A practical step-by-step guide

How to conduct your own UX expert review

Your team can conduct a useful expert review internally, particularly when it knows the audience, business goals and service well. This guide explains how to define the scope, review the user journeys and turn observations into prioritised improvements.

10 min read

Before you start: define the scope

Do not begin by “reviewing the entire service”. Choose one clearly defined part of the service and one or two user journeys that matter to both the user and the organisation. This might be creating an account, finding the right product, submitting an application or completing a purchase.

Write down the following before the review begins:

  • who the user is and what they are trying to achieve
  • where the journey begins and when it is complete
  • which devices and screen sizes are relevant
  • which business goals the journey should support
  • which constraints, roles or exceptions need to be considered

Prepare realistic test data

Use realistic data, accounts and situations. An empty demo account may hide problems that appear only when the history is long, roles have different access levels or the user has already completed part of the task.

Prepare at least one normal situation and several deviations. What happens if a field is completed incorrectly, the connection is lost, the search returns no results, a payment is declined or the user goes back halfway through the process? A digital service must remain understandable when everything does not go according to plan.

Work through the service as a new user

Put internal knowledge aside as far as possible. Read only what is actually on the screen, and do not use shortcuts you know because you helped build the service. Stop at every step and ask:

  • Do I understand what this page or step is helping me do?
  • Is it clear what I can do and what will happen next?
  • Does the service use language and concepts the audience knows?
  • Do I receive confirmation when an action is complete?
  • Can I correct an error without losing work or starting again?
  • Is important information visible before I have to make a choice?

Repeat the journey on mobile when mobile use is relevant. Do not merely narrow the browser window: use touch, the on-screen keyboard and the actual space available on a mobile screen.

Use principles as support, not as the final answer

A set of usability principles gives the review structure. Nielsen's ten heuristics are a good starting point, but you can also use established patterns, accessibility requirements and your own design principles.

Do not record a finding simply because the interface deviates from a rule. Always explain which specific task becomes more difficult, who is affected and what the consequence is. Context determines whether the deviation is a problem.

Learn more about the principles in What is a heuristic evaluation?

Document every finding in the same way

Use a consistent template so that findings can be compared and handled as work items. One finding should describe one problem, not an entire page of loosely related observations.

  1. SituationWhere in the journey does the problem occur, and what is the user trying to do?
  2. ObservationWhat in the interface is unclear, missing or behaving unexpectedly?
  3. ConsequenceHow might the problem affect the user's understanding, trust or progress?
  4. RecommendationWhich specific changes could resolve or reduce the problem?
  5. PriorityHow important is this finding compared with the others, and why?
Example of a documented finding

The user receives no confirmation that the change has been saved.

Consequence: The user may press the button repeatedly, leave the page in doubt or contact customer service to verify the result.

Recommendation: Show a clear confirmation in the same context as the action. Also update the button state while saving is in progress.

Prioritise by consequence, not irritation

Not everything that looks unfortunate is important. Prioritise findings by the severity of the consequence, how often the situation occurs and how central the task is. A small ambiguity in the payment step may matter more than a more visible design problem on a rarely visited page.

  • Critical: The user cannot complete an important task or may make a serious incorrect choice.
  • Important: The problem creates significant uncertainty, extra work or abandonment.
  • Should be improved: The problem weakens the experience but does not usually prevent the task.

Write one short reason for each priority. The reason, not the label or score, makes the prioritisation useful to the team.

Ask someone else to challenge the assessment

Review the findings with a colleague who did not take part in the evaluation. Ask them to check whether the problem can be reproduced, whether the consequence is reasonably explained and whether the recommendation resolves the underlying cause.

The team should also distinguish between what you know and what you assume. If a finding depends on an assumption about the audience's needs, language or behaviour, mark it as a hypothesis and test it with people from the target audience.

Turn the report into action

Finish with a short presentation for the people who will decide, design and build. Begin with the most important journeys and the most serious findings. Discuss technical constraints and decide which recommendations should become work items.

Preserve the screenshot, consequence and reasoning when the finding is moved into the backlog. If only a short task title remains, the insight that made the work important is often lost.

When is an outside perspective worthwhile?

An internal review has one clear advantage: the team knows the service, audience and constraints. The same knowledge is also a weakness. It is difficult to see which explanations, concepts and shortcuts a new user does not know.

An outside perspective is particularly useful when the service is business-critical, the team disagrees about priorities, a major redesign is approaching or you want to check the internal review before important decisions are made. But do begin yourselves: a systematic internal review is far better than leaving known problems untouched.

Professional foundation

The assessments draw on more than 15 years of experience with digital products, established interaction design and usability principles, and the specific context of each service. See also ISO 9241-11 on usability and Nielsen Norman Group's heuristic evaluation workbook.

Do you have a service I should review?

Request an expert review

Briefly tell me which service you would like reviewed and which questions you want answered. I will reply with a proposed scope, delivery time and price.

Keep reading