Why Cost Per Lead Stopped Telling You Anything
The number every business watches is the one most detached from whether the campaign made money, and privacy changes since 2021 made it worse rather than better.
Ask most owners how their advertising is performing and they quote a cost per lead. It is the number on the dashboard, it moves when things change, and it feels like the answer.
It is not the answer, and in some accounts it moves in the opposite direction to profit.
The problem in one example
Two campaigns. One produces leads at forty dollars, the other at ninety.
Every dashboard says the first is better. But suppose the cheaper leads book calls a third of the time and close at one in ten, while the expensive ones book most of the time and close at one in three.
The expensive campaign is producing customers for a fraction of what the cheap one costs. And if you optimise toward the visible number, you will move budget onto the campaign that is quietly losing money, then wonder why revenue did not follow volume.
This is not a hypothetical edge case. It is the default outcome of optimising toward the metric closest to the ad and furthest from the cash.
Why it got worse after 2021
Apple’s App Tracking Transparency, released with iOS 14.5 in April 2021, meant the platforms lost sight of a large share of what happens after a click. Flurry’s tracking showed the overwhelming majority of US users declining permission.
So the platforms moved to modelling. The conversions in your dashboard are increasingly estimates rather than counts, which is fine for optimisation and misleading if you treat them as a ledger.
Meanwhile the one number that never degraded is the one in your own systems: which enquiries became customers, and what they were worth. That data got relatively more valuable at exactly the moment most businesses were still watching the platform.
The metrics that actually connect to cash
Cost per qualified lead. Same as cost per lead, but counting only the ones that matched your criteria. This one change reorders most accounts.
Cost per held call. Booked is not held. No show rates vary enormously by campaign and by how the offer was framed.
Cost per closed deal, by campaign. The number that decides budget. It requires your CRM and your ad accounts to be connected, which is why most businesses do not have it.
Payback period. How long until a customer has repaid the cost of acquiring them. This determines how fast you can afford to grow.
Why almost nobody has these
Not because they are complicated. Because the source of each is a different system.
Spend is in the ad platform. Leads are in a form tool or an inbox. Qualification lives in somebody’s head or a CRM field they update when they remember. Closed revenue sits in accounting. Connecting the four means a lead keeps its campaign source all the way through to the invoice, and in most businesses that chain breaks at the first step.
Where AI closes the loop
Two specific jobs, and both were impractical before.
Qualifying every lead at the moment it arrives. A person cannot do this at three in the morning or at volume, so qualification either happened late or not at all, which is why cost per qualified lead was a theoretical metric. AI does it on arrival, against your written criteria, and marks the record accordingly.
Keeping the source attached. Every conversation, from any channel, written to one contact record with its campaign source intact, so a closed deal months later still knows which ad produced it.
Once those two exist, the useful numbers become available without anybody assembling a spreadsheet, and budget decisions stop being made on the metric that happened to be visible.
What to do first
Write down what qualified means, in a list somebody could apply without asking you. That single document is the precondition for everything above and most businesses have never produced it.
Then make sure the campaign source survives into the CRM and onto the deal. Then judge campaigns on held calls and closed deals, and let cost per lead be a diagnostic rather than a target.
Define qualified before you measure anything
Every metric above depends on one document that most businesses have never written: what a qualified lead actually is.
It needs to be a list somebody could apply without asking you. Sector, size, geography, the type of work, budget range, timing, and who has to be involved for a decision to happen. Specific enough that two people reading the same enquiry reach the same answer.
Until that exists, cost per qualified lead cannot be calculated, because qualified means whatever the person looking happened to think that day. Writing it takes an hour and it is the precondition for every other improvement on this page.
Payback period decides how fast you can grow
The metric owners most often skip is the one that governs everything else: how long until a customer has repaid what it cost to acquire them.
A business recovering acquisition cost inside a month can reinvest almost immediately and grow at the speed of demand. One taking nine months is funding that gap out of working capital, and its growth rate is capped by cash rather than by market.
Two businesses with identical cost per closed deal can therefore have completely different capacity to scale. If you are deciding whether to increase spend, this is the number that answers it, not cost per lead.
Judge over a full cycle
One caution about all of this. If your sales cycle is three months, a campaign judged after three weeks is being judged on the wrong evidence.
Early metrics measure the top of the funnel. Cost per closed deal cannot be known until deals have had time to close. Businesses that switch campaigns off at week four are frequently killing the ones that were about to work, and doing it using a number that was never designed to predict.
Sources
Apple, App Tracking Transparency framework. Introduced with iOS 14.5, released 26 April 2021, requiring apps to obtain user authorisation before tracking activity across other companies apps and websites.
Flurry Analytics, iOS 14.5 opt-in rate tracker, daily measurement from launch.
AI Optimize connects the ad accounts, the qualification and the CRM so you can see cost per closed deal by campaign rather than cost per form fill. That work sits under Paid Ads Management and Reporting & Data.
Related reading
Your Landing Page Is the Campaign
Most advertising budgets are spent optimising the ad and pointing it at a page nobody rebuilt. The page is where the money is decided, and it is usually the least examined part of the account.
What Apple's Privacy Change Did to Your Ad Targeting
The single largest shift in paid acquisition this decade was not a platform update. It was a permission prompt, and most advertisers still run accounts built for the world before it.
WHAT WE BUILD