Sales teams that run Odoo still collect leads the analog way: a stack of business cards from an expo booth, a folder of card photos on a rep's phone, a WhatsApp thread of snapshots after a conference. Someone then keys each one into crm.lead or res.partner by hand — misspelling company names, dropping VAT numbers, and letting warm contacts go cold while they sit in a data-entry backlog. Odoo core has no card-to-contact capability at all: there is no OCR pipeline, no image-to-field mapping, and the standard lead form assumes a human is typing. Even the closest off-the-shelf apps.odoo.com module tends to ship a fixed field map and a single OCR engine you cannot swap, which rarely matches a specific country's card conventions or a team's existing lead-qualification workflow.

ECOSIRE builds a made-to-order module that closes that gap end to end. We add a capture surface — a mobile-friendly OWL widget and a QWeb-rendered upload action — so a rep photographs a card and the image lands as an ir.attachment on a new record. A dedicated model (for example card.scan, extending models.Model) then calls an AI recognition service to detect the card's text regions and classify them into name, job title, company, phone, mobile, email, website, and postal address. A set of @api.depends compute methods and normalization rules map the raw extraction onto real Odoo fields, resolve or create the res.partner company, de-duplicate against existing partners by email and phone, and spin up a crm.lead (or a contact under an existing company) with the source, medium, and campaign already stamped. Low-confidence fields are flagged for a one-click human review rather than written silently, so the data your reps trust is data a person confirmed.

Everything is engineered to your instance, not bolted on. Access is governed by ir.model.access.csv plus record rules so a rep only sees their own scans while managers see the team's; automated actions (base.automation) can route a freshly created lead to the right salesperson, tag it, or trigger a follow-up activity. The module ships as a proper addon with its own __manifest__.py, versioned views, and demo data, and it exposes its create-lead path over Odoo's XML-RPC/JSON-RPC API so an existing badge scanner, event app, or kiosk can push cards in without touching the UI. We build against Community or Enterprise as your edition requires — calling out any Enterprise-only dependency up front — and target whichever of Odoo 17.0, 18.0, or 19.0 you run.

Delivery is build-to-order. After a short scoping call we confirm the exact field map, the recognition provider and language coverage, your de-duplication rules, and which teams and pipelines the leads flow into, then send a fixed quotation. Typical delivery is 2 to 4 weeks from confirmed scope, and you receive the module on a staging instance for UAT before it ever touches production. Pricing starts from $299 (indicative, single-company base scope); multi-company routing, additional card languages or scripts, integrations with an external event/badge system, and bulk migration of an existing backlog of card images all increase the quoted scope.