Guide7 min readUpdated
Heuristic Evaluation: What It Is and How to Run One
Heuristic evaluation is an expert review of an app or website against usability principles. Learn the principles in plain words and how to run one.
By Tithi Mondal, UI/UX Designer
Imagine a founder says her fintech app loses people at the KYC step. A reviewer opens the app on a phone and goes through it slowly. On the document upload screen, the button says "Proceed". There is no message that the photo was received. After tapping, nothing seems to happen for six seconds. Two usability principles are broken on one screen, and no user was needed to see it. That kind of review is called a heuristic evaluation.
A heuristic evaluation is an expert review in which a person checks each screen of an app or website against a list of common usability principles and writes down every problem. A "heuristic" is simply a rule of thumb. The review is quick because it needs no test participants.
What is a heuristic evaluation in simple words?
A heuristic evaluation is a structured expert review of a product, done against a fixed list of usability rules. You can also call it an expert review of an app, or a usability checklist review.
The reviewer walks through key tasks, such as signing up, searching, paying or booking. At every screen the reviewer asks: does this follow the principles? Each problem is written down with the screen name, the principle it breaks, and how serious it is.
This is different from usability testing, where real people try the product and you watch them. In a heuristic evaluation, an expert uses knowledge and experience in place of users. For a wider review that also covers content and speed, see the UX audit guide on this site.
What are the usability heuristics, in plain words?
The usability heuristics are about ten widely used principles for judging whether a screen is easy to use. They are written in many forms, but they carry the same ideas. Here they are in plain language, each with an example.
- Show what is happening. The app should always tell the person where they are and what it is doing. Example: a progress bar during KYC upload, and a message "Photo received".
- Use the customer's own language. Words and symbols should match what the customer knows. Example: "Pay by UPI" instead of "Initiate transaction".
- Give people control and a way out. A person should be able to go back, cancel or undo. Example: a "Cancel order" option before the order is accepted.
- Be consistent. The same thing should look and work the same way everywhere. Example: the main button is always the same colour and in the same place.
- Prevent errors before they happen. Make the mistake hard to make. Example: a phone number field that accepts only 10 digits and shows a number keypad.
- Let people recognise, not remember. Show options instead of asking people to recall them. Example: show saved addresses instead of asking for the address again.
- Work for new and experienced users. Beginners need guidance, and frequent users need shortcuts. Example: a "Repeat last order" button for regulars.
- Keep it simple. Remove information that is not needed on the screen. Every extra line competes with the important line.
- Help people recover from errors. Error messages should say what went wrong and how to fix it, in plain words. Example: "The OTP has expired. Tap here to get a new one."
- Offer help where it is needed. Short hints near the field are better than a long manual. Example: a one line note under "PAN number" that says where to find it.
You can add checks for accessibility, language and slow networks. My posts on accessibility in app and web design and designing for low-end phones and slow networks give good extra points.
Why does a heuristic evaluation matter for a business owner?
A heuristic evaluation matters because it finds many obvious problems quickly and at lower cost than a full study. You can run it on a live product, on a prototype or on a competitor's app.
It is useful when you have a drop in sign-ups or orders and do not know where. It is also useful before a redesign, to list what must be fixed, and before you spend on user testing. There is no need to pay for testing sessions to discover a missing back button.
The result is a list of problems with a severity rating, so that you fix the serious ones first. This is much easier to plan than a vague feeling that "the app is confusing". If you are deciding between fixing and rebuilding, app redesign or rebuild will help.
How do you run a heuristic evaluation step by step?
You run one by choosing the tasks, going through them screen by screen against the principles, writing each problem down and then rating how serious it is. A small app can be reviewed in one to two days.
- Pick 3 to 5 important tasks. For example: create an account, find a product, pay, track an order, contact support.
- Choose the device. Review on a real mid-range Android phone and on a laptop if the site is used on both. Most Indian traffic is mobile.
- Go through each task slowly. On every screen, check the ten principles above. Note the problem, not a fix, at first.
- Record each problem in a table. Use the columns shown below.
- Rate severity. Use a simple scale: high means people cannot finish the task, medium means they are slowed or confused, low means a small annoyance.
- Ask a second person to review separately. Two reviewers find more problems than one, because each person notices different things.
- Combine the lists and agree on priorities. Remove duplicates and put the high problems first.
- Suggest fixes. Write one clear fix for each high and medium problem.
Here is how one row of the table might look.
What does a good heuristic evaluation report contain?
A good report is short, specific and ordered by severity. The reader should be able to act on it the same day.
- The tasks and devices that were reviewed.
- A list of problems with the screen name, the principle broken and a screenshot reference.
- A severity rating for each problem.
- A suggested fix for every high and medium problem.
- A short list of what already works well, so the team does not undo it.
- A note on what the review could not tell you, such as whether real users want a feature.
What mistakes make a heuristic evaluation weak?
The biggest mistake is believing that an expert review replaces talking to users. It does not. An expert can spot a broken rule. An expert cannot tell you whether your customers want the product.
- One reviewer only. One person misses many problems.
- Reviewing without a task. Wandering around the app finds fewer problems than following a real task.
- Reviewing only on a laptop. Many problems show only on a small phone.
- Listing taste as a problem. "I dislike this colour" is not a heuristic problem. "The text cannot be read in sunlight" is.
- No severity. A list of 60 equal problems cannot be acted on.
- Only criticism. Note what works, so that it is not broken by mistake.
Can a small project skip a heuristic evaluation?
Yes, a very small project can skip a full review. If you have a one-page site with one action, a quick check against the principles, using a short checklist, is enough.
It is most useful for apps and websites that already exist and are not performing. It is also useful before a redesign. If your product is brand new, a review of the wireframes can catch problems before any design is built. Read wireframe vs mockup vs prototype to see where that fits.
This kind of review fits the Understand stage of my process, before the flow is mapped or new screens are drawn. If you have a live app or website with a drop-off you cannot explain, you can see my UI/UX design service or write to me through the contact section.
Frequently asked questions
How many reviewers do I need for a heuristic evaluation?
Two or more is better than one, because each reviewer finds different problems. For a small app, two reviewers working separately is a good start.
What is the difference between a heuristic evaluation and a UX audit?
A heuristic evaluation checks screens against usability principles. A UX audit is wider and may also cover content, speed, accessibility and analytics.
Is a heuristic evaluation cheaper than usability testing?
Usually yes, because it needs no participants and no sessions. But it cannot replace testing with real users.
Can I do a heuristic evaluation on my own app?
You can, using a checklist, but you know the product too well and will miss problems. A fresh pair of eyes finds more.
Working on something like this?
Tell Tithi what you are planning. You get a clear scope, a fixed price and a timeline before any work begins. I reply to every enquiry within 24 hours.