Build time per test case

What it means

The person-time to take a test case from a requirement to a passing, trustworthy automated test in a particular tool: reading the scenario, building the steps, wiring up test data, and running it until it is green for the right reasons. It is entered per vendor because tools differ enormously here.

In Detailed mode there is one figure per complexity tier; in Simple mode one average stands in for all three. Each tier also carries a scratch / migration mode: migrating an existing case typically takes 40–60% of the time of authoring it from scratch.

Why it matters to the cost

This is where codeless and AI-assisted tools make their case, and where a framework you build yourself pays its bill. At 300 cases, the difference between an average of 15 minutes and 4 hours a case is roughly 75 hours against 1,200 — at the default loaded rate of about $67 an hour, that is $5,000 against $80,000 before anything has run. Measure it in a proof of concept on your own application; it is the most valuable number you can bring to an evaluation.

Typical values

By category, for an average case: AI-native tools around 15 minutes; codeless enterprise tools 30 minutes to 2 hours; script-based tools around 2 hours; a homegrown framework around 4 hours; manual case authoring about 30 minutes.

Typical values describe a category of tool — AI-native, codeless, script-based, homegrown, manual — never a named product. Why we do not price named products.

Where it appears in the model

Every name is a link to its definition. See the whole model.