I have a set of cards, in this case people. Imagine a dating site where the user wants to find people who are both near to them geographically and have a high match rating. I'm not sure what the most useful way to present them in a list is.

Ordering them by match alone is definitely a bad idea, because the best matches might be thousands of miles away. I don't want to just order by distance because I want people who are good matches and a reasonable distance to be a priority over bad matches who are very close.

I know this could be presented as an algorithm problem, but in this case I'm looking for how it might be effectively presented or UX workarounds to achieve the same effect.

2 Answers 2


I think the most important part here in terms of the UX is the freedom and control you give to the users and the information you give to them.

Give the user the feeling that they are in control and act according to their preferences. In this case it sounds to me that users are going to make a trade-off: this person matches 100% but lives far away vs. this person is very close to me but the match might be imperfect.

If the user has the ability to let the system know what the maximum radius is in which the system searches for matches, or what the minimum required match % is, the system can take these values into account and show matches that are preferred by the user.

Another solution is thinking in a regression, however, thinking in regression is fairly difficult for most people. But you could give your users the option to set attribute weights. For example, a slider or number value to indicate how important proximity is or how important a match is:

  • Proximity: 3
  • Match: 4

will result in: y = proximity*3 + match%*4. In this way a score is calculated and the results are ordered, but most importantly for your question, the user knows where this calculation comes from because the user set the attribute weights.

Finally, think about decision-making strategies to help your users make a deliberate decision (Bettman, Payne, Johnson are researches who identified a few; such as, satisficing, weighted additive, equal weights, lexicographic etc.)


In a card format style/layout, even if you tried to order by two different properties, it is very difficult for users to make the visual comparison. For example, if you want to order by compatibility match then you might present something on the top right hand corner as a number value. However, if you want to order by proximity then you might have to show a distance value on the top left hand corner. So while it is possible for users to visually scan one value or the other, it is difficult for them to try and scan both easily.

This is one of the main design considerations for the card pattern, which is usually why a table design pattern is easier for comparison of multiple values that you may want to filter/sort/search by.

The alternative is to have the filter/sort/view selected by the user first (e.g. compatibility), and then order by the second value (e.g. distance).

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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