Research / Explainers / 7 of 8

The words become a vehicle first

Shoppers do not type the way a catalog is written. They type a two-digit year, a model with no make, a misspelling, or no year at all. The Fitment Search Engine reads a vehicle out of those words first, and only then asks the catalog what fits. This illustration replays that order on four invented searches and a small invented catalog.

  • StatusIllustration
  • Published2026-10-11
  • Example textInvented
  • Embedded onFitment Search Engine
  • State in the address?q=
  • Codelib/explainers/fitment.ts
  • Protocolv1.0

IllustrationExplainer

The words become a vehicle first

A fitment search settles the vehicle before it looks for a part.

the shopper types08camryfront bumperyear2008inferredmakeToyotainferredmodelCamryreadpartBumpersreadthe catalog, checked against the vehicleCamry 2007-2011Front bumper coverin stockCamry 2007-2009Grille assemblylowCamry 2012-2014Front bumper coverother yearsCamry 2007-2011Headlight, driver sideoutCorolla 2009-2013Front bumper coverother modelCivic 2016-2021Grille, gloss blackother makeCivic 2019-2021Front bumper coverother makeAccord 2018-2022Grille assemblyother makeF-150 2015-2017Headlight, halogenother makeF-150 2018-2020Headlight, LEDother makeAltima 2016-2018Mirror, power, heatedother makeAltima 2019-2022Mirror, powerother make
The shopper types

year2008 inferred

makeToyota inferred

modelCamry read

partBumpers read

Parts that fit3 of 12

Bumpersin stock

Grilleslow

Headlightsout

The vehicle was settled before the catalog was searched. Bumpers comes first because the shopper asked for it; everything else that fits follows, grouped by part type.

Illustration. The searches, the catalog and the stock words are invented, and each reading is written out by hand for the figure. The engine's own rules for reading words are not shown.

The sentence it teaches

A fitment search settles the vehicle before it looks for a part.

Method

  1. Read the words. Each word is offered to four slots: year, make, model and part. A two-digit year is read as a model year, a model name implies its make, and a near miss is matched to a model the catalog actually carries.
  2. Settle the vehicle. When the year, make and model are known, the search goes ahead. When the words fit more than one vehicle, the engine asks the shopper to pick one instead of guessing.
  3. Check the catalog. A part fits when its make and model match and the year sits inside the range its title and tags state. Everything else is set aside, with the reason.
  4. Return what fits, grouped. The part group the shopper asked for comes first, everything else that fits follows by part type, and stock is read from the store when the page loads.

What this does not show

  • It does not show how often real searches are read correctly. No run is reported here.
  • It does not show the engine's rules, word lists or data files.

State in the address

The figure keeps its state in the query string, so a particular view can be linked and the page still renders completely without it.

  • ?q= the search selected, by name

The same figure is embedded on the Fitment Search Engine tool page.

Cite this explainer

OptiVis AI Labs. The words become a vehicle first (explainer). 2026. https://optivisai.org/research/explainers/fitment-search

@misc{optivislabs2026explainerfitmentsearch,
  title = {The words become a vehicle first (explainer)},
  author = {{OptiVis AI Labs}},
  year = {2026},
  month = oct,
  howpublished = {\url{https://optivisai.org/research/explainers/fitment-search}},
  note = {A fitment search settles the vehicle before it looks for a part. Illustration with invented example text.}
}