The trap in the question
There is no such thing as “your actual size.” Size is not a property of your body. It is a label a brand attached to a garment, and two brands will attach different numbers to the same shirt.
A 2025 survey of 687 shoppers across the US and UK found 91% said their size depends on the brand. That is the honest starting point.
So what should an app actually give you?
Measurements. Not a size. A useful tool returns values that hold up regardless of who made the garment: chest, waist, hip, shoulder, neck, inseam, arm.
The reason is visible in the data. Tellar Research measured more than 1,000 garments across 50 UK brands against their own published charts. Only 64% matched within 1 cm, and average within-brand variance was 4.2 cm — wider than many size steps. Chart accuracy also fell as sizes got larger: 71% at UK 8-10, dropping to 54% at UK 20-22.
If a brand's own chart disagrees with its own garment more than half the time, inferring your size from that chart is guesswork with extra steps.
When an inferred size is useful
It is a reasonable filter when you already know a brand runs small, or when you are browsing and want a starting point. It fails badly as a precision tool.
Fit Analytics surveyed 1,400 consumers and found 91% said sizes vary by brand. Among people who returned size-related items, 55% said the item was too small and 40% too big — a strong signal that the label is often simply wrong rather than the shopper misjudging.
The practical approach
- Measure once properly and store the numbers.
- Use measurements to filter or compare, not a size label to decide.
- When a brand's chart matters, measure the actual garment if you can.
- Read fit reviews, which describe the cut, not the label.
Bottom line
Any app can produce a size. Very few can justify it. The durable number is 34 inches of waist, not a medium.
If you want the whole picture, start with how accurate these tools actually are, then read why labels keep disagreeing.