What's the best kitchen scale for food logging?
A 1 g-resolution scale in the £12–£20 range.
Accuracy is effectively solved at every price point in this category, so buy on display legibility, tare placement and auto-off delay rather than on claimed precision.
We bought seven kitchen scales from about £12 to about £70 and tested each against calibration weights at 10 g, 100 g, 500 g and 1000 g — ten readings per point, cold and after twenty minutes of use.
The result is boring and worth stating
All seven were within 1 g at 100 g and within 2 g at 500 g.
The £12 scale and the £70 scale disagreed with the reference by amounts that are, for food logging, indistinguishable. Accuracy is effectively solved at every price point in this category.
That means the money is buying something other than accuracy, and it is worth knowing what.
What actually differs
Display legibility from standing height. You are looking down at it from about a metre with a bowl in the way. Two of the seven had displays that were unreadable at that angle, which is a design failure rather than a preference.
Tare button placement. You press it constantly. If your hand has to move off the bowl to reach it, you will notice every single time.
Auto-off delay. This is the one nobody advertises and the most consequential. Three of the seven switched off quickly enough to lose a running total while we fetched something from the fridge. That is genuinely infuriating and no spec sheet mentions it.
Capacity. 3 kg is the practical minimum. A large mixing bowl can consume a kilogram before any food goes in.
Why this belongs in a report about measurement
Because a scale is the only instrument in food logging that is not estimating.
A photo estimate infers portion from an image. A database entry tells you what 100 g of something contains and then asks you how much you had. A scale answers that second question directly, and for the foods you can put on it — which for most home cooks is the majority of their calories — it removes the largest single error source in the whole process.
That is not an argument against food-logging apps. Most of what people eat cannot be sensibly weighed: restaurant meals, shared dishes, anything somebody else cooked. That is precisely where an estimator earns its place, and independent benchmarking of those tools reports errors ranging from roughly 1% to over 12% depending on the tool.
It is an argument for weighing the things you can weigh, which costs twelve pounds once.
The maintenance nobody does
None of the seven shipped with a calibration weight.
A 500 g reference weight costs a few pounds. Checking your scale against it once a year takes ten seconds and is the only way to know whether it has drifted — which cheap scales do, over years.
That is the entire maintenance regime for this category, and it is more than almost anyone does.
What we would buy
A 1 g-resolution scale with at least 3 kg capacity, in the £12–£20 range, chosen in a shop where you can see the display and press the tare button before paying.
Add a 0.1 g mode only if you weigh coffee or spices. It is irrelevant for weighing chicken.
Also considered (and didn't win)
Scales with a 0.1 g mode for coffee and spices · Scales with a detachable, dishwasher-safe platform · Scales with a pull-out display for use under large bowls
Frequently Asked Questions
Do I need an expensive kitchen scale for tracking food?
No. We tested seven against calibration weights at 10 g, 100 g, 500 g and 1000 g, ten readings each. All seven were within 1 g at 100 g and within 2 g at 500 g, across a price range of roughly five to one. Accuracy is effectively solved in this category, which means the money buys build quality and convenience rather than better numbers.
What resolution do I need?
1 g is sufficient for food logging. A 0.1 g mode is useful for coffee, spices and supplements, and irrelevant for weighing chicken. Capacity matters more than people expect — 3 kg is the practical minimum, because a large mixing bowl can consume a kilogram of that before any food goes in.
Why does weighing food matter if my app estimates well?
Because weighing is not estimating. Even a very good photo estimate is inferring portion from an image, while a scale reads it. For foods you can put on a scale — which for most home cooks is the majority of their calories — weighing removes the largest single error source in a food log, and it costs about twelve pounds once. Estimation earns its place for restaurant meals, shared dishes and anything you did not cook.
How do I check whether my scale is still accurate?
A 500 g calibration weight costs a few pounds and takes ten seconds to use. Cheap scales drift over years and none of the seven we tested shipped with a reference weight. Checking annually is more maintenance than almost anyone does and it is the only way to know.
References
How we picked. What's The Best Report follows a documented winner-selection methodology and editorial policy. We accept no affiliate revenue. See our no-affiliate disclosure.