RDDT — Q4 FY25 Earnings Review

Bert Hochfeld | TickerTarget / HHHYPERGROWTH Date: 2026-02-25 Quarter: Q4 FY25 (ending December 31, 2025) Prior coverage: None — initiating coverage with this review


Summary

I haven't covered Reddit in the newsletter before, but after reviewing the Q4 FY25 print, I can say this is one of the cleaner earnings reports I've seen in a long time. Reddit delivered $726m of revenue, +70% YoY — that's the sixth consecutive quarter above 60% growth, against a comparable of +71% from Q4 FY24. The company beat its guide by $66m (+10%), maintained the pattern of deliberately undemanding estimates that has averaged an 11.2% upside across all four FY25 quarters, and simultaneously delivered Adj EBITDA of $327m at a 45.1% margin — itself a $42-52m beat vs. the implied guide. The EBITDA beat is the more important signal: at 60% incremental margins, Reddit's cost structure is essentially fixed infrastructure costs against a revenue base that compounds at whatever rate ARPU × users dictates. FCF for the quarter was $264m, a 36.3% margin — the Rule of 115 (70% growth + 45% EBITDA) is rare company at any valuation, but at 10.9x EV/TTM Revenue with zero debt and $2.5B of net cash, the market is undervaluing what Reddit has become. This is a monetization unlock story at an early enough stage that the most interesting chapters are still ahead of it. The thesis is simple: international ARPU at $2.31 vs. US ARPU at $10.79 is a 4.7x gap that will not persist. Closing a fraction of it across 68.9m international DAUq generates more incremental revenue than any single product investment management can make. Initiating with a buy.


Headline KPIs


Layer 1 — Revenue Growth Trajectory

Reddit's revenue growth has been the defining feature of its first two years as a public company. The progression from +48% in Q1 FY24 to +71% in Q4 FY24 to +70% in Q4 FY25 against that very comparable is not a performance that shows up in many companies at this scale. When revenue doubles from ~$1.3B in FY24 to $2.2B in FY25 at that growth rate, you are witnessing a monetization unlock in real time.

The decomposition matters: DAUq grew +19% YoY while revenue grew +70% YoY. The 51-point difference is entirely ARPU-driven. Global ARPU expanded from $4.21 to $5.98 — a +42% YoY improvement. The monetization machine is outrunning the user growth engine by a wide and widening margin, which is the correct sequencing for an ad-supported business that historically charged nothing for access to one of the most searched content repositories on the internet.

The sequential pattern is equally compelling:

Quarter Revenue QoQ ARPU (Global) ARPU YoY
Q1 FY25 $392m -8.3% $3.63
Q2 FY25 $500m +27.4% $4.53
Q3 FY25 $585m +17.0% $5.04
Q4 FY25 $726m +24.1% $5.98 +42%

Q1 FY26 guide of $600m midpoint implies -17.4% QoQ decline — that is the standard Q1 seasonality pattern for ad-supported businesses and entirely consistent with Reddit's own history. Q1 FY25 declined -8.3% QoQ from Q4 FY24's $427.7m, so this year's -17% QoQ decline is slightly steeper in absolute percentage terms, but the business is 65% larger. The $600m Q1 FY26 midpoint implies +53% YoY growth against the $392.4m Q1 FY25 base — deceleration from the 60-70% run rate, but still exceptional at this revenue scale.

The guidance pattern is relevant here. Four consecutive quarters of beats averaging +11.2% above midpoint is a well-established management policy of deliberate understatement. If that pattern holds, Q1 FY26 actual revenue likely lands in the $650-665m range, implying closer to 65-68% YoY growth. That would represent no deceleration at all. Until management demonstrates a quarter where the beat cadence breaks, I treat the guide as a lower bound.


Layer 2 — Margin Analysis

Reddit's margin profile is one of the most attractive I've seen in the ad-supported space. The company runs infrastructure on third-party cloud — no data centers, no capital intensity to speak of (CapEx was $3m in Q4 on $726m of revenue). The incremental cost of serving another impression is essentially zero. That structural reality flows through directly to the financials.

Gross Margin: 91.9% in Q4, 91.2% for FY25 (+70bp YoY). The minor YoY compression vs. Q4 FY24's 92.6% reflects higher cost-of-revenue from infrastructure growth as absolute revenue scaled, not a pricing problem or a unit economics deterioration. Six consecutive quarters above 90% is a benchmark that very few software-adjacent businesses achieve.

EBITDA Margin: The trajectory from 29.4% in Q1 FY25 to 33.4% in Q2 to 40.3% in Q3 to 45.1% in Q4 is a continuous march upward. FY25 full year EBITDA margin of 38.4% compares to 22.9% in FY24 — a 15.5-point expansion in a single year. The mechanism is straightforward: total adjusted costs grew +35% in FY25 while revenue grew +69%, creating 34 points of revenue outperformance over cost growth. At scale, the gap compounds.

SBC as %: A concern in social media businesses generally. Reddit's SBC peaked as a percentage of revenue in FY24 (driven by the IPO-related $595.5m charge in Q1 FY24 alone). In FY25, SBC was $387m, or 17.6% of revenue — down materially from the distorted FY24 figure. Q4 FY25 SBC was $91m (12.5% of revenue), sequentially declining from $93.6m in Q3. Management guided "high teens" SBC as a percentage of revenue for FY26 — similar to FY25. Diluted shares actually declined QoQ with the buyback program in progress. This is heading in the right direction.

Operating Leverage: G&A headcount ended FY25 at the same level as year-end FY23 — zero headcount growth over two years on the back-office side while revenue grew from $249.8m per quarter to $726m. That is operational discipline worth noting. The investment is concentrated in R&D and Sales — both of which produce returns that are measurable in-year (3-6x in the CFO's framing for sales force expansion).


Layer 3 — FCF and Profitability

FY25 FCF of $684m against FY24 FCF of $216m is a 3.2x improvement in a single year. FCF margin of 31.1% for the full year, and 36.3% in Q4, places Reddit firmly in elite company among hypergrowth businesses. Most 70% revenue growers are burning cash or generating negligible FCF; Reddit is doing both at once with zero debt and $2.5B on the balance sheet.

The Rule of 115 (70% growth + 45% EBITDA) is a metric I use to identify companies operating at the intersection of growth and profitability that the market consistently undervalues. In my experience, businesses at Rule of 115 — particularly those that have crossed this threshold recently and are still early in their monetization arc — tend to be re-rated meaningfully over 12-24 months as the FCF trajectory becomes undeniable to generalist investors. Reddit crossed this threshold in Q4 FY25.

The FCF to GAAP Net Income relationship is also notable. GAAP Net Income of $252m vs. FCF of $264m in Q4 means Reddit's GAAP earnings are not heavily distorted by non-cash charges that erode cash generation. The CapEx is essentially nil. The business is generating real cash.

The $1B share repurchase authorization, while representing only ~4% of market cap at current prices, signals management confidence in the sustainability of FCF generation. This is not a company repurchasing shares while burning cash — it is a company generating $684m of FCF in a year and returning capital while maintaining $2.5B of liquidity for organic investment and M&A optionality.


Layer 4 — Leading Indicators

Bullish:

Bearish:


Layer 5 — Management and Guidance Credibility

Guidance cadence is the cleanest signal: Four consecutive quarter beats averaging +11.2% above midpoint, with Q2 FY25 at an extraordinary +19% (the AI licensing pop). The range is consistent: +7.5%, +19%, +8.3%, +10%. Excluding Q2's anomaly, the mean is +8.6%. Management is deliberate about sandbagging — not because they lack visibility (they say 40-50% of orders are written in-quarter, with March the largest month), but because the guide is a floor, not a forecast.

The EBITDA guide is similarly conservative. Q1 FY26 guide of $215m midpoint on $600m revenue implies 35.8% EBITDA margin. If revenue beats to $660m at 40% EBITDA margin (a reasonable assumption given Q4 was 45%), actual EBITDA would be $264m — 23% above guide. The pattern suggests management is also deliberately under-guiding on margins, not just revenue.

Huffman credibility is mixed. On the commercial side — business model, monetization mechanics, product investments — I find him credible. The beat cadence supports it. On community relations, the 2023 API controversy and prior incidents show a willingness to prioritize business interests over community trust. For an ad-supported platform where authenticity is the core value proposition, this is a structural risk worth monitoring. The new CPO (Maria Angelidou-Smith) arriving in Q1 FY26 is a positive — it suggests Huffman recognizes the execution gaps in product velocity.

COO Jennifer Wong is more consistently credible on the specifics of advertiser dynamics, vertical mix, and ad stack progress. Her framing of the large advertiser opportunity ("still scratching the surface," "majority penetration in minority of LOBs and geographies") is consistent with the financial trajectory.

The logged-in/logged-out reporting sunset (final reports in Q2 FY26) is worth monitoring. The rationale given — that the distinction matters less as product treats both user types equally — is plausible. But the timing coincides with a period when management will be managing against harder comparables. Removing a data quality signal when results are peaking invites skepticism. I'll watch the Q2 FY26 reported WAUq/DAUq numbers carefully for any discontinuity.


Valuation Re-mark

Current snapshot (Feb 25, 2026):

Metric Value
Market Cap $26.5B
Net Cash $2.477B
Enterprise Value ~$24.0B
TTM Revenue (FY25) $2.203B
EV/TTM Revenue 10.9x
TTM Adj EBITDA $845m
EV/TTM EBITDA 28.4x
GAAP Net Income TTM $530m
P/E (TTM GAAP) ~50x
Annualized Q4 EPS (GAAP diluted) 4.96(1.24 × 4)
P/E (annualized Q4 run rate) ~26x
PEG (Q4 run-rate P/E / ~53% fwd growth) ~0.50

Cohort comparison: For companies growing revenue at 60-75% YoY with EBITDA margins at 40%+ and demonstrable FCF generation, the market historically prices EV/S in the 15-20x range. AppLovin (APP), the closest comparable on growth rate and margin profile, trades at 20-25x forward EV/S — and AppLovin has substantially higher revenue concentration risk (gaming) and no authenticity moat. Reddit at 10.9x EV/TTM Revenue represents a 30-40% discount to where this growth/margin profile should trade.

The discount has a partial justification: Reddit is 23 months post-IPO with a shorter track record as a public company, its Google search dependency is unquantified, and the FY26 comparison base will require high-40s to mid-50s% growth just to hold absolute revenue growth rates. But none of these factors justify a sustained 35% discount to cohort. A company that generates 36% FCF margins at 70% revenue growth with zero debt and $2.5B net cash trades at 10.9x EV/TTM. That is one of the more compelling mismatches I see in the current market.

NTM framing: If FY26 revenue lands at $3.0-3.3B (implying 36-50% YoY growth — conservative to base case given Q1 guide beat pattern), EV/NTM Revenue is 7.3-8.0x. At 40% EBITDA margins in FY26, that implies EV/NTM EBITDA of 18-20x. For a business that is still growing at 30-50%+ with dominant competitive positioning in its niche, that is cheap.

The ARPU gap as terminal value framing: US ARPU is 10.79/quarter(43.16/year). Meta's US ARPU runs approximately $55/year. Reddit is at 78% of Meta's US monetization rate with a user base that skews more researched and intentional in purchasing behavior. If Reddit closes the gap to 90% of Meta US ARPU over 3-4 years while US DAUq grows even modestly at 8-10%, US revenue alone could approach $2.0-2.5B annually — more than the entire FY25 total. International is additive on top of that, and Reddit Answers + Reddit Max are early options on incremental revenue streams not in any current model.


Recommendation

I am initiating coverage of Reddit with a buy recommendation and a target weight of 5-7% in the high growth portfolio. The investment case rests on three compounding mechanisms:

  1. US ARPU expansion from $10.79 toward Meta parity. Not fully priced. Measurable quarterly via ARPU/DAUq US. If this stalls, the thesis weakens.

  2. International monetization catch-up. $2.31 Intl ARPU vs. $10.79 US is a 4.7x gap that machine translation and ad stack maturation in the UK, Germany, France, and Australia will close over 3-5 years. This is visible in the data: Intl ARPU grew +38% YoY in Q4, faster than the US rate.

  3. Incremental optionality (Reddit Answers, Reddit Max, Data Licensing). None of these are in current consensus models at scale. They don't need to contribute materially to justify the current price.

At 10.9x EV/TTM Revenue for a Rule of 115 company with zero debt, $2.5B net cash, and a four-quarter guide-beat cadence averaging +11.2%, the beer has not gotten any colder. The risk-reward is compelling.

What would change the thesis: A miss on US ARPU (below $10.50 in Q1 FY26, given the seasonal Q1 dip), sustained deceleration in active advertiser count growth below 50% YoY, or evidence that lower-funnel revenue is reverting toward mid-funnel rates without new offsetting growth. None of those are visible in Q4 data.

Weight: 5-7% in high growth portfolio. Not yet a concentration-limit name — Google search dependency is a tail risk I cannot quantify — but the valuation case is strong enough to warrant meaningful exposure now.