The Problem

Manual processes and inconsistent data capture

Invoices and supporting information arrive in different formats and through different channels.

Much of the work involved manually extracting and entering data into spreadsheets or internal systems. This was time-consuming, repetitive and prone to inconsistency.

The issue was not simply verification. It was the effort required to make the data usable in the first place.

Limited visibility of spend and supplier behaviour

Without a centralised system, trusts had no clear way to understand:

• Which agencies were most expensive for specific roles
• When demand for cover peaked during the year
• How spend compared across schools within the trust

This made it difficult to identify patterns such as:

• Repeated use of higher-cost agencies
• Late bookings leading to inflated rates
• Duplicate or overlapping bookings

The lack of visibility meant inefficiencies persisted.

Reactive decision making under pressure

Hiring decisions were often made quickly, without access to relevant data.

Trusts defaulted to familiar agencies or immediate availability, rather than making informed cost-based decisions.

Over time, this led to increased spend with limited understanding of why.

The Solution

The Curve designed and built a platform to structure and centralise agency recruitment data, using AI to extract and standardise information from invoices while removing the need for manual data entry. This created a consistent dataset across all agency activity, which was integrated into real-time reporting to provide visibility of spend, supplier performance and cost patterns. By replacing fragmented processes with a single, structured system, Multi-Academy Trusts reduced administrative effort, improved cost control and established a scalable foundation for managing one of their largest variable expenses.

Our Approach

We began prioritising structured data capture and visibility before introducing automation.

Understanding and structuring the process

We worked closely with Edaro to understand how agency recruitment and invoicing worked in practice.

This involved mapping how data flowed through the system, from invoice receipt through to reporting, and identifying where manual effort and inconsistency were introduced.

The goal was to create a structured process that reduced friction and made data consistently usable.

Reducing manual effort through intelligent data capture

A key priority was removing the need for manual data entry from invoices.

We introduced an AI-supported approach that extracts key information from invoices as they are uploaded, including candidate details, dates, rates and hours worked.

This does not replace human judgement, but it significantly reduces the time required to process large numbers of invoices to ensure data is captured in a consistent format.

What was previously a repetitive administrative task is now handled automatically, allowing teams to focus on higher-value work.

Creating a single, reliable view of agency spend

Once captured, invoice and timesheet data is structured and made available through a central reporting layer.

This provides trusts with a real-time view of their agency activity and spend.

They can now:

  • Compare agency rates across similar roles

  • Identify higher-cost suppliers

  • Track spend patterns across the academic year

  • Understand how different schools within the trust are performing

This visibility is what enables better decision making.


Enabling better decisions, not just automation

The value of the platform is not just in processing invoices, but in what the data enables.

By having a consistent, reliable dataset, trusts can identify trends and inefficiencies that were previously hidden.

This includes:

  • Recognising when late bookings lead to higher rates

  • Identifying agencies that consistently charge more

  • Understanding where demand is increasing and planning accordingly

This shifts the model from reactive administration to proactive spend management.

The Results

01.

Significant reduction in manual processing: Edaro estimates that MATs using the platform are saving up to 5% of total agency recruitment spend.

02.

Improved visibility and control: MAT leadership now has real-time visibility of agency spend across roles, suppliers and schools. Patterns that were previously hidden are now clear, enabling trusts to challenge costs and manage budgets more effectively.

03.

More informed, data-led decision making: Decisions that were previously made under pressure and based on limited information can now be supported by data.

By introducing AI-supported data extraction and eliminating fragmented manual processes, Edaro significantly reduced the risk of: inconsistent data capture, limited spend visibility, inefficient supplier selection and uncontrolled agency costs

Their Thoughts

“Hundreds of invoices were being checked by hand. Now the AI does it automatically, catches the discrepancies before payment goes out and feeds everything into management reports so we can actually see what we're spending and challenge agencies on it. We believe we're saving MATs up to 5% of their agency spend. For schools that can't afford books, that makes a real difference.”

Gary Redman

Founder and CEO