Marketing KPIs
A marketing KPI is a quantified measure of marketing performance: cost to acquire a customer, return on ad spend, organic traffic, repeat purchase rate, lifetime value. Teams pick a handful, put each on a dashboard, and track it over time. Problems begin when one KPI is elevated above the rest and read on its own, because a single number contains no context about its cause. We document [what single-point metrics cost](/insights/why-strategic-initiatives-fail#single-point-metrics) in the strategic initiatives article.
How it actually works
The traditional approach selects a few KPIs and tracks each in isolation, one tile per metric. That approach assumes a single metric tells the whole story. A single metric in isolation reports what happened. The cause, the significance, and the appropriate response come from the relationships between metrics, and which relationships matter changes with where a customer is in their journey.
When a team optimizes one metric in isolation, the sequence is consistent: a number moves, a wrong cause gets assigned, budget and headcount get committed to that wrong cause, and the initiative scales before anything was proven. OKRs show a version of the same problem when they count activities completed without measuring the effect on the bottom line. Profitability can fall in the same quarter marketing reports its highest ROAS, because the two numbers measure different things.
At Old School Labs, two products told opposite stories depending on which number was read. Product B had the healthy ROAS and recovered the cost of acquiring a customer in four months. Product A cost twice as much to acquire a customer and came in below 1:1 on the first purchase, so any team optimizing on ROAS alone would have cut it immediately. Product A recovered its acquisition cost in two months, half the time of the product that looked twice as healthy on the dashboard, because those customers came back on a much shorter cycle, and they stayed around three times longer. Over the life of a customer it returned orders of magnitude more margin than Product B. A single-metric read would have killed the most profitable segment in the business. The full case is documented in [the strategic initiatives article](/insights/why-strategic-initiatives-fail#return-on-ad-spend-roas).
In practice
Paired metrics means no metric is read alone. A team can know its customer acquisition cost and its lifetime value and still not know its payback period: how long it takes to make back the cost of acquiring a customer. CAC alone reports what a customer costs, and LTV alone what a customer is worth; read together with timing, the pair shows whether the business can afford its own growth. Payback is measured in months of contribution margin rather than revenue: a customer generating $100 a month at a 30% contribution margin repays $300 of acquisition cost in ten months, while the revenue-based version reports three, understating the wait by exactly the margin structure. At a publicly traded company, a 2% dip in organic traffic came up in a QBR shortly after a website restructuring investment of over a million dollars. Four teams, website, strategy, marketing, and technology, were pulled in to find the cause. Cross-channel context resolved the question: paid advertising had increased 300% that quarter, and total website traffic was up 800% year over year. The 2% was noise.
Where we come in
We build paired, shared measurement systems for a specific business model. At Old School Labs, the cohort with the lowest ROAS in the business, 0.6, was the most profitable, because its margin was high and its customers returned. The Potbelly turnaround followed this structure: efficiency was proven at level budget before spend was scaled. With Equinix and Digital Realty, two companies that look identical on paper, we built separate measurement strategies because their models and go-to-market motions differ.
Start a Revenue Health Pre-Assessment →See it in action
Related terms
- Key performance indicators (KPIs)
- Quantified measures of performance against a goal, such as CAC, ROAS, or repeat purchase rate.
- OKRs
- Objectives and key results, which often track activity and output completed without measuring revenue impact.
- Leading indicators
- Early signals that appear months or years before revenue moves, such as a rising CAC.
- Lagging indicators
- Outcome measures that confirm results after the fact. Revenue is a lagging indicator, the last number to move.
- Paired metrics
- Two or more connected metrics read together, so a change in one is judged against the context the others supply. A single metric reports what happened; the pair supplies why.
- North star metric
- A single metric a team elevates above all others. Without other metrics as context, it has the same isolation problem as any single KPI.