Your own record, read as a body rather than a spreadsheet.

A rise of 3.1 mmol/L is a statistic. Peaking at 11.2 an hour after eating, spending fifty minutes above your target range and coming back under it at ninety-five minutes is the same meal described in the language the condition is actually lived in.

Wellovue measures each meal’s response on its own curve — time to peak, minutes above target range, time back in range — and then averages across meals. A meal that was not watched for long enough to say when glucose came back is left out of those timings rather than guessed at, and the page tells you how many meals it could actually measure.

Meals followed by activity are compared against meals that were not, with both curves drawn against your target range. The finding says plainly that this is an observed association rather than a controlled comparison, because the two groups differ in more than the walk.

If the difference looks worth knowing, you can test it: a short experiment where the walk is assigned at random rather than chosen on the days you already feel well.

Meals after 20:00 are compared with earlier ones, read on your own clock rather than the server’s. Waking glucose is described across the 05:00–09:00 window, with how much it varies between days and how many mornings sat above your target range.

A persistently raised morning pattern is flagged as worth raising with your clinician. Wellovue surfaces it; it does not interpret it, and it never suggests what to do about it.

HbA1c is taken quarterly, so a thirty-day window would show one result or none. Lab trends are read over a longer span and say which span they used, because it is not the one the rest of the screen is showing. A trend across results recorded in two different units is refused rather than drawn.

Wellovue is not a medical device and is not a substitute for clinical care. It does not diagnose, prescribe, or calculate insulin doses, and it will not advise you during a hypo or a hyper. Some subjects are gated behind a clinician conversation regardless of what the data shows.

Findings for type 1 and gestational diabetes are not offered. The engine models type 2 physiology, and running it over a record it was not built for would produce confident answers from the wrong model of a body. That gate stays until clinical review lifts it.

How this works walks the seven steps from a reading to a checkable finding. Diabetes intelligence defines the terms. The clinician report is what you can take to an appointment.