Earnings Analysis Q3 2026

Meta Q3 2026 Earnings: Price per Ad Up 14% as Impression Growth Slows to 10%

Meta reported third quarter 2026 results on October 21, 2026: $64.6B in revenue, up 26%, above its own $61B to $64B guidance. Ad impressions grew 10% and the average price per ad grew 14%. Put next to the first two quarters of the year, that is a change in where Meta's growth comes from. In Q1 it came mostly from showing more ads. In Q3 it came mostly from charging more for each one. For anyone buying Meta inventory through Black Friday, that is the number that matters.

November 13, 2026 7 min read Meta Ads, CPM, BFCM

The numbers

  • Total revenue: $64.60B (+26% YoY), above the $61B to $64B guidance range
  • Advertising revenue: $63.06B (+26%)
  • Ad impressions: +10% YoY
  • Average price per ad: +14% YoY
  • Family daily active people: 3.64B on average for September 2026 (+3%)
  • Costs and expenses: $51.61B (+68%), including a $9.89B legal charge tied to the August 2026 multistate attorneys general agreement
  • Income from operations: $12.99B (-37%); $22.88B excluding the legal charge
  • Net income and EPS: $9.86B and $3.83 per diluted share; $7.40 per share excluding the charge
  • Capital expenditures: about $34.9B in the quarter, more than the $32.8B of cash generated from operations
  • Headcount: 66,775 (-15% YoY)
  • Q4 2026 revenue guidance: $73B to $76B
  • 2026 capex guidance: $135B to $145B (low end raised from $130B)
  • 2026 total expenses guidance: $175B to $179B, raised from $165B to $169B to absorb the legal charge

A note on the EPS comparison. $3.83 is up 265% on Q3 2025's $1.05, but that year-ago quarter carried a one-time tax charge, and this one carries a one-time legal charge. Neither year's reported EPS describes the ad business. Revenue, impressions and price per ad do.

The shift: 2026 growth went from volume to price

Meta's ad revenue growth is roughly impression growth multiplied by price growth. Here are the three quarters of 2026 side by side.

Quarter Revenue growth Ad impressions Average price per ad
Q1 2026+33%+19%+12%
Q2 2026+28%+14%+12%
Q3 2026+26%+10%+14%

Source: Meta quarterly earnings releases for Q1, Q2 and Q3 2026. All figures are year-over-year.

Impression growth has almost halved in two quarters, from 19% to 10%. Price growth went the other way. In Q1, volume did most of the work. In Q3, price did.

Meta's own explanation on the call: the price increase came from "ad performance gains and improvements in macro conditions relative to Q3 of last year," partly offset by strong impression growth on lower-monetizing surfaces and regions. Its 10-Q says impressions grew in every region, especially Asia-Pacific. Read plainly: advertiser demand is rising faster than the supply of ad slots, and the fastest-growing supply is in the cheapest places.

What advertisers should actually take away

+14% price per ad is not your CPM, but it points the same way

Meta reports average price per ad across every surface, format and country. It is not a CPM or a CPC, and the mix moves it. If impressions shift toward cheaper regions, the global average understates what is happening in expensive ones. With most new inventory coming from Asia-Pacific and lower-monetizing surfaces, a US or European advertiser should assume their own price increase was at least the global 14%, then check. The earnings release does not break price out by region. In Q1, Meta's presentation slides did: US and Canada ran at +14% and Europe at +19% against a global +12%. Pull your Q3 2026 CPM against Q3 2025 by country.

Q4 guidance says the auction gets tighter, not looser

Meta guided Q4 revenue to $73B to $76B, assuming a currency headwind of about one point. Against Q4 2025, that range works out to growth in the low-to-high twenties on our arithmetic. If impression growth stays near 10%, the rest has to come from price again. Q4 is already the most expensive quarter of the year. Budget BFCM on the assumption that Meta CPMs run well above last year's, and decide now which campaigns you would cut first if they do.

Nearly 10 million advertisers used a generative AI ad tool

Meta said nearly 10 million advertisers used at least one of its generative AI ad tools in Q3, and that new features will keep arriving primarily through the Advantage+ suite. Two things follow. Creative volume in the auction is rising, so creative fatigue arrives sooner. And "ad performance gains" is the first reason Meta gives for higher prices: when the system converts better, advertisers bid more, and the gain is shared with Meta. Measure Advantage+ on incremental results, not on the platform-reported lift.

Audience growth is 3%. The rest is monetization

Family daily active people reached 3.64B, up 3%, slightly under the 3.65B analysts expected according to CNBC. Revenue grew 26% on a user base growing 3%. Meta is earning more per person through more time spent, more ad load and higher prices. That is fine for Meta, and it is a reminder that reach on these apps is close to saturated. New scale has to be bought at higher prices or found on newer surfaces.

WhatsApp is becoming a revenue line

Family of Apps other revenue was $1.1B, up 59%, driven mainly by WhatsApp paid messaging and subscriptions. It is small next to $63B of advertising, but it is the fastest-growing line Meta reports. If you run click-to-WhatsApp ads or paid conversations, expect more product and more pricing changes there.

The legal charge and capex are investor stories

The $9.89B legal charge cut operating income by 37%. Capital expenditures of about $34.9B exceeded operating cash flow, leaving free cash flow slightly negative. Meta also announced a non-voting Class C share to be distributed as a stock dividend, which works like a 2-for-1 split, with trading expected in late January 2027. CNBC reported the shares rose about 2.5% in extended trading. None of it changes what your ads cost this quarter. One detail worth noting: headcount is down 15% year over year, and Meta named third-party AI token costs among its cost drivers.

Paying more per ad raises the cost of every outage

Earnings do not mention reliability, so here is what our monitoring logged in the same quarter. On July 16, 2026, Meta's Ads Delivery was disrupted from 09:40 to 11:47 UTC, and Meta's status RSS feed did not publish the incident until about 90 minutes after its own dashboard showed it. On September 22, Meta logged another Ads Delivery disruption in the middle of the US afternoon.

Last year the pattern was worse at exactly this time of year: Meta logged 11 incidents between September 15 and November 10, 2025, the run-up to Black Friday. We do not know whether 2026 repeats it. We do know that at 14% higher prices, an hour of broken delivery or wrong reporting costs more than it did a year ago.

Details: the July 16 delivery outage, the September 22 incident, and what happened before BFCM 2025.

What to check in your account this week

  1. 1 Compare Q3 2026 CPM to Q3 2025, by country. If yours rose less than 14%, you are beating the average. If it rose more, find which placements carry the increase.
  2. 2 Set a BFCM CPM ceiling before the week starts. Decide the cost at which each campaign stops being profitable. It is easier to hold that line when it was written down in advance.
  3. 3 Check creative fatigue faster. With millions of advertisers generating variants, frequency and CTR decay deserve a weekly look in Q4, not a monthly one.
  4. 4 Validate Advantage+ with a holdout. Platform-reported conversions and incremental conversions are different numbers. Higher prices make the difference more expensive.
  5. 5 Have an outage plan for peak week. Know who gets told, which budgets get paused and where the status information comes from. Meta's RSS feed lagged its own dashboard by about 90 minutes in July.

Sources

Financial figures are from Meta's earnings release and 10-Q. Analyst expectations and the share price move are from CNBC. Incident data is from AdStatus monitoring.

Know about a Meta outage before your CPA tells you

AdStatus checks Meta's official status sources every 5 minutes, including the dashboard data that showed the July outage before the RSS feed did, and posts each incident and its all-clear in Slack or by email. It covers Google Ads, Microsoft Advertising, Pinterest, Amazon Ads and Shopify too. 7-day free trial, then $10/month.