AI goods receiving: acceptance by photo

Intake that leaves nothing to argue about with suppliers

A storekeeper photographs pallets, boxes and labels with a phone or a handheld scanner with a camera. The model counts packages, reads the labels, spots crushed or wet packaging and checks the delivery against the order. The intake record with photos and a discrepancy report goes to your ERP (1C) or WMS.

Choose your scale →

What gets lost at intake

  • The truck is at the dock and the driver is in a hurry. The storekeeper counts boxes on the run and signs the waybill.
  • A week later crushed packs turn up inside the pallet. The supplier replies that nobody complained at intake.
  • Goods arrive with the wrong SKU or a short shelf life. Nobody notices until an order is being picked.
  • Photos are taken at intake, but they stay on a personal phone. A month later the shot of a specific delivery is impossible to find.

What changes after launch

  • 01Discrepancies with the order are visible right at intake
  • 02A report with photos is ready before the truck leaves
  • 03Your ERP and WMS record what actually arrived

How the system works

  1. 01

    Load the order

    The system pulls the expected delivery from 1C or your WMS: items, quantities, SKUs, expiry dates. The storekeeper does not need the paper waybill.

  2. 02

    The storekeeper takes photos

    The app prompts what to shoot: the whole pallet, the labels, close-ups of damage. A blurry photo is flagged for a retake right away, while the truck is still at the gate.

  3. 03

    The model reads the photos

    It counts packages, reads label text and barcodes, and marks dents, tears and signs of moisture on the packaging.

  4. 04

    Check against the order

    It compares what arrived with what was ordered: shortages, mixed-up items, wrong SKUs, shelf life below your limit.

  5. 05

    The storekeeper decides doubtful cases

    If a label is unreadable or the damage is unclear, the storekeeper confirms on screen. The final call belongs to them, not to the model.

  6. 06

    Record it in your systems

    The intake, discrepancies and the report with photos go to 1C or your WMS. Photos are linked to the delivery and easy to pull up for a claim.

Which scale is yours?

One task, three different projects. Pick the one that looks like you and press “I want this” — your request arrives tagged, and we send an estimate in 1–2 days.

Small business

common start

One warehouse or one intake dock, deliveries every day, and supplier claims rest on your word against theirs.

What we build

  • One intake point and one group of goods or raw materials
  • Phone or handheld app with capture prompts
  • Package counting, label reading, damage detection
  • Discrepancy report with photos in PDF
Integrations
Export to 1C or Excel, Telegram alerts
Where it runs
Cloud: no infrastructure needed on your side
Timeline
Guideline: 4–8 weeks to the first working version

Mid-size business

common start

Several warehouses or intake points, many suppliers, and the whole intake process runs through 1C or a WMS.

What we build

  • Separate scenarios for raw materials, packaging and finished goods
  • Orders pulled from and results written to 1C or WMS
  • Order check: shortages, mixed-up items, expiry dates
  • Supplier reports: who has discrepancies more often
  • Support contract: we add new suppliers and label formats
Integrations
ERP (1C), WMS, Telegram, email for claims
Where it runs
Cloud or your own server
Timeline
Guideline: 2–4 months to the first working version

Enterprise

Distribution centres in several regions, one intake policy and security requirements for storing photos.

What we build

  • One set of intake rules for every warehouse
  • Roles: storekeeper, shift lead, procurement
  • Audit log: who confirmed an intake and with which discrepancies
  • Models on your servers
  • Service under an SLA
Integrations
WMS, 1C:ERP, supplier portals, internal APIs
Where it runs
Your servers or a closed perimeter: delivery photos stay inside
Timeline
Guideline: 4+ months, starting with a pilot at one warehouse

Pilot from RUB 500,000. Beyond that we price by task and hours.

Questions

Can the model read any label?

A clear, undamaged one usually yes, barcodes included. A crumpled, wet or steeply angled one, not always. In that case the app asks for a retake or passes the photo to the storekeeper.

Will the model see boxes inside the pallet?

No, it only sees what is in the frame. That is why we design the capture routine with you: which sides to shoot and where checking the labels is enough.

How does it connect to our WMS?

Through an API or file exchange, whichever suits your system. We write a connector for your 1C configuration or WMS. Timeline and cost depend on the task and hours; we estimate them within 1–2 days.

Will taking photos slow intake down?

Intake gains a few photos per pallet. In return the storekeeper does not recount everything by hand or write the report manually. We test the capture routine at your warehouse during the pilot and simplify it if it gets in the way.

More in this area

A quote in 1–2 days

Name the process that eats time. You get a pilot range, not a 40-slide deck.

Presentation ↓ PDF

Or message us

WhatsAppTelegram