Every month the same headline: food inflation was such-and-such per cent. And every month plenty of people look at their receipt and think not in my house. It isn't a feeling. The two numbers measure different things, and understanding the difference is the first step to deciding your shop instead of just enduring it.
What the CPI actually measures
Your national statistics office calculates the Consumer Price Index every month. To do it, it maintains a basket of representative items and collects their prices from shops across the country. It then works out how much the total cost of that basket has moved against the same month a year earlier.
The part that rarely gets explained is the second one: not every product counts equally. Each item enters with a weight taken from a household spending survey — how much the average household actually spends on it. If households spend far more on meat than on tea, meat moves the index far more than tea does.
It's a good macroeconomic indicator. It just isn't your indicator: it describes the average household's spending, and the average household doesn't exist. More to the point, it tells you nothing you can use on Saturday in the supermarket aisle.
Five reasons your number is different
1. You don't buy the average basket
This is the big one. If you're vegetarian, a rise in meat barely touches you. If you drink coffee every morning, a rise in coffee hits you squarely even though it carries little weight in the index. Your basket has its own weights: yours. Two people under the same official CPI can be living through very different inflation depending on what they buy.
2. It's a national average, and prices aren't national
The same product doesn't cost the same in two chains, two regions, or sometimes two branches of the same chain. The CPI averages all of that away. You shop in three or four specific places, and what matters to you is what happens there — which is also, conveniently, the information you can actually act on.
3. The index adjusts for quality; your wallet doesn't
When a product changes — new formulation, different packaging, different quality — the index tries to separate the part of the price move that is a quality change from the part that is a pure price rise. That's the right thing to do when measuring prices. But at the till you pay the whole amount, not the adjusted one.
4. Shrinkflation is felt long before it's seen
When a pack goes from a kilo to 900 grams at the same price, that's an 11% rise per kilo. The index catches it, because it works in price per unit of measurement. You, on the other hand, see the same price on the label and just notice the pack runs out sooner. It's real inflation that takes months to register consciously. There's a separate guide on price per kilo or litre.
5. You substitute — and substituting changes your own basket
When something rises sharply you buy less of it, switch brand, or drop it. That cushions the blow to your total spend, but it also means your spending no longer measures what it measured last year: comparing this month's total against a year ago doesn't tell you how much prices rose, it tells you how much your spending changed. Two different questions. Both matter, but they have to be separated.
Now the question that's actually useful
Knowing your inflation is 7% rather than 4% is satisfying for about thirty seconds. It changes nothing. The percentage is a diagnosis: it confirms you weren't imagining things, and that's where its usefulness ends.
The actionable question is a different one: where is my money actually going, and how much of that is up to me? Because you can't change a percentage, but you can change a list of specific products. It almost always turns out that:
- The rise is concentrated. Not "everything" goes up. A handful of products do, and usually not the ones you'd guess. That's where the decision lives.
- Frequency beats percentage. A 40% rise on something you buy twice a year moves nothing. An 8% rise on something you buy weekly does. Sorting by real impact in pounds or euros per month completely reorders your priorities.
- Part of the rise isn't a rise, it's the shop. A branded product is identical everywhere; if it costs more in one place, that isn't inflation, it's a gap you could close any week.
- And part of it isn't price, it's quantity. Buying more and throwing more away is expensive without anything having gone up.
That's what being in control means: not a number that depresses you, but knowing exactly where the money goes so you can decide what to do about it — switch brand, switch shop, buy a different size, or do nothing because now you know it doesn't matter. Saving is a possible consequence, not an obligation; what it can't keep being is a monthly surprise.
Why almost nobody gets that far
Because by hand it's miserable work. You need a price history for every product you buy, in every shop, over months. That means keeping every receipt, typing every line into a spreadsheet, and then solving the genuinely hard part: recognising that "WHL MILK 6X1L", "Whole milk 6-pack" and "MILK W. 6UN" are the same product, and that when it switched to 5×1L you have to compare per litre, not per pack.
None of that is intellectually difficult. It's simply too tedious to sustain for more than a fortnight, which is exactly why most people end up with the feeling instead of the control.
Where to go next
The method for breaking your spending down and seeing which part is up to you is in where your grocery money actually goes. And the one comparison behind almost every decision in the aisle is in price per kilo or litre.
The CPI answers "how much have prices risen nationally?". It's well built and it does its job. But the question you take home is a different one, and only your own receipt can answer it.