Data and the Transformation of Recovery Audits

Data and the Transformation of Recovery Audits

Recovery audits are great at getting back money lost to overpayments, but wouldn’t it be better if the money was never lost at all?

Any organization serious about protecting margin has already built recovery audits into their finance processes. What’s changed is the way audits are conducted. The speed at which audit information can be turned into useable data, and the analytical insight that can be gained, are shifting the focus of audits from backward-looking to preventative.

 

Stopping profit loss in a digital world

Traditional recovery audits are post-payment reviews that look to recoup funds lost due to errors, overpayments and under-deductions. By conducting careful analysis of vendor payments, companies aim to minimize leakage from a steady drip-drip of losses.

At the end of the process, a large company will often recover millions of dollars in lost profit. Now advances in artificial intelligence (AI) and machine learning are enabling finance teams to go a step further.

Analysing data is now enriching the audit process by identifying exceptions and validating them against all the information companies capture about vendors, prices, environmental data and regulations.

 

The data-driven recovery audit

Modern recovery audits start by considering all of the source-to-pay (S2P) data a company has access to. Once that information is collected and consolidated, it can be analyzed to uncover:

 

  • Funding shortages for specific promotions
  • Incorrect application of payment terms
  • Incorrect pricing
  • Missing rebates and discounts
  • Duplicate payments
  • And more

 

With AI in the mix, recovery audit analysis tools can rapidly apply logic that would have lived in the heads of senior auditors or been hidden in spreadsheet macros. Now those decades of experience and expertise can be translated into algorithms that apply it on-demand.

 

The Benefits

 

Fixing leakage where it lives

To head off erroneous, fraudulent, or non-compliant payments, businesses need a way to assess risk from all the information gathered during the recovery audit process.

An effective solution uses a methodology to measure risk based on intelligence gathered from previous audit experience, from which businesses can adapt and learn.

 

Making claims management simpler

The ability to find, assess and provide backup for claims can also facilitate payments back from the supplier.

Being able to send a claim online for review and approval leads to faster issue resolution and minimizes the possibility of ‘vendor abrasion’ – aggravating otherwise solid supplier relationships with claims that turn out to be false-positives.

 

Enabling a broader view

Critical information that can help identify overpayments exists in vendor databases and payments systems, but it can also be found in other business systems.

By considering relevant data from across the enterprise like email, collaboration platforms, and CRM, transaction data can be enriched by unexpected information about vendors, prices and contract terms.

 

Managing unstructured data

Unstructured file types like email attachments, images  and PDFs can bury (or make it time-consuming to access) loads of essential information. Physically opening contracts, invoices, images and receipts to verify each transaction isn’t scalable.

A recovery audit solution that applies machine learning can understand and extract key data points like dates, amounts, terms and spend categories from unstructured documents, regardless of format and irrespective of language and country.

 

Stop leakage before it happens

There’s now an opportunity to start identifying sources of leakage earlier. Traditional recovery audits discover errors six months to a year after the fact. With systems in place to determine the risk areas for leakage, there’s an opportunity to catch overpayments earlier: directly after a transaction, while they are underway, or even before they occur.

Artificial intelligence and machine learning can detect vital information across multiple document formats, which can help make recovery audits more efficient and accelerate recoveries for clients.

On its own, however, technology can’t deliver the nuance, discretion, experience and professionalism needed to manage supplier relationships or identify opportunities to make processes better.

To really be effective, AI-powered solutions have to draw from learnings and information acquired over decades of recovery analysis and real-world engagement. That historical information will add important context to data held on contracts and payments systems, enabling better risk assessment.

 

Want to learn more?

Check out the How Artificial Intelligence and Machine Learning are Transforming Recovery Audit white paper or the How Data and Technology are Transforming Recovery Audit webinar to find out how AI and machine learning are transforming the industry to help prevent leakage from occurring in the first place. 

Share this post