Two of the most reliable datasets in this business moved in opposite directions this year, and almost nobody has noticed. Understanding why cost per lead is rising on one platform while falling on the other is worth more to your budget than any industry average.
Meta got dearer. Search got cheaper. If you run both and treat them as one pot of money, your split is now working from prices that no longer exist.
This is for you if your Meta figures crept up and you are trying to work out whether it is you.
Why cost per lead is rising on Meta
The clearest evidence comes from Meta itself, which has no marketing angle in its own filings.
For the quarter ending 30 June 2026, Meta reported average price per ad up 12% year over year, with ad impressions up 14%. More inventory, sold at higher prices, which is a demand story rather than a performance one.
That has a direct consequence for you. An account changed in no way at all costs roughly a tenth more this year than last. Before you conclude your creative has stopped working, subtract that.
Crucially this is not a blip. Meta reported the same 12% increase in the preceding quarter, so two consecutive quarters point the same way.
Why search cost per lead went the other way
WordStream’s 2026 study of 13,474 US search campaigns found the overall average cost per lead falling to $66.69, which it describes as the first decrease in five years.
The mechanism matters more than the headline. Clicks did not get cheaper. Conversion rate increased for 87% of industries, so the same traffic produced more enquiries.
That is a genuinely different kind of movement. Meta’s change is a price change imposed on you. Search’s change is a performance improvement achieved by advertisers, which means it is a route available to you rather than something that happened to you.
What a rising cost per lead means for your budget
Nobody in this space seems to be planning around this, which makes it an advantage while it lasts.
| Direction | Cause | Can you influence it? | |
|---|---|---|---|
| Meta | Up ~12% | Higher price per ad | No, only your efficiency |
| Google search | Down | Conversion rate up for 87% of industries | Yes, directly |
A budget split decided eighteen months ago now rests on stale prices. That does not mean move everything to search. It means the arithmetic deserves redoing, on closed business rather than on enquiries.
Remember search leads convert better once they arrive. Ruler Analytics, from 110 million sessions across 13 industries, measured paid search at 5.4% against 2.11% for paid social. A dearer enquiry that closes more often is frequently the cheaper customer.
Separating market movement from your own account
Do this before you change anything, because it takes ten minutes and prevents expensive mistakes.
Take your Meta figure against the same period last year. Subtract roughly 12% of the increase as platform pricing. What is left is yours, and that is the part worth investigating.
On search, there is no market increase to subtract this year. So if your search costs rose, the cause is entirely internal, and that is a much sharper signal than it would have been in any other recent year.
Then check whether you raised budget. Scaling pushes you into more expensive parts of the auction, and that effect stacks on top of platform inflation rather than replacing it. If you did both, expect a rise well beyond 12%, which is normal auction behaviour with its own fixes.
What helps when cost per lead is rising
You cannot negotiate with an auction. You can change what you put into it.
Creative is the first lever on Meta, because it now does the job audience targeting used to do. A genuinely new angle moves your figure more than any setting, and refreshing on a rhythm beats refreshing when something breaks.
The optimisation event is second. Most accounts I audit are optimising on the wrong one, usually because it was set once years ago and nobody looked again. Rising prices punish that mistake harder than flat prices did.
Conversion rate is third, and it is the lever search advertisers used to bring their whole market down this year. Improving the page a lead lands on lowers cost per lead without touching a bid.
One thing I would not do: cut Meta spend purely because the platform got more expensive. If the channel was profitable at your ceiling, a 12% price rise usually leaves it profitable, just less so. Decide with the ceiling, not with the direction.
The trap of reacting to a rising cost per lead
Two responses come naturally and both usually make things worse.
The first is rebuilding the account. Performance dips, someone concludes the structure is tired, and a working setup gets torn down. All the learning built up in the bidding goes with it, and the new build spends weeks getting back to where it started. Do not rebuild because performance dipped for four days.
The second is cutting budget. If the channel is profitable at your ceiling, a 12% price rise usually leaves it profitable, just less comfortably. Cutting spend on a profitable channel to protect a metric is the sort of decision that looks responsible in a report and costs money in the bank.
There is a third response that people skip because it is dull. Check whether the leads are actually being counted. Meta browser tracking has been unreliable since Apple’s changes, and an account can look 20% worse than it is for months.
Ask one question before acting: is the enquiry still worth more than it costs? If yes, the rise is an efficiency problem, not an emergency. Efficiency problems get solved calmly over a few weeks. Emergencies get solved badly on a Friday afternoon.
What to change this week
Four steps.
Pull your Meta cost per lead against the same quarter last year and subtract about 12% of the increase before drawing any conclusion.
Do the same for search, where there is nothing to subtract, so any rise points inward.
Redo your channel split using this year’s prices and judge it on cost per closed customer.
Then check the age of your top Meta creative and which event your campaigns optimise for, because those are the two fastest fixes available.
The wider piece on judging lead cost covers where the ceiling sits, what a Facebook lead costs right now has the Meta detail, and the search figures cover the other half. If you want the split worked out against your own numbers rather than an average, send them to me.