Set up the rule-based size recommender Pro

How it works, in one paragraph

The recommender is rule-based: it reads the measurement ranges you already put in your size chart (e.g. size M = chest 96–101 cm) and compares the shopper's own measurements against them. The whole calculation runs instantly in the shopper's browser using the chart data — no AI model, no external API, and no measurement data ever leaves the page or reaches a third party. That's also why it costs nothing extra to run, whatever your traffic.

Step 1 — Enable the recommender on a guide

1. In your Shopify admin, open FitKit and click the guide you want to add the recommender to.
2. Turn on the Size recommender toggle for that guide.
3. You enable it per guide — only guides where it makes sense (garments with measurable dimensions) need it. A ring size chart or a text-only guide can stay without it.

Step 2 — Choose 2-3 measurement columns

FitKit's table builder supports three column types: Size (the label column — S, M, L, or 38, 40, 42), Measurement (numeric ranges like 96–101), and Text (free notes). The recommender only works with Measurement columns.

1. Select the 2 or 3 measurement columns the shopper will be asked for. Pick the ones that actually determine fit for the garment: for a shirt, chest + waist; for pants, waist + hips (+ inseam if your chart has it).
2. Two well-chosen measurements beat four confusing ones — every extra field lowers the share of shoppers who complete the form. Three is the maximum.
3. Make sure each selected column has a value or range filled in for every size row. A size with an empty cell can never be matched.

Step 3 — Understand how range matching works

When the shopper submits their measurements:

1. For each selected column, FitKit finds the size row whose range contains the entered value. Example: chest 98 cm falls inside M's 96–101 cm range → M matches on chest.
2. The size whose ranges contain all of the shopper's values is recommended. In the common case (a well-built chart with contiguous ranges), that's a single clear answer, displayed instantly.
3. If the measurements don't all land on the same row — chest says M but waist says L — the shopper is between sizes, and FitKit recommends the larger of the two.

Between sizes → size up

Recommending the larger size is the standard fit rule, and it's the right default for returns: a slightly roomy garment gets kept and worn; a tight one goes straight back. The same logic applies when a single value sits exactly on the boundary between two ranges — FitKit resolves upward.

Step 4 — Test it like a shopper

1. Open a product the guide is assigned to on your storefront and click the Size guide link.
2. Run three quick checks: a value squarely inside one size's ranges (should return that size), a value on a boundary or split across two sizes (should return the larger), and the cm/inches toggle (same body, same recommendation in both units).
3. If a result surprises you, check the chart row for that size first — nine times out of ten it's a typo or a gap in a range. See troubleshooting for the rest.

Once the recommender is live, the analytics on your dashboard (also Pro) show guide opens and the add-to-cart rate after opening, per product — so you can see whether removing size doubt actually moves the needle on each item.