Working out what your groceries actually cost
Everyone knows groceries have gone up. Almost nobody can say which things, by how much, or whether the price on the shelf today is a good one. Your receipts already know.
Ask most households what they spend on food and you’ll get a number that’s out by twenty per cent in either direction. Ask which items have gone up and the answer is a vibe: “meat, I think? And cheese has gone stupid.”
That’s not carelessness. It’s that the information is genuinely hard to hold. A weekly shop is forty-odd items, the prices move independently, half of them are on special some weeks and not others, and the pack sizes change underneath you. Nobody can track that in their head, and the bank statement, the one number that is easy to get, tells you the total and nothing else.
The frustrating part is that the data exists. You are handed it, on paper, every single week, and then you throw it away.
A receipt is a price observation with a date on it
Photograph the receipt instead of binning it and you have, for every line: what it was, what it cost, which shop, and when. Do that for a couple of months and you have something no supermarket app will give you, which is your own price history, for the things you actually buy.
That history answers the questions the total can’t:
- Is $8.50 a good price for this? Not in the abstract, but against the eleven times you’ve bought it. If the lowest you’ve seen is $6 and you’ve been paying $7.20, then no.
- What’s actually driving the increase? Not “groceries”, but the six lines doing most of the moving.
- Is this “special” a special? A sticker means the price is lower than the price they were charging last week. Your own history says whether it’s lower than what you normally pay, which is a different question and often has a different answer.
In Harvenna it’s three steps: photograph the receipt, check it read everything correctly, confirm. Nothing counts towards your price history until you confirm, so a receipt you started fixing and abandoned can’t quietly skew it. After that, every item on your shopping list shows the last price you paid and the lowest you’ve ever seen, with a chart behind it of what each shop charged and when.
The trick with pack sizes
There’s one price movement that pack-price tracking cannot see, and it’s the one most likely to be happening to you.
The chips stay $5.50. The bag goes from 200g to 175g. The shelf price is unchanged, your receipt is unchanged, every chart is flat, and you’re paying 14% more per gram. Do that across a category over two years and it’s a substantial, entirely invisible increase.
Catching it means watching what you pay per gram, not per packet, and looking for a change that sticks rather than one odd week. Harvenna works out the per-gram price for each food at each shop, then compares the last couple of months against the couple before that, so a one-off doesn’t set anything off but a real rise does. If the packet shrank, it tells you the old and new size. If the price simply went up (loose fruit and veg, where there’s no packet to shrink) it says that instead.
That’s part of the price insights on the Family plan, alongside “this is cheap right now”, “this is dear right now”, and seasonal dips.
Where the money is actually going
Once receipts are confirmed, the big questions get easy and stop being guesses: what went to each shop, which foods take the biggest share, what the monthly trend looks like, how this week compares to your recent average.
Two details in how Harvenna reports that are worth knowing, because they’re the sort of thing that makes a number quietly misleading:
- The four-week average always divides by four, even if you’ve only been scanning for two weeks. So it also tells you how many of those four weeks actually had spend, and that’s what distinguishes a new household from a quiet month.
- Top foods are counted by line, not by purchase. A food that appears twice on one receipt counts twice, because that’s what the receipt says happened.
Neither is exciting. Both are the difference between a chart you can act on and a chart you half-trust.
What it’s for
The point of all this isn’t austerity. It’s that a decision made with the number is different from one made without it.
Knowing that meat is 40% of your grocery spend doesn’t mean stopping. It means a leftover night is worth more than you thought. Knowing that a food is at its lowest price of the year means buying two. Knowing your usual weekly figure means a $260 week is information rather than a fright.
And it makes the meal plan cost-aware in a way that isn’t preachy. Harvenna’s planner already uses your observed prices as one input when it scores a week: a gentle nudge in the ranking, never shown as a score and never the reason a meal is chosen or rejected. Nobody needs their dinner rated on value for money. But given two options it’s otherwise indifferent about, it may as well pick the cheaper one.
Getting started
The honest answer on effort: it’s one photo per shop, and it takes about fifteen seconds. You need a handful of receipts before the history says anything useful, and a couple of months before the trend lines mean much. It’s a slow-burn feature, in that the value arrives later than the work does, which is exactly why almost nobody does this by hand.
Receipt scanning, price history, insights and spend analytics are on the Family plan. Manually setting a usual price on a shopping item, and the weekly budget you measure against, are free.
If you only take one thing: keep the receipts. Even if you never scan a single one, the household that photographs them has an answer available to it, and the household that bins them has a vibe.
Harvenna turns your grocery receipts into a price history for the things you actually buy, so “is this a good price?” has an answer.