Skip to content
C—01 — Case studyBeta
Aurelia.

Turning a dead business card into a live relationship record

Building a capture-to-follow-up pipeline where optical character recognition, retrieval-grounded enrichment, and relationship scoring all have to agree — in about three seconds, on a phone.

Study recordC—01
Subject
Aurelia
Sector
Relationship intelligence · B2B SaaS
Period
2025—
Status
In beta on Android. iOS in progress.
Property
www.tryaurelia.app
Operator
Cognimit Technologies LLP
§ 01Context

What existed
before this.

Contact capture apps solve the wrong half of the problem. Digitising a card is a solved technical task; remembering why the contact mattered is not. We built Aurelia around the second half and treated scanning as the entry point rather than the product.

The problem, as constraints

  • 01Optical character recognition on a photographed card produces noisy, unordered text with no field semantics.
  • 02Enrichment from the open web is unreliable and will confidently return the wrong company if unconstrained.
  • 03A relationship score is worthless if the user cannot see why it moved.
§ 02Approach

How we
built it.

The shape of the solution, stated at the level a reviewing engineer could argue with. Nothing here is a feature list.

01

Structured extraction runs server-side over the raw text so the field mapping can be improved without shipping a new app build.

02

Enrichment is retrieval-grounded with mandatory source attribution, so every claim in a company brief can be traced to a page.

03

Scoring is derived only from observable signals — data completeness and interaction recency — so it can be explained in one sentence.

04

The free tier deliberately excludes model calls, which keeps unit economics viable without a paywall on the core capture flow.

§ 03Decision records

Every call,
with what we rejected.

A decision without a named alternative is a preference. These are the calls that shaped the product, each with the option we turned down and the reasoning that decided it. Where one has since looked wrong, it stays in the record.

01

Server-side extraction rather than on-device

Why

Field mapping accuracy improves weekly. On-device parsing would have locked each improvement behind an app store review cycle.

02

Mandatory source attribution on enrichment

Why

An unsourced company brief is indistinguishable from a hallucination. Attribution makes the output auditable by the user.

03

Tiered access to model calls

Why

Inference cost scales per scan. Bundling it into a free tier would make growth directly loss-making.

§ 04Outcome

Where it
stands now.

Stated qualitatively where a number does not honestly exist yet. Market figures carry their source; we do not attach an improvement percentage we cannot evidence.

Result

  • Card to structured record in roughly three seconds
  • Over forty fields extracted and enriched per contact on paid tiers
  • Six-stage workflow from capture through to a shareable hosted profile

Status — In beta on Android. iOS in progress.

What it runs on

Android applicationServer-side OCR normalisationRetrieval-grounded enrichmentHosted web profiles and NFCSubscription billing across three tiers
The venture behind this studyAureliaYour network, with a memory.Relationship intelligence · B2B SaaS

Other case studies

Same discipline, applied to your platform

Decision records like the ones above are a deliverable on every engagement, written as the work happens rather than assembled for marketing afterwards.

WhatsAppStart a chat