Designing Case Study Metrics: Choosing Numbers That Mean Something

Why vanity metrics fail on consultancy case studies

Most consultancy case studies lead with numbers that impress other consultants but leave prospects cold. A metric like "increased engagement by 47%" means nothing unless the reader knows what engagement was, how it was measured, and why it mattered to the client's business.

The goal of a case study metric is not to prove you are clever. It is to help a prospect picture the same result happening inside their own organisation. That requires choosing figures that connect to revenue, risk, time, or cost.

Start with the client's original problem

Before you select any performance figure, return to the brief. If the client came to you because a checkout flow was losing sales, the headline metric should be conversion rate or revenue per visitor, not a design award or a satisfaction score.

Use a simple filter when reviewing your data:

• Does this number map to a business outcome the client already tracked?
• Can a prospect verify the logic without a lengthy explanation?
• Would the client be comfortable repeating this figure in their own board meeting?

If the answer to any of these is no, the metric belongs in an appendix, not in the headline.

Pair a primary metric with a supporting metric

One number alone can look suspicious. Pair a commercial result with a behavioural or operational one. For example, "checkout completion rose 18% while average order value stayed flat" tells a more credible story than a single percentage.

"One believable pair beats five impressive but disconnected figures."

This approach also protects you when a client cannot share exact revenue data. You can highlight a directional change plus a process improvement, such as reduced support tickets or faster time to launch.

Use a table to show before and after clearly

Prospects scan case studies quickly. A compact table makes the transformation obvious without forcing them to read paragraphs of context.

Explain the method behind every number

A figure without a method is a claim. Add one sentence explaining how you measured it, over what period, and against what baseline. This builds trust and prevents prospects from dismissing your results as cherry-picked.

Finally, choose three to five metrics per case study and no more. A focused set of numbers that mean something will always outperform a long list that means nothing.

1. Identify the client's core business problem.
2. Select one primary outcome metric.
3. Add one supporting operational metric.
4. Show before and after in a table.
5. State the measurement method in one sentence.