What blended metrics measure
A blended figure ignores who claimed the credit. You take every marketing cost in a period — media spend, agency or freelancer fees, tools — and divide it by every new customer, order or qualified enquiry your own systems recorded in that same period. Blended cost per acquisition is that division; blended return on ad spend is revenue over the same total cost.
The numbers come from your accounts and your order or CRM records, not from any advertising platform. Nothing is attributed, nothing is modelled, and nothing can be counted twice. That is the whole point: a blended metric is the one figure in marketing that a finance team recognises without translation.
Why blended metrics matter
Because platform reports cannot be added up. Google, Meta and your email tool can each claim the same buyer, so the sum of their conversions exceeds the number of customers the business actually gained. Blended figures sidestep the whole argument by counting outcomes once, at the level where money is made.
They also catch effects that channel reports miss. Paid social often lifts branded search; a campaign that looks mediocre in its own dashboard can still move the blended number, and a channel with a beautiful reported ROAS can be adding nothing at all. When the platforms all claim improvement and the blended figure stays flat, something in the reporting is wrong — which is the everyday form of an attribution discrepancy.
Where blended metrics go wrong
They are diagnostic, not directional. A blended cost per acquisition tells you the programme is drifting; it never tells you which campaign caused the drift. Using it alone leads to panicked across-the-board cuts, and the channel that was actually working gets cut with the rest.
The second problem is definition drift. If one month counts every enquiry and the next counts only qualified ones, or media spend is included in one period and agency fees in another, the trend line is meaningless. Mixing new customers with repeat purchases is a related trap: retention revenue flattering an acquisition metric hides a rising cost of winning anyone new.
Timing catches people out too. Spend lands this month; the sale it caused may land next month. On a long consideration cycle, a blended figure read week by week will swing for reasons that have nothing to do with performance.
How to act on it
Fix the definition once and write it down: which costs are in, which outcomes count, and over what period. Then track the blended figure monthly as your headline, with platform reports underneath it as the explanation of movement rather than the verdict. Most useful reporting dashboards put both on one screen, which is a standard part of reporting and dashboard work.
When the blended number moves and the platform numbers do not, resist the temptation to trust whichever story is more comfortable. Check the obvious causes first — a price change, a stock problem, seasonality, a landing page break — and if the disagreement persists, settle it with an incrementality test rather than another argument about dashboards.
