Current problem: Our free-text search results aren't working sufficiently; for instance, a search for "living room lamp" displays a product that contains "Living" as a brand-line name, which is neither a "living room" item nor a "lamp."

The goal: As a first step towards improvement, we aim to update only the UX/UI without coding. Based on the data, users primarily use free-text search, and categories are the most used filters. Selecting categories could lead to the closest results, so we'll try to allow users to select the closest categories, not only search by free text.

Notes: There are 23 categories and sub-categories, and the numbers depend on the selection. Users can choose multiple categories/sub-categories and combine them with other filters.

Question: I've created a mid-fidelity prototype in Figma, and I would like to test if this prototype provides a better experience in finding the right product. Although it's impossible to set up all events and flows in Figma due to numerous categories and possible flows, I could limit the user's navigation or, in a worst case, use dummy categories. Would this be enough to test and obtain validated results? Additionally, if there is any platform for conducting complex search testing without coding, I would like to know.

Thanks in advance!

  • 1
    This doesn't sound like a prototype is the primary thing to test, tbh. The issue is the existing code doesn't work in the way you want. But if the code is already built then you should probably build the alternative UI front-end and do some usability testing with that new version. Because what if the figma version works great for users but the actual coded version of it works differently? What would the prototype testing have proven?
    – JonW
    Commented Dec 18, 2023 at 15:39
  • I second @JonW. Unless you have a completely different UI, this will require testing the coding side, so I can't see how to do it without coding. Maybe if you post the Figma mockup we can see if it's possible to change the logic
    – Devin
    Commented Dec 18, 2023 at 16:40
  • Btw, the category might be the most used filter because your search isn't tuned to what people are looking for. I'm looking for a living room lamp, the wrong results show up, so I use the categories to try and narrow down to interior design items. Sounds like the flow your users might be going through. Commented Dec 18, 2023 at 19:12

1 Answer 1


To your question, I find that Axure is significantly better for high-fidelity prototypes than any other solution in the market (having used many of the prototyping tools that are out there). It supports building a working prototype that would match your described scenario (a categorized and searchable dataset). However, the biggest issue you described is not one that a prototype would solve on its own - if your existing search returns results that are not useful for your users, then it seems like you need to look into optimizing the search tool before exploring other areas. For that, analyzing logs so you can see the correlation between search phrases, search results, and how your users used those results, would help you way more than a prototype.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.