A marketing KPI is a measure selected to judge progress towards a specific business objective. For a lead-generation business, qualified leads, acquisition cost and sales from those leads can be useful KPIs. For an online shop, order conversion, contribution after advertising and repeat purchases may be more relevant. Start with the decision you need to make, then define how the number will be calculated.
Six measures and what they can tell you
| Measure | Calculation | Use and limitation |
|---|---|---|
| Lead qualification rate | Qualified unique leads ÷ unique valid leads × 100% | Shows fit against your agreed criteria. It changes if those criteria change. |
| Cost per qualified lead | Defined campaign costs ÷ qualified leads attributed to that campaign | Compares the cost of suitable enquiries. State whether production and agency fees are included. |
| Lead-to-customer rate | New customers from a lead cohort ÷ leads in that cohort × 100% | Shows commercial progression. Give the cohort enough time to buy. |
| Customer acquisition cost | Defined sales and marketing acquisition costs ÷ new customers acquired | Supports acquisition-budget decisions. Match the cost scope and acquisition period. |
| Revenue ROAS | Attributed sales revenue ÷ advertising spend | Shows revenue per advertising euro or dollar. It does not deduct product costs or all marketing expenses. |
| Customer retention rate | Customers from the opening cohort still active at period end ÷ opening cohort × 100% | Shows how many existing customers remain. Define what “active” means for your business. |
In Google Ads, ROAS uses reported conversion value relative to advertising cost. That value need not always be sales revenue, so label the value you actually send. See Google’s explanation of target ROAS. A 4× revenue ROAS means four units of attributed revenue for each unit of advertising spend, not four units of profit.
Example: cheap leads can become expensive qualified leads
Consider two fictional campaigns with the same €2,000 advertising spend and the same qualification rules. Campaign A generates 100 valid leads, of which 20 qualify. Campaign B generates 60 valid leads, of which 30 qualify.
- A: €20 per raw lead; 20% qualification; €100 per qualified lead.
- B: €33.33 per raw lead; 50% qualification; €66.67 per qualified lead.
Campaign B costs more per initial enquiry but less per qualified enquiry. That is a reason to investigate B further, not proof that B is more profitable. Compare subsequent customers, margin and sales effort before shifting the whole budget.
Write a definition before building a chart
For every KPI, record the event, inclusion rules, source, period, owner and decision it supports. For example: “A qualified lead is a unique company enquiry accepted by sales because it matches our service area, has a relevant need and provides usable contact details. Duplicates, job applications and supplier pitches are excluded.”
Then write the counting rule. Are several contacts from one company one lead or several? Is qualification recorded on the enquiry date or acceptance date? Keep those choices stable or mark the date of a change. Otherwise an apparent improvement may simply reflect a new definition.
Do not mix a cohort with a calendar total
If 100 leads arrive in January and ten of those leads buy by March, that cohort has a 10% lead-to-customer rate by the March cutoff. Dividing all March customers by January leads produces a different and potentially misleading number because some March customers may have enquired much earlier.
Use calendar reporting for workload and spending, and cohort reporting when you need to follow the same group through a buying process. Mark incomplete recent cohorts rather than treating delayed sales as failures.
Keep the dashboard tied to action
Choose one main outcome and a few supporting measures. For a team trying to acquire suitable customers, a compact view could include qualified leads, cost per qualified lead, cohort close rate and acquisition cost. Add response time if slow follow-up is a suspected constraint.
Clicks, impressions and visits can help diagnose distribution or landing-page problems. They become distracting when presented as success without a connection to the objective. Record what changed, check the data collection and make one clear decision at each review: continue, investigate, adjust or stop.
