Tomás kept a spreadsheet of his urine ketone strip results for eleven weeks, colour-coded, and it told him almost nothing.

He did it because it felt like control. Every morning, first thing, he'd dip the strip and compare the colour against the little chart on the bottle. Dark purple meant good. Light pink meant he'd messed up. He'd then adjust his day around the colour, cutting a few more carbs here, fasting a bit longer there.

What he was actually tracking, it turns out, was how much water he'd drunk and how long since he last peed. Some mornings the same exact diet produced a dark result, and some mornings a pale one, and he spent eleven weeks chasing a signal that was mostly noise.

Three measurements, three different questions

People treat "testing ketones" as one thing. It's three things, and they don't tell you the same story.

Urine ketones. Ketone bodies your body made and then excreted. Cheap, painless, and a rough indicator of whether you were in ketosis at some point since your last bathroom trip. The problem is exactly that: it measures what you threw away, not what's circulating. As you become better adapted, your body excretes less of the ketones it makes, so urine strips trend lighter even when your blood ketones are perfectly good. That is why Tomás's colour got paler over eleven weeks and why he drew the wrong conclusion from it.

Breath acetone. One of the three ketone bodies your body produces, acetone, is volatile and comes out in your breath. Handheld meters measure it and give you a real-time number without a pinprick. They're more consistent than urine and far better than peeing on a stick, but they still drift with hydration, and the readings don't map cleanly onto blood numbers.

Blood beta-hydroxybutyrate, or BHB. This is the one clinicians use, and it's the reference method rather than a proxy for it. A finger-stick gives you a number in seconds, the value measured in millimoles per litre, and it reflects what's actually circulating. A dedicated blood ketone and glucose meter costs more than the other two, and the strips are a recurring cost, which is what puts people off. It's also the only one that will reliably tell you the difference between "adapted" and "starving."

If you're making decisions off a measurement, the measurement needs to be about the thing you're changing.

A hand resting beside a glucose meter on a table
A finger-stick reading is the reference number, not an estimate.

The numbers, and what they actually mean

A few reference points, so the graph on the meter means something.

For most adults eating normally, blood ketones sit below 0.5 mmol/L. Nutritional ketosis is generally described from 0.5 mmol/L upward, with many people working in the 0.5 to 3.0 range. The highest range, above roughly 8 mmol/L, is not a goal to chase and is a sign of trouble for people with type 1 diabetes, where it can indicate diabetic ketoacidosis and requires medical attention. Those are different conditions with similar names, and confusing the two is a genuine safety issue.

Blood glucose matters alongside it, because the pair tells you more than either does alone. Someone with a ketone reading of 1.0 and a glucose of 85 is in a different situation from someone with the same ketone number and a glucose of 160. Context is the whole point.

0.5
mmol/L and up
The usual threshold for describing nutritional ketosis

When measuring is genuinely worth it

Not everyone needs a meter. Plenty of people do well on feel and consistency, and adding a device to a diet can turn it into an anxiety project. The cases where a number genuinely earns its cost:

  • You've stalled for weeks and can't explain it. A reading tells you whether you're actually in ketosis or have drifted out without noticing.
  • You're chasing a specific metabolic goal rather than weight, and you want to see whether your changes are moving the number.
  • You eat out often and the hidden carb content is unpredictable. Testing after a restaurant meal is the only way to learn what a dish actually did.
  • You're on medication for diabetes or blood pressure and your clinician has asked you to track. In that case the meter isn't optional and the existing clinical relationship should drive the decisions.
  • You're experimenting with something new, like higher protein, or a change in meal timing, and you want to know whether it changed your ketone levels.

The case where it's not worth it is the one Tomás was caught in: checking daily for reassurance, with no decision attached to the reading. A number you don't act on is just a mood.

A measurement without a decision attached to it is entertainment.

The low-carb notes, without the nonsense

One email a week. Real numbers, real weeks, unsubscribe whenever.

How to test without turning it into a second diet

If you do buy a meter, use it the way you'd use any instrument: to answer a question you already have.

Test at consistent times. Ketones vary across the day with meals, exercise and sleep, so a reading at 7am and a reading at 9pm aren't comparable. Morning, fasted, is the common baseline.

Write down more than the number. The ketone reading, the glucose reading, what you ate the day before, how you slept, whether you trained. On its own, a single number is trivia. As a row in a log with context, it's a signal.

Don't test daily forever. Test when there's a question. A week of readings around a change you made tells you more than three months of daily colour-matching.

Remember that the reading doesn't tell you whether you're losing fat. Ketone levels are not a fat-loss scoreboard. Someone can be deeply in ketosis and eating at maintenance, and lose nothing. The meter answers "am I in ketosis," not "is this working."

A finger-stick meter that reads ketones and glucose from the same device is the version worth the money, because the pair is the useful unit. If you want the reference standard rather than a proxy, a blood glucose and ketone monitor that reads from a finger-stick is what that looks like, and it replaces the guesses you've been making with a number you can act on.

If you want to go deeper, reading your ketone numbers over time and what exogenous ketones can and cannot do pick up where this leaves off.

Sources

  • Blood ketone and glucose monitoring for nutritional ketosis: clinical and consumer measurement methods. Keto-Mojo. https://keto-mojo.com/
  • Effect of low-fat versus low-carbohydrate diets on energy metabolism. National Institute of Diabetes and Digestive and Kidney Diseases clinical protocol, 2020. https://cdn.clinicaltrials.gov/large-docs/08/NCT03878108/Prot_SAP_000.pdf
  • Diabetic ketoacidosis: symptoms and when to seek care. Centers for Disease Control and Prevention. https://www.cdc.gov/diabetes/about/diabetic-ketoacidosis.html
  • Symptoms during initiation of a ketogenic diet: a scoping review. Journal of Human Nutrition and Dietetics, 2025. https://pmc.ncbi.nlm.nih.gov/articles/PMC11978633/

This article is educational, not medical advice. Blood ketone meters are health devices; the readings are not a diagnosis. If you have diabetes or take medication, discuss monitoring with a qualified professional before you start tracking.