José Ramón Vergara
Share your situationES
Back to work
Own product · Digital commerce

Huella Limpia

A recommender that helps pet households choose a robot vacuum using clear, reviewable criteria.

Local prototype under validation

Role
Own product · Strategy and build
Stack
HTML · CSS · JavaScript · JSON · Node.js · Python
Real screenshot of the Huella Limpia prototype home page, with links to its comparison and recommendation tools
Functional prototype screenshot
ContextOwn product · Digital commerceBuiltFive-question recommender for eight eligible models
01

The problem

Pet households must compare hair pickup, carpets, self-emptying, consumables and electrical compatibility across many product listings. A quick recommendation is weak if the exact model or offer has not been verified.

02

The system

I built a catalog and a five-question recommender, with product pages and guides that explain the decision criteria. The content records sources and review dates; the prototype excludes uncertain models and hides stale prices. It remains under validation before launch.

HTML · CSS · JavaScriptJSONNode.jsPython
SignalSystemDecision
Own product under validation

A useful recommender knows when to hold back.

The prototype combines a short questionnaire, comparable product pages and editorial guides. It shows eligible models while excluding those whose identity or compatibility still needs evidence.

  1. 01

    Understand the home

    Five questions capture needs such as pets, carpets and maintenance. Answers narrow the options without asking for personal data.

  2. 02

    Compare with sources

    Product pages and guides explain features and limits, with sources and review dates. The catalog separates observed facts from items still to be confirmed.

  3. 03

    Protect the decision

    The site hides stale prices and keeps models with unresolved SKU or voltage questions out of the recommender. Launch waits for commercial checks and visual usage rights.

If you are building oneThis case documents a functional prototype, not sales or market results. A commerce recommender is only as reliable as the current offer and the evidence for the exact model.

Real screenshot of the Huella Limpia recommender with its first question about household pets
01Recommender · first question
03 · What I built
03

What I built

What the prototype can already demonstrate.

  1. 01

    Five-question recommender for eight eligible models

  2. 02

    18 product pages and eight guides in the prototype

  3. 03

    Two models excluded over identity or compatibility questions

  4. 04

    Stale prices hidden automatically

Does your product need to help people decide better?