Meta's Q3 2026 earnings, released October 21, put a number on something a lot of businesses running Meta ads have been feeling for months: Meta ad costs are climbing, and it's not because more people are seeing your ads.
Ad revenue hit $63.06 billion, up 26% year over year. Split that into price and volume, and the story changes. Impressions grew just 10% worldwide, the slowest pace Meta has reported in five quarters. The average price per ad rose 23% in the US and Canada specifically.
If you run Meta ads for a living, that's not a headline. That's your invoice.
What Did Meta's Q3 2026 Earnings Actually Show?#
Meta's ad business grew almost entirely on price, not reach. Total ad revenue rose 26% to $63.06 billion, and the US/Canada price rise is the third straight quarter of acceleration: 14%, then 20%, then 23%. The regional split matters more than the global number.
| Region | Price-per-ad growth, Q3 2026 vs. Q3 2025 |
|---|---|
| US & Canada | +23% |
| Rest of World | +24% |
| Europe | +10% |
| Asia-Pacific | +1% |
Meta also booked a $9.89 billion charge tied to an August 2026 settlement with 52 state attorneys general over teen mental health, which pulled operating income down 37% even as revenue climbed. That charge is a legal cost, not an ad-delivery or pricing change, and Meta's quarter also included Muse Wonder, its first frontier model, with no figures tying it to ad performance. At WebEpex, we've tracked CPM movement across client Meta accounts since Q2 2026, and this quarter's US/Canada acceleration matches what we started seeing two quarters ago.
Why Did Meta Ad Costs Rise While Reach Didn't?#
My read is that this is auction economics, not a new fee. More advertisers are bidding into a flat-to-slow-growing pool of inventory, a lot of it through Advantage+ and other AI-assisted ad creative tools that Meta says are driving "substantial incremental advertising revenue," its words, with no number attached. When demand outpaces supply in an auction, price absorbs the gap.
I'd also flag this as my opinion, not Meta's claim: slower impression growth could mean Meta is quietly tightening inventory in some placements to protect ad quality, not purely a demand story. We don't have proof of that either way, and Meta's release doesn't say.
Does This Change Your Meta Ad Budget?#
If you're spending under roughly $2,000 a month and your account sits outside the US or Canada, a 23% regional swing barely touches you. Global blended price growth is nowhere near that steep, and normal week-to-week noise in your own account will outweigh it. You can stop reading here.
If you're a US or Canada advertiser running meaningful always-on spend, three straight quarters of accelerating price growth is a real line item now, not something to wait out. We've managed Meta ad spend for US and Canada clients through all three of those quarters, and the ones we didn't actively adjust for saw cost-per-result climb right alongside CPM.
The Trade-Offs Nobody's Advertising#
Pro: Meta's AI targeting does more than it did two years ago. On accounts with clean conversion-event feeds running Advantage+, we've watched cost-per-result hold steadier than CPM alone would suggest. That's real.
Con: Meta gave zero advertiser-outcome numbers in this release, no conversion data or ROAS, nothing beyond "substantial incremental revenue." We're being asked to trust that AI targeting offsets rising prices on no evidence beyond the revenue line that benefits from us believing it.
One more con: slower impression growth, even at 10%, leaves less room to scale reach cheaply if a campaign leans on broad awareness over tight retargeting. That's a real constraint for brand-building budgets. Direct-response budgets feel it less.
How We're Adjusting Client Meta Budgets Right Now#
We pulled 90-day CPM and cost-per-result trends across every active client Meta account the week this release dropped, split by region, to see who was genuinely exposed to the US/Canada run-up and who was just feeling normal volatility. Two accounts were actually hit. The rest weren't.
For those two, we shifted budget weighting toward Advantage+ placements with tighter conversion-event feeds, the thing that seems to make the AI targeting earn its premium, moved part of spend to WhatsApp click-to-message where CPMs are still lower in our accounts, and set harder CPA ceilings so a bad auction week caps itself automatically. It's the same reserved-and-forecasted logic we used when Meta tested premium-priced delivery windows with Ads Super Burst earlier this quarter, applied to everyday spend instead of a launch moment.
At WebEpex we build and manage Meta ad campaigns for GCC, European, US, Canadian and Indian businesses, and this is the first quarter that work has meant an explicit "defend CPA, not CPM" playbook instead of just watching the dashboard. We're not pulling budget off Meta over this; the platform still converts. We're just not assuming the AI tools quietly cancel out a 23% price rise without checking the receipts account by account.
What I'd Tell a Client This Week#
Pull your last 90 days of CPM against cost-per-result, not CPM alone. If cost-per-result rose in lockstep with CPM, your targeting absorbed nothing, and you should test Advantage+ with a cleaner conversion feed before anything else. If cost-per-result held steadier than CPM, the AI targeting is doing real work for you, and the price rise matters less than the headline suggests. Either way, set a hard CPA ceiling this week. It costs nothing and caps your downside while the auction sorts itself out.
None of this calls for panic. It calls for checking the account, not the headline.
If you want a straight read on whether this is actually hitting your account or just normal noise, send me what you're running and I'll tell you plainly. Two minutes, no pitch. Reach out here.
