The technology is not the obstacle
Measurement technology works today, and it is cheap. The obstacle is that retailers have decades of accumulated data in the wrong shape.
What retailers actually hold
Purchase history keyed on a size label. A customer is recorded as having bought a medium in a particular brand, not as having a 78 cm waist.
That data cannot answer the question. It records what people chose, which is influenced by the label, the photo, and marketing as much as by body shape.
Why the historical data is a trap
If recommendations are built on what similar customers bought, they reproduce historical sizing behaviour, including its errors. Brands that graded inconsistently for years have trained models on inconsistent labels.
Research on over 1,000 garments across 50 UK brands found only 64% matched their own chart within 1 cm. Training on those labels teaches the model the error.
The consent problem
To know a customer's body, you need them to give you their body. That requires explicit consent for sensitive personal data, which under Nigeria's Data Protection Act 2023 means a higher standard than ordinary consent.
Very few shoppers will hand a retailer their measurements for a first purchase with a brand they do not know. Trust is the real currency here.
The incentive problem
Returns cost retailers real money, but so do fitting rooms, and so do the returns that a free-return policy encourages. Some return volume is profitable under current unit economics.
That is an uncomfortable thing to say, but it explains why a problem this visible has persisted this long.
What has actually changed
- Measurement is now cheap enough that it is a checkbox rather than a project.
- Customers now expect to input data to get personalised results.
- Return rates have risen enough to reach board level: $849.9 billion in 2025.
- Some brands do publish finished garment measurements, which is the single most useful change.
The honest read
Adoption is happening at the edges rather than the core. Fit recommendation tools exist, but they work off purchase history and grading rules, not off measured bodies.
The genuinely useful innovation would be publishing finished garment dimensions. That costs a retailer nothing and removes more guesswork than any recommendation engine.
Further reading: what brands should publish and the chart problem in full.