"How did the campaign do?" is a simple question that can be surprisingly difficult to answer. Different platforms report different figures, objectives were never written down, and nobody is quite sure which numbers matter. Most of these problems can be avoided with a little planning before launch.
Agree the question you are trying to answer
Every campaign should have a primary objective that everyone understands. Is it to generate enquiries for a particular service, introduce the brand to a new audience, or encourage existing customers to return? The objective determines which measures are relevant. Without it, reports tend to include every available metric and explain very little.
Choose a small set of measures
It is useful to separate measures into three groups:
- Outcome measures that relate directly to the objective, such as qualified enquiries or sales.
- Leading indicators that suggest whether the campaign is on track, such as visits to key pages or engagement with important content.
- Diagnostic measures that help explain performance, such as click-through rates, cost per click or frequency.
Reports are easiest to use when they lead with outcomes and use the other measures to explain them.
Check the tracking before you spend
Test every conversion action you plan to rely on. Submit test enquiries, check that events are recorded once and only once, and confirm that the information reaches the systems that need it. Make sure your consent banner and tracking setup work together, so optional tools only run when visitors have agreed.
Understand the limits of the data
No measurement is perfect. Privacy choices, ad blockers, cross-device behaviour and differences in how platforms attribute conversions all mean that reported figures are estimates rather than complete records. Being open about this from the start helps set realistic expectations and avoids decisions based on false precision.
Where possible, compare platform data with your own records, such as enquiries received or sales recorded, to get a more rounded view.
Set a review rhythm
Decide how often results will be reviewed and what kinds of decision each review is for. Early reviews might focus on technical issues and obvious problems. Later reviews can look at trends and opportunities. Agreeing this in advance reduces the temptation to react to short-term fluctuations.
Write it down
A one-page measurement plan, covering the objective, key measures, data sources, known limitations and review schedule, is usually enough. It gives everyone a shared reference point and makes the eventual answer to "how did it do?" far more useful.
