Skip to content
  1. Home
  2. Glossary
  3. Ad spend
Spend & efficiency

Ad spend

How much money went into advertising over the period — the number every efficiency calculation starts from, and the one that almost never matches what the bank took.

Formula
sum of spend across connected ad accounts
Unit
Currency
Direction
Lower is better
Across periods
Adds up across periods
In KO
spend

Spend is the money advertising cost you over a period. It sits in the numerator of nearly every efficiency metric — CPL, CP QL, ROMI all divide or subtract it.

It looks like the most reliable figure in the report: nothing to tag, nothing to attribute, it simply sits in the ad account. Which is why nobody checks it — and why it almost always means something other than the reader assumes.

What is counted

Spend = sum of spend across all connected ad accounts for the period

Three questions every spend figure answers one way or another, and they decide whether two numbers can be compared at all:

QuestionPossible answers
Whose spend is itthe ad account · the bank statement · the agency invoice
In which currencythe account’s · the report’s at the daily rate · at the period rate
With or without taxplatforms in different countries report it differently

Where it parts ways with reality

The ad account is not the bank. The account holds delivered budget, the statement holds charges. Between them sit prepayments, credits, refunds for invalid traffic, the platform’s fee and the agency’s. A 5–15% gap between “spend” in a report and money leaving the account is normal, not a bug; assuming there is no gap is the bug.

The date of spend is not the date of money. A platform books delivery to the day of the impression in its own timezone, and corrections for invalid clicks arrive days later. A report for yesterday, pulled yesterday and pulled a week later, shows different numbers for the same day. That is not broken data — that is how platforms close a period.

Currency means a rate, and a rate means a date. A dollar account and a euro report agree only after conversion, and the result depends on whether the rate is taken on the day of delivery or at the end of the period. Two correct reports on the same campaigns can differ by percentage points for this reason alone.

Not all spend is visible. Barter, seeding, a contractor running their own account, paying an influencer directly — all cost money, and none of it is in the connected accounts. Any “result per spend” metric for such a channel comes out dazzling, because the denominator is incomplete.

How KO computes it

In KO spend is an additive metric: it arrives from ad platform data as the daily spend of an ad and sums upwards through any grouping — campaign, channel, breakdown, period. There is no formula to recompute, so there is nothing to get wrong in aggregation: a sum stays a sum at every level of the report.

One thing to keep in mind when reading spend in a breakdown: it is not attributed. Leads and sales are tied to a channel through touches; spend is tied to a campaign by the platform itself. So a channel row always carries its full spend and only the result that could be linked. That is where a systematic bias comes from: the worse the tagging, the worse paid channels look, and the channel is not what is at fault.

Organic and direct have no spend — there it is absent, not zero. KO does not compute cost-per-result metrics on those rows at all, rather than printing a zero.

  • CPL — spend divided by the number of enquiries.
  • CP QL — the same division, one funnel step later.
  • ROMI — what the spend returned, not what it cost.