Give shoppers size confidence that holds after delivery.

A size recommendation shoppers can trust, honest about how sure it is, and judged on what stays sold.

Start with a conversion review How it works

I’m 5’10” and usually a medium. Which size in the Canvas field jacket?

Medium, with moderate confidence. Most shoppers your height keep the M. A few size up to layer under it, so if you’ll wear a mid-weight layer, take the L. The size chart has the exact measurements.

From past orders and return reasons for this style

What the shopper sees: a size, how sure it is, and when to size up.

Sound familiar?

“True to size?” was asked 27 times about one jacket.
The size chart is correct and still doesn’t help.
The same two sizes keep coming back for exchange.

A wrong size costs the return, the shipping both ways and often the customer.

What changes.

TodayWith the engine

Advice: A size chart and a quizA recommendation with how sure it is
Thin data: A confident guessThe chart and a range when the data is thin
Judging it: Quiz completionsContribution after returns and exchanges
Claims: Fewer returns, promisedMeasured by size and by product

How it works. 5 steps, 4 stages.

AI1
Rules and automation3
Your team1
Approval

1Gather
1. Rules and automation
  Join sizing, orders and return reasons
  Garment measurements by size, what shoppers kept, exchanged or returned, and why.
  - Shopify
2Recommend
1. AI
  Explain a size in plain words
  Takes height, build and how it will be worn, and returns a size, a confidence level and a reason.
3Fall back
1. Rules and automation
  Fall back when the data is thin
  Below a set number of orders for a size, the shopper sees the chart and a range, not a confident guess.
2. Your teamApproval
  Tune the rules
  Category owners set the thresholds and review styles that keep coming back for fit.
4Judge
1. Rules and automation
  Judge on what stays sold
  Held-out shoppers, and contribution after returns and exchanges, by product and size.
  - GrowthBook

Return reasons sharpen the next recommendation

What you get. Real work, every week.

Fit data by style

2 of 4styles with enough data to recommend

Canvas field jacketEnough orders in M and L
Merino crew
Trail pantWaist and inseam: thin data, chart shown
Waxed field jacketReturn reasons not tagged

Where a recommendation is safe to show. Elsewhere the chart does the job.

Exchanged for fit, by sizeCanvas field jacket

MMost orders4%Recommend
LShoppers who layer size up6%Recommend
SFew ordersThinShow the chart

The cohort view the category owner reads each month.

How you’ll know it’s working.

Contribution after returns
A size recommendation only helps if the order stays sold.
Exchanges and fit returns, by size
Shows where the recommendation or the chart is wrong.
Repeat purchase
A size that fits brings the customer back.

Baselines agreed before we build.

Works with what you have.

Commerce

Shopify
True Fit

Support

Gorgias

We configure first: A clear size chart and fit notes drawn from your own return reasons come first. A specialist fit vendor is worth pricing once your data shows a real gap.
We build when: You have enough orders and return reasons per style to learn from, and fit is a main reason things come back.

What a first build includes.

Built in your accounts, around your team. Scope, fee and baseline agreed before we start.

1Design
2Build

Start with a review Scope this build

We build

Garment sizing joined to orders and return reasons
A fit card with a confidence level
A chart fallback for thin data
A readout by product and size

We start from

Garment measurements by size
Orders and return reasons
Product pages and size charts

From your team

Owner: Head of ecommerce, with merchandising
Involves: Ecommerce
Merchandising
Customer service
Product development

Questions teams ask.

Will this reduce returns?

We can’t promise that. A peer-reviewed 2025 study of a Swedish apparel retailer’s size finder linked its use to slightly higher returns. It was observational, so it can’t show cause. So we measure contribution after returns, by size, on your own orders.

Do we need a fit vendor?

Not at first. Your own orders and return reasons go a long way. A specialist such as True Fit makes sense when you want its data and the cost is justified, and we’d compare it with the simpler version first.

Who owns it once it’s running?

You do. It’s built in your accounts, documented, with training for the people who run it. Ownership usually sits with your head of ecommerce, with merchandising. We can keep running it with you, or step back.

How do we start?

Usually with a conversion review: two to four weeks, a fixed fee credited to the build, and a plan that’s yours. If the job is already clear, we can scope the build directly.