From Approved Brief to Live Page: The KPI Marketing Teams Forget 

Two teams say a campaign took three days to launch. One started counting when the request arrived. The other started after copy, design and approvals were complete. Their numbers look comparable. Their processes are not. 

Campaign time to live is useful only when the clock has an agreed meaning. Without that, a faster result may come from moving the starting line rather than improving the work. 

Start with the decision the metric should support 

What are you trying to learn? A marketing leader may need to know how much notice a campaign requires. A web owner may want to understand production performance after a usable brief arrives. A finance partner may care about the effort and external cost of repeat work. 

Use two elapsed-time measures rather than forcing one number to answer every question. Request-to-live begins when the business asks for the campaign. Approved-brief-to-live begins when the agreed minimum inputs are complete and approved. Both finish when the configured journey is published and passes its acceptance checks. 

The first reveals the whole experience of getting work done. The second focuses attention on execution after readiness. Keep them together so upstream waiting does not disappear from the report. 

Define the finish line in customer terms 

A page going online is not necessarily the end of a campaign launch. The button may point to an old form. The receiving team may get a request without the product being selected. The acknowledgement may promise a response nobody has agreed to provide. 

Choose a finish event that reflects the actual job: published page, correct content, working action, verified receipt, and agreed measurement. The specific checks will vary. Their definition should not change halfway through a comparison. 

For a content-only update, the finish line can be narrower. For a new enquiry journey, it needs to include the action. Compare like with like rather than treating every website task as one standard unit. 

Do not confuse elapsed time with labour 

Elapsed time is the duration between two timestamps. Active effort is the time people spend doing the work. Three people working together for one hour produce three person-hours of effort, not three hours of elapsed time. 

Waiting can overlap with active work elsewhere in the same launch. Record the intervals that block the next necessary step, but do not add every team’s waiting estimate to create a fictional total. A simple timeline is often clearer than a large spreadsheet. 

Use separate fields for: 

• Request received and complete brief approved. 
• Journey published and verification completed. 
• Active effort by role, including corrections. 
• The main blocking interval and its reason. 
• Rework, defects and any changed scope. 

Compare recurring work, not unlike projects 

A new website and a minor campaign variant should not share the same performance benchmark. Group tasks by meaningful complexity: routine text update, existing-template variant, new component or new integration. 

For each group, review several comparable examples where available. Show the typical result and the range, not only the fastest launch. Note urgent exceptions and missing data. A rushed launch that requires extensive fixes may be fast on the clock and expensive in practice. 

These marketing operations metrics are especially helpful when the team is deciding what to standardise. Frequent, bounded tasks with long waits are different candidates from rare, genuinely novel work. 

Pair the clock with a guardrail 

A campaign execution of KPI should not reward publishing incomplete work. Pair time to live with one quality condition, such as correct updated instances or valid test-enquiry receipt. Add customer-team effort when a supplier is involved. 

A useful review question is: did this comparable launch become quicker without shifting more work to another team or lowering quality? That question is more demanding than celebrating a shorter build phase. 

Start with a prospective log if historical timestamps are unreliable. A clearly defined measure collected from today is worth more than a precise-looking estimate assembled from memory. The goal is not to prove that the team was slow. It is to make the next improvement visible.