Summative Evaluation
Research conducted at the end of a development cycle to measure the finished product's success against defined criteria. The goal is assessment, not iteration.
Definition: Research conducted at the end of a development cycle to measure the finished product's success against defined criteria. The goal is assessment, not iteration.
Summative evaluation is research conducted at the end of a development cycle to assess how well the finished (or near-finished) product meets its goals.
When It Happens
Summative evaluation occurs as you approach a release—with a high-fidelity prototype or Minimum Viable Product (MVP). This is the final quality check before launch.
Purpose
The goal is assessment: measure performance against benchmarks or success criteria.
| Aspect | Summative Evaluation |
|---|---|
| Timing | End of development |
| Goal | Measure success |
| Question | "How well does it perform?" |
| Sample size | Larger, for statistical confidence |
| Outcome | Go/no-go decision, benchmarks |
Limitations at This Stage
By the time you reach summative evaluation, fundamental changes are often prohibitively expensive. You can typically only inform "cosmetic" or component-level changes—not save a product built on a flawed foundation.
Metrics Often Used
- SUS scores
- Task success rates
- Time on task
- Error rates
- NPS
- Comparison to benchmarks or competitors
Related Terms
Formative Evaluation
Research conducted during development to find problems and improve a design-in-progress. The goal is to shape and refine, not to measure final quality.
Evaluative Research
Research that assesses whether a specific solution works, either during development (formative) or after completion (summative). Answers 'Does this work?' rather than 'What should we build?'
UX Test
A Core Method combining all three Building Blocks: testing task completion (effectiveness and efficiency), observing behavior and non-verbal cues, and asking questions about the experience. The most comprehensive single research method.
Mentions in the Knowledge Hub
This term is referenced in the following articles:
Research Timing and Team Foundation: When to Research and Who Does It
One of the most common points of friction is not about budget or methods, it is about timing. Your core job is to reframe research from a single, disruptive event into a continuous, value-adding loop.
When to Research: A Guide for Product Teams
Don't wait for the beta. The 3 critical moments to test: Concept (Generative), Prototype (Formative), and Live (Summative).