MEDA1302 Chap.8 Evaluation and Campaign Measurement
Evaluation and Campaign Measurement
This chapter matches evaluation evidence to the level of change promised by an objective. Activities describe what the team does; outputs describe material distributed or attention generated; outtakes describe what publics notice, understand or remember; outcomes describe changes in attitude, intention, behaviour or relationship.
Strong evaluation establishes a baseline, repeats a suitable measure, considers alternative explanations and writes recommendations from evidence. It reports metrics by their real names instead of renaming reach or clicks as trust, understanding or success.
An evaluation table should place the objective verb beside its indicator, public, baseline, follow-up, method, owner and timing.
This makes mismatches visible before launch. Collection conditions need consistency: changed wording, different recruitment routes or missing cases may weaken comparison even when the indicator is valid. Activity and output data remain useful for diagnosing delivery, but they do not inherit the name of an outcome. Mixed results support more precise decisions.
Strong reach with weak recall can indicate message clutter; strong understanding with weak behaviour may point to service friction. Qualitative evidence can explain experience but should not be reported as population prevalence. Price, capacity, season, partner activity and other influences must be considered before attributing movement to communication.
A recommendation states what should continue, change or stop and identifies the evidence and limitation behind that choice.
What this chapter covers
- 01
Activities and outputs
- 02
Outtakes and communication understanding
- 03
Attitude, behaviour and relationship outcomes
- 04
Indicator validity
- 05
Baseline and follow-up design
- 06
Alternative explanations and recommendations
Worked example · free
Audit a campaign dashboard
- +1Classify impressions and clicks as distribution or response measures.
- +1Recognise bookings as behaviour that may include people outside the public.
- +1Repeat the objective's knowledge question before and after communication.
- +1Monitor service and seasonal changes that could affect behaviour.
Key terms
- Activity
- An action performed by the communication team.
- Output
- A communication product, distribution result or immediate response generated by activity.
- Outtake
- What a public notices, understands, remembers or takes from communication.
- Outcome
- A change in attitude, intention, behaviour or relationship.
- Validity
- The degree to which an indicator represents the concept it claims to measure.
- Confounding factor
- Another influence that offers an alternative explanation for an observed result.
Evaluation and Campaign Measurement FAQ
Why are impressions not enough?
Impressions estimate potential display or exposure, not attention, understanding, trust or behaviour. They may support a reach account, but evaluation must use a measure that matches the objective. State the metric accurately and resist converting a convenient dashboard count into a stronger claim.
What is the difference between an outtake and an outcome?
An outtake concerns what people noticed, understood or remembered from communication. An outcome concerns a later attitude, intention, behaviour or relationship change. Correct recall of a timetable is an outtake; completing a booking is a behavioural outcome. Both may be useful when the campaign theory connects them.
How do I write an evaluation recommendation?
State what should continue, change or stop and cite the finding that justifies the decision. Mixed results are informative. Strong reach with weak recall may require a clearer message, while strong understanding with weak action may expose service friction rather than a communication problem.
Assessment move
Build a four-level ledger for every tactic: activity, output, outtake and outcome. Then place the objective beside the ledger and circle the measure that tests the same level. Practise identifying overclaims in fictional dashboards. For each, rename the metric accurately, add a better method and identify one confounding factor. Write one recommendation from mixed results rather than treating evaluation as a success summary.
This prepares the campaign report and gives concise material for Q&A defence. Create an evaluation table before designing the final dashboard. Include the objective verb, public, indicator, baseline, follow-up, collection method, owner, timing and likely alternative explanations.
If the objective promises knowledge, write the question that will test knowledge; if it promises trust, specify a consistent attitude measure rather than substituting coverage or clicks. Keep activity and output data because they can diagnose delivery, but do not let them inherit a stronger label. Practise with mixed fictional results.
Strong reach with weak recall may reveal message clutter, while strong understanding with weak action may reveal service friction. For each pattern, write one finding, one limitation and one decision about what should continue, change or stop. Check comparison quality by holding wording, population and collection route as stable as possible.
If different people are sampled at baseline and follow-up, describe group movement rather than claiming each individual changed. Add qualitative evidence only for a defined purpose, such as explaining why a booking route failed, and resist converting a few comments into prevalence. Finally, identify one outside influence such as price, capacity, season or partner activity that could produce the result.
State what additional evidence would help distinguish explanations. This practice makes evaluation a disciplined claim about learning rather than a list of favourable numbers.
