Evaluation is a practice. Treat it like one.
The first time I sat with a 200-page evaluation report that nobody had read, I thought it was a writing problem. It was not.
The report was the deliverable a team had been told to produce, on a budget set before anyone asked who would act on it, when, or for what decision.
Evaluation is not a report. It is a practice. It has its own methods, audiences, ethics, and craft. Like content design, it fails when it gets treated as a deliverable instead of a discipline.
The artefact is what shows up at the end. The practice is the set of choices made long before the artefact exists.
I have watched organisations spend the same money on evaluation and get different things. Some produced a report. Others produced decisions.
The difference was not budget, methodology, or evaluator. It was whether evaluation had been designed for from day one.
What gets mistaken for evaluation
Content design gets confused with wordsmithing. Evaluation gets confused with reporting. Both confusions cost organisations the same way.
When evaluation is reduced to a report, it becomes a thing you commission, a thing you deliver, and a thing that sits in a SharePoint folder. When content design is reduced to a polish step, the same thing happens.
The deliverable arrives, nobody can act on it, and the discipline gets blamed for the design failure.
The mistake is not quality. It is category. Treating a practice as a deliverable means buying the wrong thing. It means asking "what does the report look like?" before asking "what decision is this evidence for?"
The hidden design choices
Every evaluation is shaped by four questions that usually get answered by accident. Who is this for? What decisions does it inform? When does it need to land? What evidence will be defensible to the audience that matters?
These are not research questions. They are design questions. They decide whether the evaluation gets used or filed.
In my experience leading content reform, the projects that produced lasting change were the ones where someone had answered these questions before procurement. The team knew which executive would read the findings, what they were deciding, and when.
Methodology followed. Where it did not, the evaluation was usually rigorous, but it did not fit the moment it was meant to inform.
What intentional looks like in practice
Practice means choosing on purpose. A practice-led evaluation makes its design visible. It names its audience, its decision points, its measures, and its risks of misuse before any data is collected.
In content design contexts, this looks like a one-page evaluation brief written before the work begins. It includes the question the evaluation is answering, the person or panel who will act on the answer, the format the answer will arrive in, and the date by which it has to land.
It also names what the evaluation will not do. Practices have edges. A scope that says yes to everything is a scope that will not survive contact with the work.
For Australian government teams, the Style Manual's guidance on structuring content applies here too. If a reader cannot tell what to do with the document, the document has not done its job, regardless of how good the underlying analysis is.
Why it matters for the people the work serves
There is a quieter cost to treating evaluation as a deliverable. The people whose service was evaluated rarely see findings that change anything for them.
A practice asks who is harmed when findings are unclear, and designs against that harm. A deliverable asks whether the report meets the brief. They are not the same standard.
The organisations I have seen do this well share one habit. Before commissioning the evaluation, they ask what the worst-case use of the findings would look like, and they design the framing of recommendations and the dissemination plan to prevent it.
That is a content design discipline. It is also an evaluation discipline. The boundary between the two is thinner than most teams treat it.
The practical question
If you commission evaluation work, or content work that includes evaluation, the question worth asking next is this: do you have an evaluation brief that names the audience, the decision, the format, and the date?
If not, the methodology you are paying for is not going to save the work.