Eight KPI relationships most reports never compare

Most reports treat every KPI as its own line: CPA this month, CTR this month, reach this month. Read on their own, none of the three says anything is wrong, and yet the combination can hide a problem that only becomes visible once you set two KPIs explicitly against each other.

Three of the eight, in detail

A few concrete examples. CPA decomposition splits a risen CPA into its two possible causes, a dropped CTR or a risen CPC, because the follow-up step differs for each. A promise gap sets the message in the ad against what the landing page actually delivers: a high CTR with a low conversion rate is often not a targeting problem but an expectations problem. Vanity engagement flags a campaign with lots of interaction and little value, the kind of number that looks good in a report and does nothing for revenue.

A worked example

Illustrative case: a campaign's CTR climbs to 4.2 percent, well above account average, and the specialist reports it as a targeting win. Conversion rate on the same traffic sits at 0.6 percent, roughly a third of what the rest of the account converts at. Read separately, both numbers can pass review -- a strong CTR and an unremarkable conversion rate. Set against each other as a promise gap, the pattern reads differently: the ad is earning clicks the landing page is not equipped to close, which points at the ad's message or the page's follow-through, not at who the campaign is targeting.

The other five relationships

The other five follow the same logic: saturation (repeated exposure to the same people with no extra return), reach dilution (growing reach with declining relevance per person), value mix (conversions rising in count but dropping in value), frequency-versus-reach (frequency climbing while reach stays flat, a sign the audience is exhausted), and expensive visibility (a top placement that costs more than it returns in extra conversions). Eight relationships, and none of them are visible in a report that lists every KPI separately.

What they have in common

What they have in common: every relationship revolves around two metrics that almost never appear on the same row of a report, let alone get set against each other explicitly. That is not unwillingness on the part of whoever builds the report, it is simply not what a standard overview is built for. It is, however, exactly where the next decision comes from.