Case study · Web + iPhone · My own product
Choosing an EV in Australia meant forty tabs and no straight answers. I wanted one place with every model, every incentive, and a way to get to a shortlist of three.
Decisions that mattered · 01
- A wizard, not a filter. Filters assume you know what you want. Most people don’t — they know their budget, their commute and whether they can charge at home. Eight questions and a scoring engine turn that into a ranked list with reasons.
- Instrument the wizard. Every answer is tracked server-side and in GA4. The admin report shows which questions people abandon and which answers cluster — which is how the scoring got better.
- Used listings, automated. A weekly Apify import from public listings, so the used-price reality sits next to the new-price fantasy.
- Independence. No manufacturer pays for placement. Sources are cited on every page.
Stack
PHP 8.5 with a custom front controller (no framework), MySQL, Material Design 3 web components, Chart.js. PHPUnit. Apify for the weekly used-listings import. GitHub Actions → cPanel deploy.
Outcome
Live at ev-finder.com.au with a public roadmap.
Screens · 02
