Reddit just printed Rule of 115 while the stock trades at RSI 11.

The Numbers

Q4 revenue $726M (+70% YoY), beat guidance by $61M. Adj EBITDA $327M (45% margin). CapEx under $10M for the full year. $684M FCF. $2.5B cash, first-ever $1B buyback authorized. This is a capital-light advertising machine — 91% gross margins, near-zero infrastructure spend, growing 70%.

Stock is at $151, down 40% in a month. Forward P/E 19x on 50%+ expected growth.

None of this selloff is earnings-related. The report itself was +6% afterhours before the broader tech rout took over. SNAP -41% 1M, PINS -30% 1M, META +3% 1M. This is sector rotation, not RDDT fundamentals.

The Hidden Surface: Reddit Answers

Reddit's AI search product went from 1M WAU in Q1 to 7M in Q3 to 15M in Q4. Total search WAU is now 80M, up 30% YoY. Management calls it "entirely unmonetized today" with "enormous market opportunity."

This is high-intent behavior — shopping queries, lower-funnel conversion mindset. Zero revenue today. Not in consensus models. Growing 100%+ quarter-over-quarter.

Reddit Max: The Automation Play

Their Advantage+ equivalent just launched in public beta. Testing shows 17% CPA reduction and 27% more conversion volume. Management plans to use Reddit's 24 billion posts and comments as signal layer for ad optimization.

This is still "quarters out" from new advertiser onboarding at scale. If it works like Meta's automation did, it's a step-function in SMB penetration.

The Advertiser Flywheel

Active advertiser count +75% YoY, growing faster than revenue. Large customers are still at "minority percentage" of their brand portfolios. 11 of top 15 verticals grew 50%+ YoY. SMB revenue doubled.

This is land-and-expand inflecting. ARPU per advertiser has room to run.

The Bear Case

Growth is decelerating: 70% in Q4 to 52-54% Q1 guide. Management sandbags every quarter (Q4 guide was 53-55%, actual 70%), but even adjusting for that, the trajectory is down.

SBC is 18% of revenue ($387M in 2025). Cash P/E is far higher than headline 19x — probably 200x+ on real economic earnings.

Insider selling: CEO, COO, CFO, CTO all sold at $200-250 in January. Stock is now $151. None are buying at 34% lower.

AI licensing is dead — "other revenue" only +8% YoY. Huffman pivoted the narrative to "product partnership" instead of cash extraction. The LLM revenue thesis is over.

And the comp check: PINS trades at 6.8x forward P/E. If a profitable, proven social ad platform trades at single-digit multiples, RDDT at 19x isn't as extreme a dislocation as it looks. The sector is repricing.

The Setup

This is a doorway state: META 2022 (violent recovery after -60% on improving fundamentals) vs. SNAP 2022 (permanent impairment, never recovered).

RDDT's fundamentals look far more like META than SNAP. Rule of 115, 45% EBITDA margin, capital-light model, unmonetized surfaces growing 100%+ Q/Q, expanding advertiser base, $1B buyback as a floor.

But the market hasn't decided yet. RSI 11 is extreme — single-digit percentile historically. $1B buyback is 3.5% of float. Short interest 16.8%. This is either a flush-out before reversion or the beginning of a structural re-rate.

Forward P/E 19x on a 50%+ grower with 45% EBITDA margins and near-zero CapEx is objectively cheap — if the growth holds and if Reddit Answers monetizes and if the sector selloff is temporary.

Those are big ifs. But the setup is: exceptional fundamental quality at panic valuation in a sector rout. The insider selling and lack of buying at lower prices is the yellow flag. The PINS comp at 6.8x is the reality check.

What's needed: Factor regression (how much is RDDT vs. tech beta), competitive ad platform analysis (SNAP/PINS/META growth/take rates), user growth sustainability check, Reddit Answers monetization path modeling.

This isn't a slam dunk. It's a high-quality asset in a sector being repriced. The question is whether 19x is the entry or just the beginning of compression.