A1C is not a flat average of your blood sugar over the last 3 months — it's a recency-weighted average, where the most recent 30 days count more than the oldest ones. Your A1C is probably the single number your doctor watches most closely, and also one of the least understood. Most people assume it's just "the average of your blood sugar for the last 3 months." That's close, but not quite right, and the difference actually matters.
What A1C actually measures
A1C (also called HbA1c) measures the percentage of your hemoglobin — the protein in red blood cells that carries oxygen — that has glucose attached to it. When blood sugar is elevated, more glucose binds to hemoglobin. Because red blood cells live for about 2–3 months, the A1C test captures a rolling snapshot of your average blood sugar over roughly that window, not just the last few days.
Why it's not a flat average
Here's the part most explanations skip: not all of the last three months count equally. Because red blood cells are constantly being replaced, more recent blood sugar levels have a bigger influence on your A1C than older ones. A rough, commonly cited breakdown looks like this:
- The most recent 30 days — contribute the most to your A1C
- 31–60 days ago — contribute moderately
- 61–90 days ago — contribute the least
This is why two people with the identical average blood sugar over 90 days can get different A1C results if their pattern was different — someone who was high early and improved recently will generally test better than someone whose control has been slipping, even with the same overall average.
Translating A1C to an estimated average glucose
Your A1C percentage can be converted to an estimated average glucose (eAG) using a formula from the ADA (based on the Nathan et al. formula) -- in mg/dL, or directly in mmol/L using the same study's alternate formula:
eAG (mg/dL) = (28.7 × A1C) − 46.7
eAG (mmol/L) = (1.59 × A1C) − 2.59
| A1C | Estimated Average Glucose (mg/dL) | Estimated Average Glucose (mmol/L) |
|---|---|---|
| 5.0% | ~97 mg/dL | ~5.4 mmol/L |
| 5.7% (pre-diabetes threshold) | ~117 mg/dL | ~6.5 mmol/L |
| 6.5% (diabetes threshold) | ~140 mg/dL | ~7.8 mmol/L |
| 7.0% | ~154 mg/dL | ~8.5 mmol/L |
| 8.0% | ~183 mg/dL | ~10.2 mmol/L |
| 9.0% | ~212 mg/dL | ~11.8 mmol/L |
General ADA reference ranges: below 5.7% is considered normal, 5.7%–6.4% is the pre-diabetes range, and 6.5% or above on two separate tests is used to diagnose diabetes. Most adults with diabetes are given an individualized target, commonly around 7%, though your doctor may set a different goal based on your age, other health conditions, and risk of low blood sugar.
Why A1C alone isn't the full picture
A1C is a single number, but the same A1C can hide very different day-to-day realities. Someone whose blood sugar stays steady in a moderate range and someone whose blood sugar swings wildly between very low and very high can end up with the same A1C average — despite one situation being far riskier day-to-day. This is part of why more clinicians now also look at time in range (the percentage of time your glucose stays within target) alongside A1C, not instead of it.
GlucoBeacon's A1C Estimator applies this same recency-weighted formula to your logged readings — so between lab tests, you can see where your number is trending, not just guess. Pair it with the Time in Range view for the fuller picture a single A1C number can't show on its own.
The bottom line
A1C is a genuinely useful long-term indicator, but it's a trailing, weighted average — not a live readout, and not the whole story on its own. Understanding the lag helps explain why a big lifestyle change today won't show up on a lab result for weeks, and why staying consistent matters more than any single good (or bad) week.