An Angular.js/Ruby on Rails-based tool for benchmarking ProductHunt's products using Stanford natural language processing for sentiment analysis.
In self-balanced systems like Product Hunt or Reddit, sometimes, when the products are similar (they have a similar number of upvotes), it's unclear which product is better. PHBEN.CH is using Google Cloud Natural Language to evaluate customers' emotional response. The final score (ranged from 0 to 100) reflects the overall ProductHunt performance of the product. A site visitor can pick several products and review all stats available. Each time benchmark button is used, a separate page is created boosting the SEO ratings up.
We got involved with this project at the very early stage and, starting from the napkin drawings finalised it into code-ready set of Sketch files, delivered via Inspect InVision tool.
The simplest wireframes contained just a single search field, which later grew up into a full-scale mobile-optimized clickable web app prototype.
Sketch was used as the main designing tool. A quick and simple branding was put together along with the general guidelines on fonts, grid, and logo-use restrictions.
Prototype shared via the InVision web app.