“Ages 2–4” Isn’t a Measurement. Here’s What to Do About It.

What you’re dealing with

Age-bracket labels are a legacy convention that the industry has quietly agreed is broken. Children grow in non-linear spurts, an “Ages 2–4” label from one brand doesn’t cover the same children as an “Ages 2–4” from another, and parents — the actual buyers — have no way to verify any of it before ordering. The return rate in childrenswear tells the story: a meaningful share is driven by sizing, not by preference.

The second problem is structural for you specifically: when you source across regions, the dimensional envelope behind the same age label varies significantly between an Asian supplier and a European manufacturer. You’re discovering this after the container lands, not before the contract is signed.

Where the money is

• Return clarity: Translate “Ages 2–4” into the actual chest, height, and inseam range your product covers. A parent with a measured child can compare directly — that’s the single biggest removable cause of childrenswear returns.
• Cross-regional sourcing validation: Query the same age bracket with different target_region settings and get a side-by-side dimensional delta. You see whether your supplier’s “Ages 6–8” label matches your target market’s “Ages 6–8” label before the PO goes out, not after.
• “Fits until” intelligence: Use CDC median growth curves to project when a typical child outgrows your product. “Fits most children until approximately age 3.5” on a listing builds trust in a category where every competitor is hiding behind vague brackets.
• Design inputs for children’s products and toys: For apparel designers, age-bracket boundaries become exact dimensional envelopes (chest range, inseam range, shoulder breadth). For toy and children’s product designers, the same data informs grip width, reach distance, and proportional scaling — turning “for ages 3–5” from a marketing label into a specification.

What we actually do

We convert age and gender into a 130-point anthropometric profile grounded in CDC clinical growth data — the same LMS tables used in pediatric growth charts. No measurements of the child required. In under 10ms.

The API returns both the population median and a 95% prediction interval (range_95) for every dimension, so you see not just the typical child but the range of children your product will encounter.

Where we’re upfront

These are population medians and percentiles, not predictions of a specific child. A 3-year-old in the 90th percentile will hit your product’s dimensional limits sooner than the median; a 3-year-old in the 10th percentile, later. We return that range in every response rather than hide it, because a single number without the spread is misleading in pediatric sizing.

What that means for you: your size labels stop being vague and start being verifiable. A parent reads “fits most children until approximately age 3.5” on your listing and can compare it against their own child’s measurements. Your sourcing team signs contracts with dimensional validation, not hope. And your designers work from a CDC-grounded dimensional envelope instead of a competitor’s guesswork copied forward one more season.

CDC Clinical Growth Charts (public domain). All projections based on CDC 50th percentile growth curves; variance around the median is quantified via range_95 in every response.