AI-generated
Moatsight's own AI workflow wrote this edition. No person edited it.
- Written Nov 1, 2026 from 30 sources, including the 10-Q filed Oct 30, 2026.
- Argued by a bear and a bull critic.
- Before it was published, a separate AI check traced the key figures the company reported to saved copies of its filings.
Earnings
Loud claims, checked: 1 partly supported · 1 supported · 4 open
| Verdict | Claim | Who and when |
|---|---|---|
| SupportedSupported for logged-out search traffic; the split between AI answers and lower ranking of translated pages is the company's own hypothesis | Google's AI search features are cutting Reddit's trafficIn What changed | Reddit's CEO on the October 29, 2026 earnings call, and press coverage of the results |
| Partly supportedHolds for third-quarter revenue; open for future logged-in growth, which search referrals feed | Losing logged-out visitors matters little, since over 80% of monetization comes from logged-in and app usersIn What changed | Reddit management, shareholder letter of October 29, 2026 and the earnings call that day |
| Open | Reddit may stop licensing its content to Google for AI useIn What changed | The Wall Street Journal, as reported by CNBC on July 22, 2026 |
| OpenThe tracker called its finding provisional | ChatGPT has largely stopped citing RedditIn What changed | Promptwatch data reported by Search Engine Land on August 19, 2026 |
| OpenThe amended complaint was dismissed with leave to amend on August 13, 2026; a second amended complaint was filed on October 23, 2026; Reddit contests the claims | Reddit misled investors about the effect of Google's AI Overviews on its businessIn What changed | Plaintiffs in a securities class action filed in June 2025 |
| Open | A logged-in-only Reddit will shrink its audienceIn What changed | 24/7 Wall St., October 30, 2026 |
A verdict is the Moatsight view of what the filings show, not a finding of fact.
The core#
On revenue, margins and cash, the third quarter of fiscal 2026 was another beat. Revenue rose 53% to $895.0 million, net income rose 74% to $283.3 million, and free cash flow almost doubled to $362.6 million. The beat over the top of the guided range was the smallest since the listing, though: 2.9% on revenue, against 6% to 16% in the nine quarters before. The fourth-quarter guide implies growth slowing to about 40%. Underneath, price did most of the work. Ad prices rose about 34% and impressions about 15%, while US daily users fell 2%. Management now warns that, if October's rate of decline continues, total daily users could fall by a low-double-digit percentage in the fourth quarter. Profit quality is better than the adjusted figures alone suggest, since cash conversion is high and capital expenditure negligible. It is weaker than the reported net margin suggests, because Reddit pays almost no tax yet and stock-based compensation is rising again as a share of revenue. Management has a strong record on its own guidance and a weaker one on its longer promises.
The quarter against guidance and a year ago#
| Metric | Q3 2026 | Q3 2025 | Change | Guidance for Q3 2026 |
|---|---|---|---|---|
| Revenue, USD million | 895.0 | 584.9 | +53% | 860 to 870: beat by 2.9% over the top |
| Advertising revenue, USD million | 847.0 | 549.3 | +54% | not guided |
| Other revenue, USD million | 48.0 | 35.6 | +35% | not guided |
| US revenue, USD million | 718.9 | 479.6 | +50% | not guided |
| International revenue, USD million | 176.1 | 105.3 | +67% | not guided |
| Adjusted EBITDA, USD million | 402.3 | 236.0 | +70% | 385 to 395: beat by 1.8% over the top |
| Net income, USD million | 283.3 | 162.7 | +74% | not guided |
| Diluted EPS, USD per share | 1.40 | 0.80 | +75% | not guided |
| Free cash flow, USD million | 362.6 | 183.1 | +98% | not guided |
| DAUq global, million | 130.3 | 116.0 | +12% | not guided |
| DAUq US, million | 50.7 | 51.6 | −2% | not guided |
| ARPU global, USD | 6.87 | 5.04 | +36% | not guided |
Revenue growth slowed for the third quarter running: 69.7% in the fourth quarter of fiscal 2025, 69.1% in the first quarter of 2026, 61.1% in the second and 53.0% in the third (Moatsight calculations from the releases). Global users grew 12% and global ARPU 36%, and the two multiply to the 53% (1.123 × 1.363). In the US the split is starker: users fell 2%, and ARPU rose 53%.
What drove revenue: price, not volume#
The 10-Q splits advertising growth into its two parts. In the third quarter the average price per ad rose about 34% and impressions rose about 15%. Combined, they give 54%, which matches the reported ad revenue growth (Moatsight calculation). Reddit credits the price rise to "improved performance across the full funnel of ad objectives and higher advertiser demand". It credits part of the impression growth to "ad load optimization and improvements, including new ad placements", meaning more ads per user, not more users.
| Period | Ad price, change y/y | Ad impressions, change y/y | Ad revenue, change y/y |
|---|---|---|---|
| Q2 2026 | about +40% | about +17% | +63.9% |
| Q3 2026 | about +34% | about +15% | +54.2% |
Pricing growth fell six points in one quarter. Impression growth stayed in the mid-teens, even with US users falling, only because ad load rose and international users grew. Each ad load increase and each new placement can be made only once, so the impression line has less room left than the user line suggests. The COO said on the October 29, 2026 call that "we grew more by CPMs than by impressions" and that lower-funnel campaigns can carry prices that "are triple digits" per thousand impressions. That is the bull case in one sentence. Reddit's pricing is still catching up with what performance advertisers pay elsewhere. Reddit does not publish its average price per thousand impressions, so the size of the gap cannot be checked.
Testing the drivers against history#
Management's claim: search volatility "is not where our business lives". The CEO said this on the July 30, 2026 call. The company's own numbers allow a partial test. Between the first and third quarters of fiscal 2026, US logged-out daily users fell 9.9%, from 30.3 million to 27.3 million. Over the same two quarters US revenue rose from $525.6 million to $718.9 million, and US ARPU from $9.63 to $13.82. So far, the loss of search visitors has not dented US revenue. That fits the company's statement that logged-in and app users account for "over 80% of the monetization" (third-quarter letter). The test has a limit. The decline only accelerated late in the third quarter, so its revenue effect would show mainly in the fourth quarter and after.
The fourth-quarter guide is the real test. The guide is $1,000 million to $1,030 million, up 38% to 42%. Management warned that total DAUq "could fall in Q4, by a low-double-digit percentage" from 130 million. Suppose the fourth-quarter average lands between 113 million and 117 million, a 10% to 13% fall. Then the midpoint of the guide needs global ARPU of $8.66 to $8.95, up 45% to 50% on the $5.98 of the fourth quarter of fiscal 2025 (Moatsight calculation). In the third quarter ARPU rose 36%. The guide therefore assumes that revenue per user accelerates while the user base shrinks. This is plausible if the users lost are mostly logged-out web visitors who see few ads. It is a stretch if they are not. Management has met or beaten every guide it has given, which argues for the first reading.
How fast licensing really grows. Other revenue grew 35% in the third quarter, up from 15% in the first quarter and 24% in the second (Moatsight calculations from the 10-Q notes). Remaining performance obligations, mostly licensing contracts, fell from $143.7 million at December 31, 2025 to $62.4 million at September 30, 2026. That is what happens when contracts near their end without renewal: the revenue keeps arriving, while the backlog behind it runs off. The growth of the line says nothing about 2027.
What the headline hides#
Adjusted EBITDA leaves out the largest cost after people's cash pay. Stock-based compensation was $131.3 million in the third quarter, up 57% from $83.5 million, and 14.7% of revenue. With related payroll taxes it was $136.4 million, the figure the adjusted EBITDA reconciliation adds back. That was up 46%, slower than revenue, and down from 16.0% to 15.2% of revenue year on year (shareholder letter). The trend within 2026 runs the other way, though: 11.9% of revenue in the first quarter, 13.3% in the second and 15.2% in the third, as grants made at higher share prices vest. The CFO lowered his 2026 outlook from the high teens to "the low-to-mid-teens" on the July 30, 2026 call.
| Quarter | Revenue growth y/y | Adjusted EBITDA margin | GAAP operating margin | SBC, share of revenue | Free cash flow margin | Free cash flow after SBC, margin | Incremental adjusted EBITDA margin y/y |
|---|---|---|---|---|---|---|---|
| Q4 2025 | 69.7% | 45.1% | 31.9% | 11.7% | 36.3% | 24.6% | 58.0% |
| Q1 2026 | 69.1% | 40.1% | 27.6% | 10.3% | 46.9% | 36.6% | 55.6% |
| Q2 2026 | 61.1% | 42.6% | 28.8% | 12.5% | 32.4% | 19.9% | 57.7% |
| Q3 2026 | 53.0% | 44.9% | 29.2% | 14.7% | 40.5% | 25.8% | 53.6% |
Over the four quarters to September 2026, free cash flow was $1,198.1 million (38.8% of revenue). After deducting stock-based compensation of $385.8 million it was $812.3 million, or 26.3%. That second figure is the one our valuation uses. The incremental margin, the share of each new revenue dollar that reached adjusted EBITDA, fell to 53.6% from 55% to 58% in the three quarters before. Costs are now growing closer to revenue. The shareholder letter shows adjusted costs up 41% against revenue up 53%, a ratio of 1.3 times. In the second quarter the ratio was 1.6 times. Sales and marketing rose 63%, led by "user and brand marketing expenses" (10-Q), as Reddit bought users to make up for lost search traffic.
Net income carries almost no tax. Pre-tax income was $289.0 million and tax $5.8 million, a rate of 2.0%. Reddit has a full valuation allowance against its US deferred tax assets, including federal net operating losses of $1.7 billion at the end of 2025 (10-K). At a 21% tax rate, third-quarter net income would have been about $228 million and diluted EPS about $1.13 per share instead of $1.40 (Moatsight calculation). Interest on the $3.03 billion of cash and securities added $26.8 million, 9.3% of pre-tax income.
How to read this: The 10-Q says there is "a reasonable possibility" that Reddit releases "a significant portion" of the valuation allowance within the next 12 months, which "would result in a significant income tax benefit in the period". The allowance stood at $779.0 million at December 31, 2025. A release would lift GAAP net income and EPS sharply in one quarter. It would bring no cash, and it would be followed by a normal tax charge in later quarters. Reported EPS growth around that quarter will not reflect the business.
Cash conversion is genuinely high. Capital expenditure was 0.1% of revenue, and free cash flow over the four quarters was 121% of net income. The gap is mainly stock-based compensation, which is a cost but not a cash cost. Receivables rose $139.3 million in the first nine months of fiscal 2026, the main working-capital drag as the business grows.
The share count is flat, not falling. Fully diluted shares were 206.3 million at September 30, 2026, against 206.1 million a year earlier. Without the third quarter's buyback they would have been 207.0 million (shareholder letter). Before all of 2026's buybacks, gross dilution in the first nine months was 1.2%, the CFO said on the October 29, 2026 call. The buybacks have absorbed it.
What management said that is new#
- Search referrals. The shareholder letter states that referral traffic "was down meaningfully in Q3, with the decline accelerating later in the quarter". In October referrals fell "by more than 1 million referrals on average per week". The CEO said on the October 29, 2026 call that the main cause is Google's "AI Overviews or AI Mode... what started as a feature is now the full search experience", and that machine-translated pages are ranked lower outside the US. Three months earlier, on the July 30 call, he had called search referrals "choppy" and search "volatile". The change of tone is large.
- Logged-in-only Reddit. "This is a test, but we are going to make this transition," the CEO said on the October 29 call. Tests start in select markets outside the US in the fourth quarter. He expects "we will likely be trading reach for engagement".
- Paid marketing. The CFO said on the October 29 call that US web marketing "has proven effective at bringing users to Reddit, but low retention rates have hurt ROI". Spending is shifting to app installs and brand campaigns.
- User reporting. Reddit had said it would stop reporting logged-in and logged-out users after the second quarter (fourth-quarter 2025 letter). It kept reporting them, and the CFO said on the October 29 call that "we'll continue to report logged-in and logged-out DAUq by geography".
- Licensing. "Any deal we do needs to advance our long-term strategy and has to be a fair exchange of value," the CEO said on the October 29 call, adding that the options include "not doing a deal". He gave no timing.
- A new revenue aim. "Our next internal North Star is $5 billion in revenue," the CFO said on the October 29 call.
Management track record#
Guidance against results#
Beat = the reported figure is above the top of the guided range; Met = inside the range, ends included; Miss = below the bottom. We score the first guidance given for each quarter, published in the previous quarter's earnings release. Reddit guides only quarterly revenue and adjusted EBITDA, always as ranges, and has never given full-year guidance.
Guidance, scored
Twenty guided figures in a row above the top of the range (Q2 2024 to Q3 2026)
- Q2 2024:Beat
- Q3 2024:Beat
- Q4 2024:Beat
- Q1 2025:Beat
- Q2 2025:Beat
- Q3 2025:Beat
- Q4 2025:Beat
- Q1 2026:Beat
- Q2 2026:Beat
- Q3 2026:Beat
- Q4 2026:Due
Every revenue and adjusted EBITDA figure Reddit has guided since its March 2024 listing came in above the top of the range, though the margin shrank to 2.9% and 1.8% in Q3 2026.
| Score | Period | Metric | Guidance | Actual | Note |
|---|---|---|---|---|---|
| Beat | Q2 2024 | Revenue, USD million | 240 to 255 | 281.2 | 10.3% above the top |
| Beat | Q2 2024 | Adjusted EBITDA, USD million | 0 to 15 | 39.5 | |
| Beat | Q3 2024 | Revenue, USD million | 290 to 310 | 348.4 | 12.4% above the top |
| Beat | Q3 2024 | Adjusted EBITDA, USD million | 40 to 60 | 94.1 | |
| Beat | Q4 2024 | Revenue, USD million | 385 to 400 | 427.7 | 6.9% above the top |
| Beat | Q4 2024 | Adjusted EBITDA, USD million | 110 to 125 | 154.3 | |
| Beat | Q1 2025 | Revenue, USD million | 360 to 370 | 392.4 | 6.0% above the top |
| Beat | Q1 2025 | Adjusted EBITDA, USD million | 80 to 90 | 115.3 | |
| Beat | Q2 2025 | Revenue, USD million | 410 to 430 | 499.6 | 16.2% above the top |
| Beat | Q2 2025 | Adjusted EBITDA, USD million | 110 to 130 | 166.7 | |
| Beat | Q3 2025 | Revenue, USD million | 535 to 545 | 584.9 | 7.3% above the top |
| Beat | Q3 2025 | Adjusted EBITDA, USD million | 185 to 195 | 236.0 | |
| Beat | Q4 2025 | Revenue, USD million | 655 to 665 | 725.6 | 9.1% above the top |
| Beat | Q4 2025 | Adjusted EBITDA, USD million | 275 to 285 | 327.0 | |
| Beat | Q1 2026 | Revenue, USD million | 595 to 605 | 663.4 | 9.7% above the top |
| Beat | Q1 2026 | Adjusted EBITDA, USD million | 210 to 220 | 266.0 | |
| Beat | Q2 2026 | Revenue, USD million | 715 to 725 | 804.9 | 11.0% above the top |
| Beat | Q2 2026 | Adjusted EBITDA, USD million | 285 to 295 | 342.8 | |
| Beat | Q3 2026 | Revenue, USD million | 860 to 870 | 895.0 | 2.9% above the top, the smallest beat since the listing |
| Beat | Q3 2026 | Adjusted EBITDA, USD million | 385 to 395 | 402.3 | 1.8% above the top |
| Due | Q4 2026 | Revenue, USD million | 1,000 to 1,030 | Midpoint implies about 40% growth | |
| Due | Q4 2026 | Adjusted EBITDA, USD million | 480 to 510 | Midpoint is about 49% of the revenue midpoint |
Source: Reddit earnings releases of May 7, 2024; August 6, 2024; October 29, 2024; February 12, 2025; May 1, 2025; July 31, 2025; October 30, 2025; February 5, 2026; April 30, 2026; July 30, 2026; and October 29, 2026. Percentages are Moatsight calculations.
The count is 20 beats out of 20 scored figures: every quarter since the March 2024 listing, on both metrics, above the top of the range. No guidance was withdrawn or changed within a quarter. The first quarter of fiscal 2024 has no row: the first guidance came with its results. For nine quarters running the revenue beat over the top was 6% to 16%, a margin steady enough to call the guidance deliberately conservative. In the third quarter of fiscal 2026 it shrank to 2.9%. The CFO explained on the October 29, 2026 call that some bookings could not be delivered in the last days of the quarter. A conservative guider whose cushion shrinks to under 3% is giving information, and the fourth-quarter range should be read with less of a cushion than before.
Promise register#
| Status | Promise, and what became of it | First made | Later restatements |
|---|---|---|---|
| Quietly reframed | Grow revenue twice as fast as total adjusted costsThe ratio was 1.3 times in Q3 2026, down from 1.8 times in Q3 2025; recent letters show the ratio without the target | May 2024, Q1 2024 shareholder letter: "Our management target is to grow revenue twice as fast as total adjusted costs" | May 1, 2025 call, the CFO: "The North Star here is revenue growth twice as fast as costs"; not restated since |
| On track | Adjusted EBITDA margin of about 50% in the long term44.9% in Q3 2026; the Q4 2026 guide midpoint implies about 49% | October 30, 2025 call, the CFO: the North Star is "probably in the 50 range" | July 30, 2026 call, the CFO: "The long-term goal is 50%" |
| Open | $5 billion of annual revenue$3.09 billion over the four quarters to September 2026; no date given | October 29, 2026 call, the CFO: "Our next internal North Star is $5 billion in revenue" | |
| Open | 100 million US daily users and one billion globallyUS DAUq fell to 50.7 million in Q3 2026; management warns total DAUq could fall in Q4 | April 30, 2026 call: "Our goal is to reach 100 million daily U.S. users" | July 30, 2026 call, the CEO: "our goal is to achieve 1 billion daily users globally and 100 million in the US"; not repeated on the October 29, 2026 call |
| On track | Dilution of 1% to 3% a year, 2026 at the lower endGross dilution of 1.2% in the first nine months of 2026, 0.1% after buybacks | February 5, 2026 call, the CFO: dilution "at the lower end of our medium-term guide of 1% to 3%" | July 30, 2026 call: full-year dilution at the lower end; October 29, 2026 call: "tracking well" |
| On track | Stock-based compensation in the high teens as a share of 2026 revenue13.6% of revenue in the first nine months of 2026, including related taxes | February 5, 2026 call, the CFO: SBC "in the high teens as a percentage of revenue" | July 30, 2026 call: lowered to "the low-to-mid-teens" |
| On track | $1.0 billion share repurchase programme$340.0 million used by September 30, 2026 at an average of about $155 per share; $660.0 million left | February 5, 2026 earnings release: $1.0 billion authorisation, no expiry | |
| Dropped | Stop reporting logged-in and logged-out users after Q2 2026Still reported in Q3 2026; the 10-Q cites volatile search referrals | February 5, 2026, Q4 2025 shareholder letter: phase-out later in 2026 | October 29, 2026 call, the CFO: "we'll continue to report logged-in and logged-out DAUq by geography" |
| Open | Move Reddit to a logged-in-only experienceTests in select markets outside the US begin in Q4 2026 | October 29, 2026, Q3 2026 shareholder letter: tests in Q4, then "an optimized version of logged-in-only Reddit globally" |
What the ten status words mean
- Met
- Done as promised, by the date given.
- Exceeded
- Done, and more than promised.
- On track
- Not due yet; the latest figures point to it.
- Open
- Not due yet, and the figures so far cannot settle it.
- Slipped
- Not done by the date given, and still pursued.
- Lowered
- Cut openly before its date; the new target is in "What became of it".
- Raised
- Lifted openly before its date; the new target is in "What became of it".
- Quietly reframed
- Changed without being called a change.
- Dropped
- No longer pursued or mentioned.
- Missed
- Due, and not delivered.
Source: Reddit shareholder letters of May 7, 2024, February 5, 2026 and October 29, 2026; earnings releases of February 5 and October 29, 2026; Form 10-Q for the quarter ended September 30, 2026; Reddit earnings calls of May 1, 2025, October 30, 2025, February 5, 2026, April 30, 2026, July 30, 2026 and October 29, 2026.
The register shows a pattern. Financial promises inside management's control (margins, dilution, stock-based compensation, the buyback) are being kept, and some were beaten. Promises that depend on users are slipping out of view. The user goals were dropped from the latest call, and the plan to stop disclosing logged-in users was reversed. The reversal is to Reddit's credit, since it now discloses more at the moment the data is less flattering. The cost discipline target has been allowed to lapse without comment.
The headwind and how management handled it#
The search headwind is real and largely outside Reddit's control. Google changed how its results page shows answers, and the company's own disclosures show the effect on logged-out traffic. Management's handling has been mixed. In its favour: it disclosed the October referral trend and a possible double-digit user decline before anyone forced it to, and it kept reporting the logged-in split. Against it: through mid-2026 it described the problem as ordinary volatility. The CEO said on the April 30, 2026 call that a Google algorithm change was "business as usual". It then spent heavily on US web marketing whose users, by the CFO's own account, did not stay. The securities class action over the company's earlier statements on Google is unresolved. The court dismissed the amended complaint with leave to amend on August 13, 2026, and the plaintiffs filed again on October 23, 2026 (10-Q). The claims are allegations, which Reddit is contesting.
Capital allocation#
- Buybacks. Reddit bought back 2,193,461 shares for $340.0 million in the first nine months of fiscal 2026, an average of about $155 per share (Moatsight calculation from the 10-Q). Most of it was spent in the second quarter at an average of $157.57. The October 30, 2026 close of $91.73 is about 41% below that average. The buybacks have so far offset dilution from employee awards, not shrunk the share count, and they were bought at prices that now look high.
- Cash and debt. Cash, cash equivalents and marketable securities were $3.03 billion at September 30, 2026, with no borrowings. A $500 million revolving credit facility is undrawn (10-Q).
- Acquisitions. One small acquisition in July 2024, for $19.9 million (10-K for fiscal 2024). The CEO said on the October 29, 2026 call that buying users through acquisitions is "not on our radar right now".
- Commitments. The new cloud contracts commit $1.46 billion over three years to Amazon Web Services and Google Cloud. On an annual basis that is about 16% of revenue over the four quarters to September 2026 (Moatsight calculation).
Insiders#
The CEO and the COO sell regularly under pre-arranged 10b5-1 trading plans. The CEO exercises options at $25.29 and sells 17,308 to 18,000 shares about every two weeks: for example 18,000 shares at $175.04 to $184.06 on August 14, 2026, and 17,308 shares at $155.33 to $163.72 on October 13, 2026 (Forms 4). His plan adopted on May 20, 2026 covers up to 600,000 shares. The COO sold 60,000 shares on each of September 14 and October 14, 2026, under a plan for up to 780,000 shares. The CFO's only reported transactions in August to October 2026 were shares withheld for tax. We found no insider purchases in the Forms 4 we read. The selling is mechanical and was planned months ahead, so it says little about the outlook. It also means insiders were selling in the same months the company was buying back stock at similar prices. Control is concentrated. According to his Schedule 13D/A of May 6, 2026, the CEO may be deemed to hold about 70.7% of the votes, through a voting agreement with Advance.
How far management's word can be trusted#
On near-term numbers, very far. Twenty guided figures have been beaten, and the cost, margin and dilution commitments have been kept. On the audience, less. Management described the search problem as manageable for longer than the data supported, and its user goals have been quietly set aside. Its current claims (that over 80% of monetization is safe, and that a login wall will trade reach for engagement) are reasonable but unproven. The next two quarters will test them.