Fashion & retail: four vocabularies, one single search
“Wool turtleneck sweater size M”: four pieces of information in four different vocabularies. One single query.
Six attributes, no fixed order
Fashion vocabulary mixes product family, cut, material, color, size, and season (the detail of each attribute), often in no fixed order. An engine that treats the phrase as a bag of words doesn't know that “W30 L32” is a trouser size, not a reference — or that “turtleneck” is a cut, not a distinct product.
Size, cut and material recognized
The mode rule pack activates when you pick this industry, at indexing time. It produces these annotations, each becoming a usable filter:
| Annotation | What it isolates |
|---|---|
TAILLE_m | the letter size, XXS through 3XL |
COL_ROULE | the collar cut, turtleneck or V-neck |
MAT_LAINE | the material — merino, cashmere and mohair under one annotation |
SAISON_AW25 | the season in trade format, SS25 as well as AW25 or FW25 |
JEAN_T30_32 | a pair of trousers and its waist/length size, joined whatever the order of the label |
On top of those: colour, pattern and fastening. Four colour families absorb their shades — burgundy and garnet fall in with red, navy and denim with blue —, four patterns cover the trade's wording (plain, striped, polka dot, floral), and ten synonym groups cover its spellings: t-shirt, tee-shirt, tshirt and “tee shirt” all point at the same aisle, and so do sneaker and trainer. This is the one pack where the difficulty is not reading a code, but accepting that the same garment is written ten different ways.
Three items, as indexed
| Indexed reference | What the engine extracts |
|---|---|
| Merino wool turtleneck sweater, size M | FAM_PULL · COL_ROULE · MAT_LAINE · TAILLE_mturtleneck sweater, wool, size M |
| Slim washed jeans W30 L32 | FAM_PANTALON · COUPE_SLIM · FINITION_DELAVE · TAILLE_30_32 · JEAN_T30_32slim-fit washed jeans, size 30/32 |
| Striped cotton shirt AW25 | FAM_CHEMISE · MOTIF_RAYURES · MAT_COTON · SAISON_AW25striped cotton shirt, fall-winter 25 collection |
Three vocabulary gaps, one customer
A customer searches “red sweater size 38”. Your listings say “burgundy”, “size UK 10”, and “knit”. Three vocabulary gaps, and an empty result on a product you have in stock.
One gap in three covered by the pack
The fashion pack covers one of them: burgundy falls in with red. The other two are not in the pack. You add your own — house color names, collection abbreviations — via Custom Rules, with no regex to write.
A term specific to your catalog that the pack still misses (an in-house collection abbreviation, for example)? Add your own Custom Rule from the console, simply describing what it should recognize — no regex required.
Index your fashion catalog
The getting started guide covers the full setup, from sign-up to your first indexed catalog. The documentation details every endpoint.
Watch the breakdown happen. The table above is the engine's output, not hand-written. The same mechanism runs on the home page, with no signup — on an apparel catalog: pick “Fashion & Apparel” in the selector, 1,500 references (measured 1 September 2026). Try “V-neck t-shirt black” and read what the engine pulled out of it.
Open the live engine →