Why are my Facebook ads getting more expensive?

Your Facebook ads are getting more expensive because Meta's average price per ad rose 12% year over year in Q1 2026, while the number of people using Meta grew only 4%. More advertisers are bidding for barely more attention. This is not a targeting problem or a creative problem you can fix inside Ads Manager. It's the structure of the auction, and it moves one way. The only durable answer is to build acquisition that doesn't run through the auction at all.

You didn't change anything. Same offer. Same product. Same audiences, roughly. The creative gets refreshed on schedule. And yet the number at the bottom of the dashboard keeps drifting the wrong way, month after month, and the meeting where you explain it is getting harder each quarter.

You're not imagining it, you're not doing it wrong, and there is nothing broken in your account. The cost of Facebook ads is rising for structural reasons that have nothing to do with your campaigns. Fair warning: we run an organic growth system built underneath paid ads, so we have a view. We'll be clear about which parts of this are documented fact and which parts are our argument.

Skip the agency blog estimates. Meta is a public company and files its numbers with the SEC every quarter.

In its first quarter of 2026, reported on 29 April 2026, Meta's average price per ad increased 12% year over year. Ad impressions across the Family of Apps grew 19%. Total revenue hit $56.31 billion for the quarter, up 33%.

Now the number that matters more, and it's in the same filing. Family daily active people reached 3.56 billion in March 2026, an increase of 4% year over year.

Read those two together. The price of an ad went up 12%. The number of humans available to see ads went up 4%. That's the whole story of your rising costs in two lines. Meta isn't running out of people. It's running out of new people, roughly, while the number of businesses bidding to reach them keeps climbing. When demand for a fixed thing grows faster than supply, the price goes up. That's an auction working exactly as designed.

And notice the trajectory. In full year 2025, Meta's average price per ad rose 9%. In Q4 2025 alone it was 6%. By Q1 2026 it's 12%. The rate of increase is itself increasing.

Here's what makes this so slippery to manage.

Nobody sends you an email saying costs are up 12%. There's no renewal notice. It arrives as a slightly worse week, then a slightly worse month, then a quarter where you find yourself explaining the numbers with a bit more effort than last time. Each individual movement is small enough to be noise. Small enough to blame on creative fatigue, or seasonality, or that one campaign someone left running.

Creep is worse than a jump, because nobody calls a meeting about creep. A 40% overnight rise would trigger an emergency review, a strategy session, a genuine rethink. A 1% monthly drift triggers a new batch of creative and a note to check it again next month. Same destination. One of them you notice in time.

Most founders we talk to can tell us their blended CAC to the pound and still can't say precisely when it started climbing. That's not sloppiness. That's what creep does.

When costs rise, there's a standard sequence. You've probably run some or all of it.

Fix one: refresh the creative

This works. It genuinely works, and it's the right first move. New angles, new hooks, new formats, and the numbers improve. For a while. Then they drift back. Six creative cycles in, you're running a permanent content operation to hold a number that used to hold itself.

The creative isn't failing. It's paying an ever-rising toll. You're producing more work each quarter to stand in the same place, and the treadmill speed is set by the auction, not by you.

Fix two: rebuild the targeting

Also sensible. Also increasingly futile, and not because you're bad at it. Since iOS App Tracking Transparency, the signal Meta uses to find your buyer has been substantially degraded. The platform's response has been to take manual controls away and hand the job to its own AI. That's a reasonable response to a real problem. But it means the lever you're reaching for has less travel in it every year, by design. You can't out-target an auction that's got more bidders in it.

Fix three: raise the budget

The honest one. If each customer costs more, spend more to keep the volume. Which works, and which is exactly how a business ends up where most of our clients were when they called us: paid spend has become the entire growth plan without anyone deciding it should, nothing else is load-bearing, and every increase makes the dependency deeper. We've written about the compounding cost of that dependency in detail, because the three-year arithmetic surprises people.

Fix four: change agency

The most expensive fix, and the most understandable one. The new agency runs a fresh audit, finds real inefficiencies, rebuilds the account, and the numbers improve for a quarter. Then the same drift resumes, because the drift was never about the agency. Twelve months and a re-onboarding later, you're where you started with a different logo on the report.

None of these four is stupid. They're all rational moves. The problem is they're all moves inside the auction, and the auction is the thing that's getting more expensive.

Here's the uncomfortable bit, and it isn't a criticism of anyone.

Meta's revenue target is your ad budget. Not metaphorically. Advertising revenue is essentially the entire business, and the quarterly filings tell you exactly how well the plan is going: $56.31 billion in Q1 2026, up 33% year over year.

That's a company doing precisely what it's supposed to do for its shareholders. It isn't a betrayal, and this isn't a reason to be angry about it. But it should tell you which way the price of a rented asset moves over time when the landlord is a public company with a growth target, an auction full of bidders, and roughly the same number of eyeballs each year. It moves one way.

You already know this in every other part of your business. Nobody builds a company on a supplier who sets prices unilaterally, reports the increases publicly, and has no competitive pressure to stop. Yet a startling number of good businesses have exactly one acquisition channel and it has all three of those properties.

No. Categorically not, and be suspicious of anyone who suggests it. Your ads work. They're delivering customers today. Turning off a channel that's currently paying the bills, in the hope that something slower fills the gap, is how businesses die. We say this as a company that sells organic growth: the paid stays.

The problem isn't that you run paid ads. The problem is that paid ads are the only thing carrying weight. There's a version of this where paid does what it's genuinely good at, which is fast, controllable, high-intent demand capture, while other things underneath it take a share of the load. In that version, the 12% rise still happens. It just applies to a smaller portion of your acquisition, and it doesn't set the fate of the business.

At this point most articles pivot to a paragraph about content marketing and a lead form. Let's be more honest than that. You've probably tried organic. Most founders at £50k to £250k a month have. You set up a company account, posted for a couple of months, watched nothing happen, and went back to what works. That experience was real and the conclusion you drew was reasonable.

But it wasn't a fair test, and we've written the full diagnosis in why organic didn't work for you. The short version: it had no research under it, there was nothing worth pointing to, nothing fed anything else, and it got judged on paid timelines. Same input, same expectation, worse channel. That's not evidence the channel doesn't work. It's evidence it was run backwards.

Four things, and the point is that they only work together. Individually they're commodity tactics you could hire tomorrow. Wired together they compound, which is the entire difference.

The rooms where your buyers actually decide

Your buyer doesn't start with your ad. They start with a question. They type it into Google, or ask it in a subreddit, a Slack group, a trade forum, a comment section. They read what people like them say. They shortlist. They make most of the decision. Then they click an ad and the ad takes the credit. You paid full auction price for demand that was shaped somewhere you had no presence.

Finding those rooms is research, not guesswork, and we've covered how that mapping works in where your customers actually are. Then there's the harder part: being in those rooms without getting thrown out. Doing this properly takes warmed accounts, ratio discipline and genuine participation, and it takes patience.

Content built from real buyer language

Start with the sentences you mined in the rooms, and you get content that reads the way your buyer thinks, ranks for the searches that actually convert, and gives your presence in those communities something honestly useful to point at. Fair warning: this is slow. Months, not weeks. Anything that produces results in a fortnight isn't load-bearing.

A referral engine that actually fires

The most neglected asset in most businesses, and the one that pays back fastest, because it fires on customers from every channel including the paid you're already running. Not a refer-a-friend link in the footer. A structured system: the ask, the timing, the incentive, the tracking, the follow-through. When the referral engine is running properly, every referred customer lowers the effective acquisition cost of whichever channel brought their source customer in. Including Meta.

Outbound, to carry the gap

Everything above compounds, which means it takes time. Outbound doesn't. It puts real conversations in the calendar from week one while the rest takes hold, and every reply tells you which rooms your buyers are actually in, which feeds the research.

Not what we'd sell you. What we'd tell you.

This week. Work out what percentage of your customers arrive through paid. If it's above 80%, that's the number that matters more than your CAC. Your CAC is a cost. That percentage is a risk.

This month. Look at your referral situation honestly. Not "do we have a referral programme" but "did anyone systematically ask our last fifty happy customers for an introduction, at the right moment, with a reason to do it." Almost always, nobody did. That's free money you're not collecting, and it's the fastest weight off the auction.

This quarter. Find out where your buyers actually research before they buy. Ask ten recent customers how they first heard about the problem you solve, and where they looked before they found you. Ten conversations. You'll learn more than any tool will tell you.

Sequence matters more than spend here. We've laid out the argument for why build order beats budget in full, because the fix isn't more effort, it's a different order.

Your Facebook ads are getting more expensive because Meta's average price per ad rose 12% year over year while its user base grew 4%. That's an auction with more bidders and roughly the same attention to sell. It isn't your creative, it isn't your targeting, and it isn't your agency.

Every fix available inside Ads Manager is a move inside the auction, which is why every fix works for a quarter and then drifts back. The auction is the thing getting more expensive, and you can't out-optimise it.

The ads should stay. The question is whether they're the only thing carrying the weight, and whether you're comfortable with a business where every future customer arrives through one channel you don't control, at a price that's published quarterly and only moves one way.

See what a system that doesn't run through the auction looks like →