For a few weeks I've posted the individual organs of a one-man SaaS here: the alert pipeline that decides who gets an email, the tour-predictions table that grades itself in December, the one-PHP-file analytics dashboard. Not once have I shown you the patient.
So here it is: thedjcalendar.com - tour dates for \~100 of the world's biggest DJs, pulled automatically from Bandsintown and Ticketmaster, normalized and de-duplicated, rendered as server-side PHP. Founded 2015, resurrected this year, run by one person on $3/month shared hosting.
What's under the hood:
- Plain PHP + MySQL, no framework, no page builder. Every page server-rendered.
- Artist pages at clean URLs with a 30-second hover preview, tour map, and discography links.
- Tour alerts: save an artist, get an email when they announce a show near you. Guests can give a city or ZIP and the dispatcher geocodes it once, caches it forever, and applies the Haversine math at 25 miles. Yes, 40.2336 km. I refuse to round.
- A predictions scoreboard: 61 predicted tour announcements with confidence scores, graded automatically in December whether I like the results or not.
- Browse by genre, city, venue, or a /directory of every artist as crawlable cards.
- Zero JavaScript frameworks. The interactive bits are vanilla JS, and the analytics is the one-PHP-file log parser you've already read about.
The pitch for this post: I've written the war stories here and you've been a genuinely useful roast crowd before - a previous thread got me to move 1,402 backup files out of my webroot. Now I want the same energy pointed at the product itself.
Roast it specifically:
- The signup flow. Save favorites, get alerts. Too many steps? Too few reasons? Would you hand over your email, and if not, what's missing at the moment you'd decide?
- The artist page. It's the core object of the whole site. Does it earn a revisit, or is it a listing?
- The homepage. Can a stranger tell in five seconds what this is and why they'd care, or is it DJ-insider soup?
- The business reality. Nobody's paying me anything. Alerts are free. If you ran this, what's the first thing you'd test to make it earn its hosting bill?
What would make you actually subscribe to a tour alert for an artist you love - and what would make you bounce in five seconds? I'd rather hear it from this crowd than from a bounce-rate graph in March.