Hola fellow designers ✋
I'm Dimitris Raptis, Co-founder and designer of VisualEyes! My team and I are thrilled to announce our design tool!
Currently, the design teams are struggling to survive under the countless screen iterations. Thousand design decisions are in doubt or taken under a non-clear context. Traditional A/B testing is a credible way to choose between your variants but takes time and effort to conduct it. We live in the attention era, but our design processes are full of distraction!
That's why we created VisualEyes!
The first A.I design assistant that can predict the attention in your designs. The algorithm is powered by Artificial Intelligence and neuroscience research on real eye tracking data from our platform and respective open datasets.
Available plugins for Figma, Sketch, and Adobe XD!
Today we're on Product Hunt!
So instead of testing with real users or data you suggest making design decisions based on generated data? I'd like to see comparisons of what this generates vs. real eye tracking data for the same designs to see how close it gets. Without that context this doesn't seem like a good idea.
You can check out our accuracy statement here.
The tool shouldn't necessarily replace your traditional user research process. You cannot run an A/B testing for each small design tweak, that's why some decisions are taken under non-clear context. You can use VisualEyes for small changes like these and a classic A/B test for the final vital decision.
It seems accurate enough at the site level, but if you're talking about using this for small changes I don't think it gives you that level of fidelity.
Now we develop a new feature called Areas of Interest.
It will help you quantify the heatmap results and gives back a percentage(%) of the total heatmap that is inside the specified area.
This should show you how much your tweaks affect the area driven by numbers.
Would you interest in Beta testing this feature? It should be ready by the end of this week.
This was my initial reaction to this a while ago but I'm willing to give it a chance as a tool that works in addition to real user tests.
I'd be really interested in your feedback