# RDDT Q3 was two stories — ads ripped, user growth is the real question — retained record

> Retained reddit record from reddit:ValueInvesting.

Canonical: https://highsignal.app/data/records/3facab3ae0717f04

[← sources](https://highsignal.app/data)


source: [reddit](https://highsignal.app/data/reddit) (reddit:ValueInvesting)  published 2026-10-31 02:44 ingested 2026-11-01 02:41

[original source ↗](https://www.reddit.com/r/ValueInvesting/comments/1xjyy9y/rddt_q3_was_two_stories_ads_ripped_user_growth_is/)

## retained text



```
long RDDT, so take this with that bias — and yes, posting this on reddit itself feels a little meta, but the numbers are the numbers.

RDDT reported Q3 last night. Revenue came in at $895M vs the $860–870M guide, up 53% year over year, with diluted EPS of $1.40. Ad revenue was $847M, up 54%, and thats the whole ballgame imo — ads are about 95% of the business, and they still grew 54% during the exact period search referrals were supposed to be falling apart. Adjusted EBITDA was $402M at a 44.9% margin, also above the $385–395M guide.

The number everyone here will jump on first is users, and honestly they should. DAUq was 130.3M, flat vs Q2 and up 12% year over year, below the roughly 133M people were expecting. Weekly actives were 515.2M, basically flat vs 514.6M in Q2. In the U.S., DAUq fell 2% year over year to 50.7M, down 2.5M sequentially. Last quarter U.S. DAUq only grew 6% and dipped sequentially, which was the bear case in one line — AI Overviews eat the search funnel, the funnel narrows, growth dies. This quarter that part got worse, not better.

The split inside that is logged-in vs logged-out. Global logged-in DAUq grew 7% to 53.5M and was up sequentially, while logged-out fell from 77.7M to 76.8M. The two moves basically offset. Management said logged-in and app users are the vast majority of engagement and over 80% of monetization, while the logged-out web audience is about 59% of daily users but under a fifth of monetization. That distinction matters, but it doesnt make the U.S. decline go away.

The part I think gets underweighted: ARPU. Global ARPU was $6.87, up 36%. U.S. ARPU was $13.82, up 53% and up from $11.85 in Q2, even as U.S. users fell. International ARPU was $2.43, up 32% from $2.26 in Q2, on international DAUq up 24% to 79.6M and international revenue up 67%. The gap is still about 5.7x vs the U.S. It doesnt need to close all the way to matter, every point of convergence drops almost straight to margin at 91.5% gross margins.

Other revenue, the data licensing piece, was $48M, up 35% and up from $43.3M in Q2. Its only about 5% of revenue and basically concentrated, so I dont hang the thesis on it, and there was no new deal announced — management said licensing is not guaranteed and the Google negotiation is still open. The corpus is worth more because AI needs grounding in actual human conversation, but that tension is real when Google is also the source of the referral decline.

Guidance is where I have to be straight. Q4 was guided to $1,000–1,030M revenue vs roughly $1,050M consensus, and $480–510M adjusted EBITDA vs roughly $512M consensus. And management warned search referral traffic was down meaningfully in Q3, accelerating late in the quarter, down more than 1 million referrals on average per week in October, and that if that pace continues DAUq could fall in Q4 by a low-double-digit percentage from the 130M base. They also said testing of a primarily logged-in experience starts in Q4 in select markets outside the U.S. Q3 cash was $3.03B in cash, equivalents and marketable securities, free cash flow was $363M, and they bought back about $100M of stock, so this isnt a funding story — its a funnel story.

Also worth noting the growth rate stepped down: revenue +53% ends the run of 60%+ quarters, and ad growth has gone 74% in Q1, 64% in Q2, 54% in Q3. Still fast, just not the same slope.

What would change my mind: two quarters in a row of logged-in users stalling, not just logged-out search traffic bouncing around, or ad growth dropping into the 20s while management blames the funnel. One noisy quarter on uniques wouldnt do it, but a guided double-digit DAUq drop is more than noise, so I am watching Q4 logged-in numbers first. Curious how people here read it — if you use reddit daily and never come through google anymore, does the search referral argument even describe how you got here?
```



document fetched 2026-11-01 00:40 · canonical https://reddit.com/r/ValueInvesting/comments/1xjyy9y/rddt_q3_was_two_stories_ads_ripped_user_growth_is
