A multi-dealer used car marketplace with AI-powered VIN autofill that reduced vehicle listing time from ~2 minutes to 10 seconds.
Dealerships were losing time manually entering vehicle specs during listings. The existing flow required 15+ fields, most of which could be derived from a single VIN. Sellers abandoned the process mid-way, leaving listings incomplete and costing the platform inventory.
I mapped the full listing workflow and identified that the VIN was the single highest-leverage input — it could pre-fill year, make, model, trim, engine, transmission, and more. The redesigned flow surfaces one primary action up front: scan or enter your VIN. Everything else follows automatically, with the user only confirming or correcting prefilled values rather than typing from scratch.
Instead of showing a blank 15-field form, the listing wizard walks dealers through one step at a time using plain, instructional language. Each screen has a single clear action. Error states explain exactly what went wrong and how to fix it — not just 'Invalid VIN'.
Advanced fields (custom features, pricing notes, seller disclosures) are collapsed behind a 'More details' toggle. First-time users don't face decision fatigue. Power users who want full control still have it.
After VIN lookup, the UI shifts from input mode to review mode. Pre-filled values are displayed as editable chips — tapping one opens an inline editor. This mental model (review, not fill) dramatically reduced time on task.
Each dealership has its own inventory namespace. The admin dashboard lets managers see all listings, their status, and any incomplete drafts. Status labels (Draft, Under Review, Live, Sold) are consistent across all surfaces.
Listing time dropped from ~2 minutes to under 10 seconds. The flow was tested with 15+ dealerships. Incomplete listing rate fell by over 60%. Dealers reported the process felt like 'scanning a barcode' rather than doing data entry.
See more