Currently, I redesign a data-driven platform and what is confusing is a presentation of data on a comparison view. As you can see the bar's size is relative for each value (rows), but not relative to overall data. Each dimension (row) has a different total amount so that's also not relative /proportional in comparison to another dimension(rows). I'm concerned about how to visualize data to make it more clear for the user. Currently for instance in the example below the value of 3 is more than double the size of a chart with 8 replies. So it's totally confusing...example


  • so... each row has 5 visualizations; those are related to each other, but not the row below it, correct? What is being compared in this case?
    – Mike M
    Oct 15, 2020 at 14:30
  • that's a rating. on the example, above you see a 5-star rating, another chart has ie. 10 point rating. you compare the ratings between ie, departments/ offices in different countries. That's why important is to compare them. What's important each of them has a different amounts of replies/ answers. that's what creates confusion. For the example above ie. number 3 on 5th row is bigger than number eight on 3rd row. I thought to maybe compare them by percentage overall and additionally on a side add the amount of answers/ replies? Oct 16, 2020 at 5:50
  • Is this statement correct?: 4 people gave 1 start to Dimension 1
    – Nicolas
    Oct 17, 2020 at 8:41
  • Also, what is the relationship between Dimension 1 and Dimension 2? Does it matter?
    – Nicolas
    Oct 17, 2020 at 8:41
  • Nicolas Hung. Yes, indeed you are right :) I have also 10 point scale to prepare :P The relationship between dimension one/two is that ie. it's different departments/offices which the user would compare. Oct 18, 2020 at 11:04

1 Answer 1


The challenge with data visualization design is often centred around how to present the key information as clear as possible without compromising the integrity and accuracy of the underlying data.

There is no easy answer to the question because often you want to combine one or more datasets and/or views of the data (for whatever reason), and you need to find the right trade-off between showing enough information for it to be useful, but not too much so that it becomes overwhelming.

I think the better solution is to break down the overall data visualization into smaller chunks, or to create some interactions that allow you to filter or change the views so that there is less information to process. This will also give you a cleaner design and more focus for the audience.

But you are probably working with a few different sets of requirements and constraints, so just keep in mind that it is about getting the right balance rather than something that is going to solve the problem 100% of the time.

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.