Sports assistant
Developed a product card feature that allows users to  ask any question about a product or choose from a  list of popular and relevant questions
Goals
🎯
Simplify product selection β€” provide quick and clear answers instead of complex specifications
Build trust β€” show that the product truly matches the user's needs
Save time β€” gather all the necessary information directly on the product page
Boost sales β€” increase purchase confidence and suggest complementary products
Pain points
🚧
Business objectives
  • High support load β€” numerous repetitive questions about product usage and compatibility
  • Lack of differentiation β€” competitor product cards look the same
  • Underutilized content β€” expert guides and blogs exist but are not linked to product pages
  • Missed upselling opportunities β€” no smart recommendations for complementary products
Customer pain points
  • Hard to choose β€” specifications are complex, users don't know what matters for their case
  • Low trust β€” descriptions feel like marketing, reviews are subjective
  • Information overload β€” searching across multiple sources takes too much time
  • No personalization β€” users want advice tailored to their goals ("I run on pavement in winter...")
  • Purchase hesitation β€” unanswered practical questions stop people from buying
Workflow
πŸš€
  1. Research
  • Analyzed user behavior on the product page (time on page, exit points, add-to-cart frequency)
  • Created a detailed user journey: from landing on the product page to interacting with the assistant
  • Conducted market and competitor research: studied how competitors use product recommendations and AI assistants, identified best practices and gaps
2.Prototyping and testing
  • Developed 3 interactive prototypes of the Sports Assistant with different launch scenarios
  • Conducted usability tests to evaluate ease of use, scenario comprehension, and suggestion relevance
  • Collected feedback to improve the UX
3.Design
  • Refined user interaction scenarios (suggested questions, free input, recommendations)
  • Prepared 3 widget interface options and placement variants within the product card (applied the brand color used for all AI-related blocks + added sparkles to maintain consistent AI logic throughout the app)
  • Lengthy design approval process with the stakeholder: they had their own vision for the widget, and some elements conflicted with our design system, requiring compromises (settled on the 3rd option)
4.Results analysis
  • Conducted A/B testing: product card with the widget versus the standard card
  • Collected and analyzed metrics: conversion, engagement, CTR on suggestions, purchase frequency after interaction
Workflow
πŸš€
  1. Research
  • Analyzed user behavior on the product page (time on page, exit points, add-to-cart frequency)
  • Created a detailed user journey: from landing on the product page to interacting with the assistant
  • Conducted market and competitor research: studied how competitors use product recommendations and AI assistants, identified best practices and gaps
2.Prototyping and testing
  • Developed 3 interactive prototypes of the Sports Assistant with different launch scenarios
  • Conducted usability tests to evaluate ease of use, scenario comprehension, and suggestion relevance
  • Collected feedback to improve the UX
3.Design
  • Refined user interaction scenarios (suggested questions, free input, recommendations)
  • Prepared 3 widget interface options and placement variants within the product card (applied the brand color used for all AI-related blocks + added sparkles to maintain consistent AI logic throughout the app)
  • Lengthy design approval process with the stakeholder: they had their own vision for the widget, and some elements conflicted with our design system, requiring compromises (settled on the 3rd option)
4.Results analysis
  • Conducted A/B testing: product card with the widget versus the standard card
  • Collected and analyzed metrics: conversion, engagement, CTR on suggestions, purchase frequency after interaction
Results
πŸ†
After the feature launch to production
  • The Sports Assistant implementation led to a 1.3% increase in conversion
  • Average order value grew by 8% thanks to personalized recommendations and upselling features
  • User engagement increased by 10%, with strong interest in interactive content and assistant suggestions
  • Automating answers to common questions reduced support inquiries by 15%
Results
The development of the Sports Assistant widget improved the product page experience by providing personalized recommendations and instant answers. As a result, we saw increased conversion, higher average order value, greater user engagement, and reduced support load. This feature not only strengthened user trust but also created a clear competitive advantage for the business
Conversion
+1,3%
AOV
+8%
Support load
-15%
Made on
Tilda