Advanced technology products deserve a modern UX

A better user experience is required in an advanced technology product.

Design is simply outdated and poor. User Experience must be improved

BR

Thanks for the feedback! Design can be subjective, so it’d help to know what specifically feels off to you

yes :slight_smile: it’s subjective…

So i will try something… I’ve captured some screenshots and shared with gemini, and ask it to make an analisys…

here it comes:

Certainly! Here is the UI/UX analysis of AIEndurance.com translated and refined into English.

UI/UX Analysis: AI Endurance Dashboard

Based on the screenshots, the platform feels like a powerful data engine trapped in a 2015-era web interface. It prioritizes function over form, which is common in ā€œhardcoreā€ sports science apps, but it lacks the visual polish of modern competitors like Strava, TrainingPeaks, or Athlytic.


1. Visual Design & Aesthetics (ā€œThe Outdated Lookā€)

The ā€œoutdatedā€ feeling you’re picking up on comes from a few specific design choices:

  • Color Palette & Gradients: The heavy use of saturated purples and blues, combined with linear gradients (orange-to-pink or blue-to-violet) on the workout cards, was a major trend in the mid-2010s. Modern design has moved toward flatter, more muted tones or ā€œglassmorphismā€ (semi-transparent layers).

  • Iconography: The icons (like the stick figure stretching in ā€œStretching & Mobilityā€) look like ā€œclip-art.ā€ Modern fitness apps typically use ā€œline-weightā€ icons (thin, minimalist strokes) that look more professional and less like stock illustrations.

  • Shadows and Depth: The cards use heavy drop-shadows to create depth. Current UI trends (like Material Design 3) use subtle tonal elevation or simple borders to separate sections.

2. Information Architecture (The ā€œData Dumpā€)

The biggest issue is Cognitive Load. The app tries to show you everything at once without a clear visual hierarchy.

  • Dashboard Clutter: In the main dashboard (Image 2), every widget (Aerobic Recovery, Muscle Recovery, RMSSD, HR Rest) has the same visual weight. It’s hard for the eye to know what the ā€œmost important numberā€ is for today.

  • Data Density: In the Recovery Details (Image 4), the overlapping red and blue area charts are visually ā€œnoisy.ā€ It’s difficult to quickly distinguish trends because the fill colors are too opaque.

  • The ā€œN/Aā€ Problem: Displaying ā€œN/Aā€ in bright red text (Image 8) creates a sense of ā€œerrorā€ or ā€œfailureā€ even when it simply means data hasn’t been synced yet.

3. Navigation & Interaction

  • Desktop-First Header: The top navigation bar is a very traditional ā€œwebsiteā€ header. Modern web apps often use a collapsible sidebar to give more vertical space for the actual training data.

  • Form Controls (Sliders): The plan configuration screen (Image 3) uses very basic HTML-style sliders. A modern interface would show a real-time ā€œpreviewā€ of how moving those sliders changes your predicted fitness or fatigue levels on the same screen.

  • The ā€œDeleteā€ Grid: The list of RMSSD values (Image 5) is highly repetitive. Having a ā€œDeleteā€ button visible for every single entry creates a cluttered look. Usually, these actions are hidden behind a long-press or a ā€œswipe-to-deleteā€ gesture.


4. Strengths (What works well)

Despite the dated look, there are several UI wins:

  • Calendar Color-Coding: The use of distinct colors for different disciplines (Run, Bike, Swim) is very effective for ā€œat-a-glanceā€ planning.

  • Digital Twin Concept: The comparison between ā€œActualā€ vs. ā€œDigital Twinā€ (Image 6) is a unique selling point. Even if the graph is a bit clunky, the concept is communicated clearly.

  • Compliance Heatmap: The ā€œ365 Day Historyā€ (Image 8) is a great way to visualize consistency, inspired by the GitHub contribution graph.

Summary Checklist for Improvement

Feature Current State Modern Recommendation
Typography Generic Sans-Serif Use variable fonts with better weight contrast.
Gradients High-contrast linear Subtle ā€œgrainyā€ gradients or solid flat colors.
Layout Boxy, shadowed containers Rounded corners, more ā€œwhite spaceā€ (breathing room).
Navigation Top bar (Static) Side-bar or Floating Action Buttons (FAB).

Said that, there are also some more specific UX improvements that can be included:

  • Detailed analysis should include the planned thresholds, compiance….

  • Paired activities could be shown as one activity (Plan + Execution)

  • All activities should have ā€œanalysisā€ view. This activity can be previewed in the dashboard, but I cannot open it

Just a couple of examples

BR

I’m not getting any prediction…. why?
The app should explain me why I’m not getting it

wow that is super actionable, thank you so much!

TBH I never really liked the deeper purple tones myself, hard to put your finger on it though, this is great.

Good stuff, we actually use a Material UI and modern web frameworks so maybe by making a few small tweaks we can improve a lot

we could make the link of activities vs planned workout more explicit, agree.

Do you have a Runner account? If yes and you’re doing a lot of Cycling maybe worth switching over to ā€˜Triathlete’ account. If the activity doesn’t have power we don’t show the ā€˜View Analysis’ button.

Looks like your plan is still being created or if it’s been more than a few minutes please trigger the plan search again.

No triathlete profile… in reality my goals are usually running races, but i usually do some indoor cycling and casual swimming.

yes my issue is that I don’t use power at cycling… but why not show the analysis? At least HR

Yes, it seems that my last run this afternoon has triggered new power zones… and a new plan is required… it’s fine for me… my marathon goal if very far, so I’m not obsessed with each session planned and im flexible

1 Like

You’re not wrong. Do you have any rides with both power and heart rate on some platform? If yes you could upload them either via syncing that platform or Bulk Fit file upload in Apps page. Once we do we automatically train model to estimate power from heart rate: Calculate cycling power without a power meter | AI Endurance

The reason we don’t show anything for rides without power if you don’t have the power from heart rate model is that it seems to be very much an edge case since after all the cycling workout instructions, predictions etc are in power so mainly you benefit from the platform with at least some power data.. but maybe wrong

it makes sense…. power is the main driver for AI Endurance…

but if I cannot see the analysis (or just the detailed view) AI Endurance cannot be my main platform.

Anyway… I will do more testing, since I want to give a longer try

1 Like


The pairing looks nicer now. Also improved a few other little UX things here and there

Have to say I agree on the UI, it’s just not that nice, I’m sure you could get a much more modern, polished look and feel with a bit of help from AI! I also feel that some of the data displayed is too much - my feeling is that you could simply some of it but still allow people to click through to see the detail if they want. Recovery is a prime example - the dashboard shows six recovery scores but I really just want one that tells me if I’m good to go or not (like red/amber/green), you could move all the detail onto the specific Recovery page. I’m an engineer and a geeky triathlete so normally I like lots of data but this is too much for me. Some descriptions/wordings could be a little clearer too but that’s a minor niggle (like Digital Twin I assume actually means ā€œPredictedā€)

1 Like

yes definitely planning to have a go at the design. With the data it’s a tricky one, it’s also our identity to explain things properly to you and we don’t want to water that down. I feel like there’s enough products out there where you get a ā€˜score’ and you have no idea where it comes from. I get your point that we can likely do some small changes and have both though

2 Likes

Don’t get me wrong, I like the data but I don’t always have the time or inclination to dig into it, hence why a simple recovery score that also gives a clear go/no-go is what I’d like to see as the headline, especially when I wake up first thing in the morning and I need to decide if or what I’ll do training wise that morning. It would also help broaden the appeal of your platform for people who are less knowledgeable on training and HRV etc. etc.

1 Like