How Carbonapp automates its project-to-funder pipeline: from manual data entry to smart AI matching

KEY FIGURES

0 manual data entry to publish a project

100% of projects standardised automatically

4 matching criteria scored automatically

‍

THE CHALLENGE

Carbonapp connects funders with carbon offset projects, a complex matchmaking business that relies on the quality of the information available on both sides. Yet the platform was facing two structural bottlenecks that reinforced each other.

On one side, putting projects online was slow and inconsistent: each project page was written manually, with quality varying from one contributor to another. On the other, matching funders with projects was done by hand, criterion by criterion: budget, tonnage, location, project type. A time-consuming process that was hard to make reliable and impossible to scale.

The crux of the problem: as long as projects weren’t presented in a consistent, structured way, no matching algorithm could be reliable.

Before the project began, several key challenges were identified:

• Automatically standardising and enriching project pages from raw documents
• Automating funder-project matching based on multiple weighted criteria
• Creating synergy between the two modules so that one feeds the other
• Drastically reducing manual data entry time and improving scalability

‍

THE SOLUTION

Qolaig designed an end-to-end system built around two components that work in tandem, with the quality of one determining the reliability of the other.

01. Automatic project page generation

From a 10 to 20-page project description PDF, the system automatically extracts the key data, enriches the information through web scraping, and produces structured, consistent and optimised narrative content. This content is pushed directly to the platform via API, with no manual data entry and no variation in quality. The same automatic document processing approach was applied to Pan Distribution’s contracts.

02. Smart funder-project matching

Once projects are standardised, the recommendation engine kicks in. Based on a funder’s profile, the API calculates a relevance score for each project using several weighted criteria: available tonnage, margin, geolocation, project type. The most relevant recommendations surface automatically, quickly and consistently.

03. A synergy that makes the system scalable

The two modules are designed to feed each other: enriched, standardised project pages make scoring more accurate, and more reliable matching increases the platform’s value for funders. This virtuous circle is what allows Carbonapp to grow without proportionally growing its headcount.

‍

RESULTS

• Automated publishing of project pages from a simple PDF
• 100% of projects standardised and structured in a consistent format usable by the matching engine
• Multi-criteria scoring (tonnage, margin, geolocation, project type) calculated automatically for each project
• Scalability unlocked: the volume of projects and funders can grow without adding to operational workload

‍

RETURN ON INVESTMENT

On a matchmaking platform, the quality of the matching is the product. Every hour spent manually entering a project or qualifying a funder by hand was an hour taken away from growth. By automating both ends of the pipeline, project enrichment and recommendation scoring, Carbonapp can now absorb an increase in volume without a proportional increase in operational costs. Scalability is no longer a goal on the horizon: it’s built into the architecture. The same logic applies at Liora, whose knowledge base grows richer with every interaction.

‍

FAQ

How much does an AI workshop with Qolaig cost?

The AI workshop costs €5,000 excl. VAT for one day on site. 50% of this amount is deducted from development fees if the project moves forward.

How long does it take to deploy a solution?

The first use cases are usually live 2 to 6 weeks after the workshop.

Is our data kept confidential?

Yes. It is never used to train third-party models and stays in your environment, with hosting suited to your GDPR requirements.

What did Qolaig deploy for Carbonapp?

A system that automatically generates project pages from a PDF, and a matching engine between funders and projects.

Which criteria does the matching use?

Available tonnage, margin, geolocation and project type, weighted and scored automatically.

Do projects still need to be entered by hand?

No. A project description PDF is enough: the data is extracted, enriched and published to the platform via API.