Built for the question a tracker cannot answer.

Most guidance is written for a population. It is a reasonable starting point and a poor stopping point, because the effect of a late dinner, a walk, or a change in timing varies enormously between people, and nobody can tell you in advance which side of that variation you are on.

The usual answer is to log more. But logging produces a record, and a record is not an answer. Nothing in a food diary tells you whether the walk was worth the twelve minutes.

Everything you record goes onto one timeline, with its source attached. Patterns are found by a statistical model, reported with a sample count and a confidence, and stated alongside what they do not account for.

When a pattern is interesting but uncertain, the answer is a small experiment you can actually run: the same breakfast for six days, walking after three of them. The prediction is written down first and cannot be edited afterwards, so over time you accumulate a record of how often this system was right about you in particular.

The person this is built for is the one living with the condition. Every finding is written to be read by them, not summarised for them.

Your clinician sees a version of the same thing: ninety days on one page, with what you tested and what it showed. Ten minutes is not enough time to explain a year, and a stack of screenshots does not help. This is the part that makes the appointment better.

This will not adjust your medication, calculate an insulin dose, diagnose a complication, or advise you during a hypo. Those limits are enforced in the system itself: an experiment marked as needing clinician review cannot be started without it, and a prediction, once made, cannot be quietly rewritten to look better.

It also will not claim to reverse anything. There is a lot of that about, and it is not a claim the evidence supports.

How this works goes through the whole loop, from a single reading to something worth bringing to an appointment.

The record works for every kind of diabetes: the timeline, imports from a meter or CGM, meals, medication, activity and labs, the safety rules, and the summary you bring to an appointment.

The loop is closed for Type 2 and prediabetes. A finding can propose the test that would settle it, starting that test writes down what Wellovue expects before you begin, and the result is scored against it. Neither half can be edited afterwards.

Wellovue knows which kind of diabetes it is reading for and refuses to interpret the ones it has no reviewed model for — Type 1, gestational and clinician-classified forms are recorded and not yet interpreted. Weighing competing explanations for a pattern is still being built.

If you want to use it now, bring a CSV from your meter or CGM. If you would rather wait, leave an address on the home page and we will write when there is something worth writing about.