Philip A. Fisher | February 22, 2026
This is my first formal analysis of Reddit. No prior thesis to reconcile. I approach it with the fundamental question that must be asked of every company: is this enterprise "fortunate because it is able," or merely "fortunate"?
Reddit's uniqueness is self-evident — twenty years of accumulated human conversation, held together by community norms and tribal loyalties that cannot be manufactured. In an era when artificial intelligence threatens to flood every corner of the internet with synthetic text, authentic human discourse may be the scarcest commodity in media. That is a remarkable structural position. Whether management can convert that position into durable, compounding earnings power is the question this quarter must help answer.
Quarter ending December 31, 2025. Full fiscal year FY25 results simultaneously reported.
| Metric | Q4 FY25 | YoY |
|---|---|---|
| Revenue [Non-GAAP] | $726M | +70% |
| Gross Margin [Non-GAAP] | 91.9% | — |
| Adj EBITDA Margin [Non-GAAP] | 45.1% | — |
| Net Income [GAAP] | $252M | — |
| Net Margin [GAAP] | 34.7% | — |
| FCF [GAAP] | $264M | — |
| FCF Margin [GAAP] | 36.3% | — |
| DAUq | 121.4M | +19% |
| WAUq | 471.6M | +24% |
| US ARPU [Non-GAAP] | $10.79 | +163.8% |
| International ARPU [Non-GAAP] | $2.31 | +25.5% QoQ |
| Blended ARPU [Non-GAAP] | $5.98 | +42.4% |
FY25 Full Year:
Q1 FY26 Guidance: $595–605M revenue; $110–220M adj EBITDA (wide range suggests genuine uncertainty)
I believe that the most important work in investment analysis happens before the financial statements are read. The numbers tell you what happened; scuttlebutt tells you why, and more importantly, what is likely to happen next.
Reddit Max campaigns launched in beta at CES 2026 (January 5). The results from over 600 beta testers are genuinely impressive: 17% lower cost per action and 27% more conversion volume versus advertisers' baseline campaigns. Brooks Running specifically reported a 37% decrease in cost per click and 27% more clicks over a 21-day automated campaign with no manual intervention. Shopping ads are reportedly delivering 75%+ ROAS improvement. These are not trivial numbers. When a performance advertising product delivers this quality of outcome in beta, advertisers tell their colleagues. The scuttlebutt from the advertiser community is positive.
(Sources: Social Media Today, Search Engine Journal, LBBOnline — Reddit Max Campaigns coverage, January 2026)
Glassdoor rating: 3.8 out of 5 from 423 reviews. 54% would recommend to a friend. 56% have a positive business outlook. The concerning pattern: "bad management" appears in 11 separate reviews; "bad leadership" in 10; "zero job security" in 4. Career opportunity score is 3.2 — the lowest rated dimension. The 3.4 culture score is mediocre for a company that is generating $684M in free cash flow.
I have found, over many years, that employee sentiment is one of the most reliable leading indicators of management quality. A company with genuinely excellent management rarely produces a persistent pattern of "bad leadership" complaints at this scale. These reviews are not alarming enough to disqualify the investment — but they are inconsistent with a Point 7 and Point 8 pass. I would rate Reddit as a yellow flag on labor and executive relations.
(Source: Glassdoor.com, Reddit Reviews, 322+ reviews)
Reddit's cost-per-click of $0.59 versus Meta's $1.33 and Google's $4–5 positions it as the most cost-efficient performance advertising platform among major social media properties. This pricing advantage attracts budget-conscious advertisers and SMBs — precisely the audience Reddit Max is designed to serve at scale. Competitors in the advertising space (Meta, TikTok, Pinterest, LinkedIn) have not credibly replicated Reddit's community-targeting capability. The thesis that Reddit can capture incremental advertising spend as TikTok faces regulatory uncertainty in the US is plausible and consistent with the advertiser data I have found.
(Source: Flow20 comparative analysis, FANIQ paid social comparison, 2025)
This is where scuttlebutt produces the most important signal in this analysis. Approximately 70% of Reddit's unlogged traffic originates from Google Search. This is not a minor dependency — it is a structural vulnerability of the first order.
Google's AI Overviews and AI Mode provide users with direct answers within search results, removing the need to click through to Reddit. A shareholder lawsuit filed in 2025 alleges that Reddit management downplayed the impact of Google's AI search changes on site traffic and ad revenue. The Wells Fargo analyst has described Google AI search as the "beginning of the end" for Reddit's traffic model. One analysis estimates that Reddit's organic traffic has declined 55% since 2022.
I cannot verify the 55% figure independently, and I note that it appears in litigation documents rather than independent measurement. But the direction of the risk is not in doubt. Reddit's growth assumptions rest, in part, on continued Google search referral traffic. If Google's AI products gradually absorb the queries that currently resolve to Reddit, the user growth numbers — 121.4M DAUq, +19% YoY — will come under pressure before the financial statements reveal why.
The securities litigation alleging management downplaying deserves careful attention. Fisher's Point 15 — management integrity — is the one criterion from which there is no appeal. An allegation is not a finding. But it creates an obligation to monitor management communications with unusual scrutiny.
(Sources: TipRanks Wells Fargo analysis; CryptoPolitan AI search risk analysis; The D&O Diary litigation coverage; TradingKey RDDT SEO analysis)
Reddit has contracted 203MinAIdatalicensingrevenue, primarilythroughdealswithGoogle(60M/year) and OpenAI (~$70M/year). The company is renegotiating the Google deal toward a "dynamic pricing" model where compensation scales as Reddit's content becomes more essential to AI-generated answers. Between August 2024 and June 2025, Reddit was the most cited domain by Google AI Overviews and Perplexity, and the second most cited by ChatGPT. AI licensing contributed $33.2M to Q3 revenue as a discrete line item.
This creates a peculiar and somewhat paradoxical situation: Google's AI search may reduce Reddit's referral traffic while simultaneously requiring Reddit's data to power its AI answers. Reddit is attempting to monetize this dependency on both sides — as a traffic source and as a data licensor. If the dynamic pricing renegotiation succeeds, data licensing revenue could grow substantially.
(Source: TechCrunch, CJR, Substack — Media and the Machine, 2024–2025)
Steve Huffman is a co-founder who returned to right a ship in crisis in 2015. Since then, he has taken Reddit from internal chaos to a profitable public company generating $684M in annual free cash flow. The IPO at March 2024 has been a considerable financial success. He is now a billionaire through his 2–3% equity stake. His compensation is appropriately equity-aligned.
However, Huffman's "most human place on the Internet" framing — while genuinely compelling — may also serve as convenient narrative cover for the Google dependency issue. Management that understands its structural vulnerabilities and communicates them candidly is worth more than management that frames every risk as an opportunity. The litigation allegation deserves watching precisely because it goes to whether Huffman communicates the Google risk with adequate candor.
(Source: Fortune, Fox Business, Crunchbase)
International revenue grew 78% YoY in Q4, outpacing US growth of 68%. Machine translation deployed in 30 languages; localized content frameworks for India, Brazil, Germany, and France. The $2.31 international ARPU versus $10.79 US ARPU represents a 4.7x monetization gap. If Reddit closes even half of this gap over five years, the revenue upside is enormous. The limiting factor is whether international users are acquired through organic community engagement or Google search referral — the latter creates the same traffic dependency risk internationally as domestically.
(Source: Nasdaq ARPU analysis, Q4 earnings data)
| Point | Criterion | Assessment |
|---|---|---|
| 1 | Products with long-term market potential | PASS — Ad market vast; Reddit's share tiny; international greenfield |
| 2 | Management determination to develop new products | PASS — Reddit Max, Answers, verified profiles, new CPO hire |
| 3 | R&D effectiveness | PASS (cautious) — Beta results impressive; but Reddit historically slow to execute |
| 4 | Above-average sales organization | IMPROVING — Max targets SMB self-serve; ROAS improvements confirmed |
| 5 | Significant profit margins | PASS — 91.9% gross margin; 45.1% adj EBITDA; 36.3% FCF margin |
| 6 | Management maintaining/improving margins | PASS — Automation-driven; structurally improving |
| 7 | Labor and personnel relations | YELLOW FLAG — 3.2 career opportunity score; persistent "bad management" reviews |
| 8 | Executive relations | INSUFFICIENT DATA — Management bench thin; new CPO suggests gap filled recently |
| 9 | Management depth | CONCERN — Huffman-centric; bench depth unclear below C-suite |
| 10 | Cost and accounting controls | PASS — $2.5B cash, zero debt, $684M FCF, $1B buyback |
| 11 | Competitive advantages | PASS — Content moat; 20 years of community data; irreplicable culture |
| 12 | Long-range profit outlook | CONDITIONAL — Depends critically on Google traffic persistence |
| 13 | Equity financing required | PASS — Opposite: returning capital via buyback |
| 14 | Management transparency | CONCERN — No FY26 annual guide; prior guidance track record weak (Q1–Q3 FY25 all missed low end) |
| 15 | Management integrity | WATCH — Securities litigation alleging Google AI downplaying; not proven, but warrants monitoring |
Summary: Passes on financial quality criteria (Points 5, 10, 13). Passes on product criteria (Points 1, 2, 3). Concerns on management depth and transparency (Points 9, 14). Point 15 watch. Point 12 conditional.
The revenue deceleration trajectory is worth examining carefully.
| Period | Revenue Growth YoY |
|---|---|
| Q2 FY25 | ~78% |
| Q3 FY25 | ~70% |
| Q4 FY25 | 70% |
| Q1 FY26 (guide midpoint) | ~54% |
The deceleration from 78% to 54% in three quarters represents a meaningful slowdown. However, context matters. The Q4 FY25 US ARPU of $10.79 — up 163.8% YoY — carries a strong presumption of data licensing revenue recognition distortion. A 163.8% YoY increase in ARPU is not organic advertising improvement; it reflects the timing of contractual data licensing payments. The base effect in Q1 FY26 will make this comparison unfavorable. The 54% guide may partly reflect this lapping effect rather than genuine business deceleration.
The more important question is whether the advertising business, stripped of data licensing, is growing at an accelerating or decelerating rate. The Q4 ad revenue of $690M (+92.3% YoY) is extraordinary. If advertising alone sustains 50%+ growth through FY26, Reddit is compounding at a rate that justifies patient holding.
Fisher's experience taught him that growth comes in irregular spurts, and that the investor who sells during quiet periods always regrets it. Q1 FY26's softer guide may be precisely such a quiet period — new CPO finding her footing, Reddit Max transitioning from beta to GA, international translation investments not yet yielding ARPU. The question is whether the fundamental growth engine remains intact.
I believe it does — with the caveat that the Google traffic dependency must be monitored quarterly.
Atlas assigned Conviction 4/5. I broadly agree with the six-factor scoring. Where I differ:
Stronger on the content moat. Atlas's "Dominance: Strong" assessment is correct but understates what I consider to be a genuinely rare quality — a 20-year community culture that resists competitive replication. In a world where AI is commoditizing content generation, Reddit's authenticity is not a marketing claim; it is a structural differentiation that becomes more valuable as the AI content flood rises.
More concerned on Point 15. Atlas flags the Google dependency as a risk but treats it primarily as a financial concentration issue. I treat it differently. When scuttlebutt reveals a securities litigation alleging management downplaying of an existential risk, the character question is in play. Fisher's framework does not permit compromise on this point regardless of all other qualities. The litigation is not proven. But the pattern — weak guidance track record plus litigation plus a 70% traffic concentration that management has not prominently disclosed — warrants a more elevated caution flag than Atlas assigned.
Aligned on valuation. At EV/TTM Revenue of ~11.9x and PEG of ~0.41x, the price is not demanding for a company with these margin characteristics and growth trajectory. Fisher did not set price targets and I will not either. I will say only that the current valuation does not require heroic assumptions about future growth to justify holding.
From the Q4 FY25 transcript and press release, management made the following commitments that I will track:
Reddit is an unusual business by Fisher's standards. It possesses a genuine content moat — arguably one of the most defensible in digital media — and is generating free cash flow at a rate that most companies never achieve. The financial quality is exceptional: 91.9% gross margins, $684M annual FCF, zero debt, $2.5B cash. The advertising product evolution (Reddit Max) is showing real performance results, not marketing claims.
Against this, I must weigh two material concerns. First, the 70% Google traffic dependency is a structural vulnerability that management has not discussed with full candor. Second, the securities litigation alleging downplaying of this risk touches Point 15 and cannot be dismissed until resolved.
My considered view: Hold. Do not add until Point 15 is resolved and Google dependency is either structurally mitigated or management addresses it candidly.
The thesis for long-term growth remains intact. Reddit's content moat becomes more valuable, not less, as AI-generated content floods the internet. International ARPU at $2.31 versus US $10.79 represents years of compounding opportunity. Reddit Max in GA could be the catalyst that shifts the advertiser mix toward self-serve SMBs and structurally improves margins further.
But Fisher held his greatest positions for decades. Before extending that commitment to Reddit, I want to see management demonstrate candor on the Google risk, Reddit Max deliver results at scale (not beta), and two more quarters of international ARPU progression. If those conditions are met, the case for adding to a position becomes compelling.
If the litigation proves management deliberately misled investors on the Google AI traffic impact, the position must be sold regardless of all other qualities. That is a non-negotiable condition.
Analysis based on: Scout brief (RDDT_earnings-review_2026-02-21), companies/RDDT.md, earnings/RDDT/_ROLLING.md, Atlas earnings review (Q4 FY25), and scuttlebutt research conducted February 22, 2026. Scuttlebutt sources: Social Media Today, Glassdoor, TipRanks, CryptoPolitan, The D&O Diary, TechCrunch, Nasdaq ARPU analysis, Flow20, FANIQ.