Sunday, February 22, 2015

Circular Square

A year or two ago I ran off on a tangent and made a website I thought would revolutionize grocery shopping. I called it Circular Square. Simple concept: give it your zipcode, and it will search the major chains for locations near you. It gives you their addresses, links to Google Maps, and links to their weekly sales circulars.

You can then search all those sales (or some subset) for a particular item that might be on sale. It will come back with everything matching your search. It was essentially a giant collection of scrapers with a clean  single-page AJAX front-end. It took a good bit of work to make each scraper, but I ended up with eleven or so, covering half the major chains around Nashville.

Got a shopping list, and want to know the best place to get each item this week? Problem solved. This site could save you half an hour of searching through fliers.

Oh, but it gets better: you would be able to log in and save searches! The system would then email you when something you wanted to stock up on went on sale. Love Blue Bell ice cream, but only willing to buy it if you catch the rare half-price sale? Just watch your email! And on top of that, it would search the coupon sites too! This thing was going to be beautiful. You can still go to the site and get an idea of how it went. (Saved searches didn't work yet, though.)

But I ran into a slight problem: the websites I was scraping change constantly. After a year or two without maintenance, only six of the eleven scrapers still give location results, some of those links don't work properly, and only two still give sale results. It's still pretty sweet for Aldi and Harris Teeter, though!

At the time I concluded that this wasn't going to be worth the effort to maintain. But I'm starting to reconsider. I can obviously make this site work. I even enjoy writing the scrapers, it's a fun puzzle to solve. But the real problem is that it has to be worth my time. I see a few possible business models.

1) Completely free. I just use it for my own purposes, save my own money and time, and let others use my tools if they want to with no guarantees of functionality. I'm pretty confident I wouldn't end up maintaining it very long under this model, but if I had some help it might survive.

2) Ads. The search page itself contains ads, the emails contain ads, ads ads ads. Not the insane sort that flash and blink and start talking, but some relatively unobtrusive ones. Might offset the time cost some.

3) Subscriptions. You get some subset of services for free, but having more than say three saved searches costs you. So then my question is, what would you pay for this?

4) Additional services. Maybe it makes a giant grocery list, then hires an Uber driver to go pick it all up for you. Or something marginally less crazy?

So I'm throwing it out there. What do my readers (all three of you) think? Should I pick this project back up, finish it out, and maintain it for a while? Would anyone out there use it besides me?