining, so GAAP profitability inflection in FY26-27 is under-appreciated.

The one thing to watch

The variable to watch is not seat growth or AI headlines — it is whether Q3 classifies AI-credit revenue as durable consumption rather than one-time enforcement catch-up. But be precise about the transmission mechanism: FY27 estimate revisions alone, at a constant ~6x forward-sales multiple, only get the stock to the high-$20s; the 30%+ upside case requires the credit-durability confirmation to drive BOTH the revision and a modest re-rating toward 8x. If you believe only in the revision channel, the edge here is much smaller than the headline drawdown suggests.

Key evidence (10)

Q2 2026 (reported Aug 5, 2026): revenue $370M, +48% YoY; NDR 136%; 85% gross margin; 10% non-GAAP op margin; FCF $53.2M (14% margin); cash $1.67B (Figma IR [T1]; Reuters [T2] 2026-08-05)
FY26 revenue guidance raised $40M to $1.463-1.467B (~39% growth at midpoint); Q3 guided $373-375M (Reuters [T2] 2026-08-05)
Estimates revised UP over the trailing 3 months (FY26 EPS $0.28→$0.29, FY27 $0.34→$0.36, WSJ [T2]) — a window during which the stock actually ROSE ~26%; the -69% figure is the 1-year return and no revision data spans that full year, so the correct framing is: estimates rising while the stock remains ~69% below its year-ago level
Crash timeline: Sept 4, 2025 post-IPO earnings beat but stock slumped on lockup overhang and 'priced for perfection' (Reuters [T2]); Class B lockup expired Jan 27, 2026 ([T4]); Feb 19, 2026 strong outlook sparked recovery (Reuters [T2])
AI monetization: credits enforced Mar 2026; >75% of over-limit Org/Enterprise users still consuming in April, >95% retained; Make weekly usage 60% of >$100k-ARR customers vs ~50% prior quarter (Figma IR/Q1'26 call [T1])
FY25 GAAP loss -$1.25B was IPO-vesting SBC; H1'26 SBC only $60.4M, no unrecognized option SBC remaining as of Jun 30, 2026 (10-Q via StockTitan [T4] — filing-level detail, acceptable but verify)
Gross margin compressed ~450bps YoY (85% vs ~89.5%) — AI inference costs are real (implied from reported figures)
Insider sales July 2026 (CTO ~261k shares, GC) were routine 10b5-1; no 2026 secondary offering found ([T4])
Valuation math check (verification): at ~487M implied shares and $1.67B cash, 8-9x FY27E sales of $1.82B implies ~$33-37/share (+30-46%); reaching $45 requires ~11x FY27E sales — genuine multiple expansion; revision-only at a constant ~5.9x multiple implies only high-$20s
Stock +51% off Feb 2026 low, +26% 3m, technicals bullish, RSI 56; but 20d realized vol ~107% — position sizing matters (market data, 2026-08-14)

Catalysts and risks

Catalysts

Q3 2026 earnings (~early Nov 2026): first clean read on AI-credit revenue durability and whether FY26 guide gets raised again
FY27 initial guidance (Feb 2027 earnings) — the direct test of the +24% deceleration assumption
Any disclosure breaking out AI-credit/Make revenue as a separate line
GAAP operating profitability inflection as IPO SBC rolls off (H2'26)
Competitive launches from Adobe (Firefly/Express AI), Canva, or AI-native builders that visibly target Figma's enterprise base

Risks

AI-credit revenue is one-time enforcement catch-up, not durable consumption — removes a key pillar of the raised guide
FY27 deceleration to +24% proves accurate or optimistic; growth stocks de-rate violently when 40% growth becomes 20%
Gross margin compression from AI inference costs continues (already ~450bps YoY)
Diluted share count and RSU overhang not filing-verified; post-lockup float expansion is a persistent supply headwind
EPS consensus is internally inconsistent across sources ($0.06 vs $0.29 FY26) — GAAP/non-GAAP mismatch means earnings-based signals are unreliable
~107% annualized realized volatility: sizing errors are expensive even if the thesis is right
Upside math is less asymmetric than the raw drawdown implies: revision-only upside is ~high-$20s; the 30-46% case requires multiple expansion that the market may not grant

What to watch

Q3 print: AI-credit consumption trend and whether management quantifies it as recurring
NDR trajectory (136% → anything below ~125% is a red flag)
Gross margin: stabilization at ~85% vs. continued compression
FY27 consensus revisions into year-end — the single number that closes or widens the valuation gap
Whether the forward-sales multiple holds ~6x or begins re-rating — needed for the 30%+ upside case, not just the revision
10b5-1 selling cadence and any new secondary/insider block sales as float normalizes
Adobe/Canva AI product traction and any enterprise-win/loss commentary on the earnings call

Market snapshot when the call was made 2026-08-14

Price

Return 1w	+9.2%
Return 1m	+8.6%
Return 3m	+25.6%
Return 6m	+15.4%
Return 1y	-69.0%
From 52-week high	-79.2%
Realized vol, 20d	107.7%
SPY 3m / 1y	+4.8% / +21.0%
QQQ 3m / 1y	+2.4% / +26.2%

Technicals

FIG technicals as of 2026-08-14: last close $25.42; 20d SMA 24.10 (bullish); 50d SMA 21.93; 12d EMA 24.83; 26d EMA 23.89; RSI 14 56.4 (balanced); 20d vol 106.6%; volume x0.5 vs 20d avg

Verifier notes REVISE · opinion revised once

Unsupported claims

Bull case: 'a revision-driven move to 8-9x FY27E sales implies $35-45 (40-75% upside)'. The arithmetic does not support this. Using the opinion's own figures (~487M implied shares, $1.67B cash, FY27E revenue $1.82B): 8-9x FY27E sales = EV of $14.6-16.4B, equity value of $16.2-18.0B, implying ~$33-37 per share (~30-46% upside), not $35-45. Even if FY27 revenue is revised up to ~+30% (~$1.9B), 8-9x yields only ~$35-39. Reaching $45 requires ~11x FY27E sales — multiple heroics the opinion explicitly disclaims.
Answer/key_evidence: 'estimates were revised UP over past 3 months while the stock fell.' The evidence shows the WSJ revision window is the trailing 3 months (FY26 EPS $0.28→$0.29, FY27 $0.34→$0.36), during which the stock actually ROSE +25.6% (3m return). The -69% decline is a 1-year figure; the evidence contains no revision data spanning that full year. The claim conflates two different windows and misstates the stock's direction over the revision window.

Internal inconsistencies

The bull case relies on the multiple re-rating from the current ~5.9x to '8-9x FY27E sales' — that IS multiple expansion of ~35-50% — while most_differentiated_insight states 'the entire valuation gap closes through that revision channel rather than through multiple expansion.' These two statements contradict each other: the stated price path in the bull case requires the very multiple expansion the differentiated insight disclaims.

Corrections applied

Fix the bull-case math: either restate the target as ~$33-37 at 8-9x FY27E sales (~30-46% upside), or explicitly acknowledge that reaching $45 requires ~11x FY27E sales (i.e., genuine multiple expansion) and drop the 'without any multiple heroics' framing.
Reconcile the revision-vs-multiple narrative: either the gap closes via FY27 estimate revisions at a roughly constant multiple (which yields a smaller price move than the bull case states), or via a re-rating to 8-9x (which is multiple expansion) — pick one mechanism and make the bull case and most_differentiated_insight consistent.
Correct the revision-window claim to: 'consensus estimates were revised UP over the past 3 months even though the stock remains ~69% below its year-ago level / ~79% below its 52-week high' — avoiding the false implication that the stock fell during the 3-month revision window, when it actually rose ~26%.

Sources (23 · T1 = primary filings)

Earlier and later calls on FIG

Sep 25, 2026Business strong; stock a speculative buy.Lean bullish
Sep 15, 2026Yes — but only in small size.Lean bullish
Sep 8, 2026Yes — massively, but not cheaply.No clear edgeview changed
Aug 23, 2026No clear edge at $27.No clear edgeview changed

Multiple compression, not deterioration.