After @Belfry’s very inspiring presentation on managing his fuel spend, @zeeclor has been interested in something equivalent for our supermarket spend.
@zeeclor has even gone a bit further and made a start on some code to answer his questions. So far it’s been a small group of us that have been discussing it in passing, but i think it’s a much bigger topic, and i’d like to open it up to the broader group.
@zeeclor’s focus has been on tracking what his spend has been. He’s looking at scanning receipts and ingesting the data that’s available there. @Graeme mentioned to me in a phone call recently that he believes that Woolworths offers receipt information on their web site as part of their awards program. That would be very valuable information. For instance, I wonder what i spend on onions in a year
. What about breakfast cereal, hand soap, meat, etc. All of these routine items probably all seem like they’re all on an equal footing with one another, but i reckon an annual retrospective might change your opinion. Onions might sounds like a dumb example, but we use them every time we cook. I’m sure they add up.
Anyway, what our grocery spend has been in the past is valuable, but i’m just as interested in what i need to buy “today”. Convenience is a major player here. In practice, I don’t think it’s worth visiting two shops to by my list, so i only visit one shop. I walk out of that one shop with thirty items in four bags that cost roughly $150, maybe $200. If i was equipped with the data maybe i’d learn that a second stop could have filled one of those bags in a different supermarket and saved me $15.
I’m fighting the urge to race off and start building something because any solution would only be as good as the data that it’s using. A cursory conversation with Claude was pretty demoralising. The Woolworths web site (for instance) makes no guarantee that the price you see online is what you’ll get in your shop. On the other hand, the phone applications are able to answer those questions. I assume the supermarkets have made that concession because you’re probably only going to look up one product if you’re typing with your thumbs. That all but rules out researching my hypothetical thirty items on one site, let alone comparing with the competition.
Anyway, that’s where my head’s at. I think i managed to stay off my soap box through all of that, but i went close talking about the web interface vs the phone app ![]()
Please weigh in if you also find this to be an inspiring/engaging challenge.