Deliverect launches AI solution to combat order errors and fraudulent refunds

The food tech specialist uses artificial intelligence in a solution to automate quality control in restaurant chains, with promising results in terms of revenue protection. For restaurant owners who are strongly active in home delivery of meals, this is already a blessing.

Video Deliverect

Deliverect, a Belgian SaaS platform specialized in the digitalization of catering activities, recently announced the launch of 'Resolve', a new solution based on artificial intelligence. This technology, developed on Google Cloud, is aimed at detecting and documenting errors in orders - in particular missing or incorrect products - and at providing visual evidence to dispute unjustified refund requests. A consumer who orders a meal on a platform such as UberEats or Deliveroo can file a complaint in the app if part of their order is missing or incorrect. Usually with the chance of a refund of the entire order or part of it.

According to the Belgian unicorn Deliverect, these M&I (Missing & Incorrect) errors are a major problem for chains operating on multiple sites: around 30% of refunds are said to be unjustified and account for as much as 60% of the losses associated with them.

Automate order traceability

In practice, Resolve relies on cameras above order picking stations. Each order is filmed and analyzed with AI to detect any anomalies and generate an ‘evidence package’ in case a customer complaint follows. The approach is intended both to prevent fraud and to provide constructive feedback to employees in the event of proven errors.

Deliverect’s first tests in a pilot restaurant show a 52% reduction in refund errors and a 44% reduction in refunded amounts, thanks to improvements in operational processes, through analysis of recorded visual data. The system was able to detect that 42% of the refunds claimed in one of the pilot establishments were contestable!

A strategy based on industrializing AI

This launch is part of Deliverect’s broader strategy to increase operational efficiency by automating increasingly manual tasks in restaurants. After consolidating its position in order bundling and menu synchronization, the company is now focusing on decision-making tools and quality control systems.

Zhong Xu, CEO and co-founder of Deliverect, emphasizes the importance of such solutions at a time when restaurants are faced with increasing complexity in managing their sales channels: delivery via third-party platforms, direct orders, counter service or drive-thru. By automating verification across all these touchpoints, Resolve aims to restore a fair relationship between restaurateurs and platforms.

It is initially offering its service to chains with more than 50 points of sale in the United States, the United Kingdom and France. It is compatible with the main delivery platforms (Uber Eats, DoorDash, Deliveroo).

Now confirm on a larger scale

With a 95% success rate for disputes on refunds and up to 70% protection of potential lost revenue, the figures from the pilot phase are certainly encouraging. Now it remains to be shown that they can also be generalised. Not to mention that the use of cameras and AI analysis tools in food production environments can also raise practical and ethical questions in terms of privacy, labour law, social acceptance, etc.

But Deliverect is determined to consolidate its role as a provider of intelligent solutions for modern catering. In that regard, 'Resolve' is a relevant and interesting case study for the use of artificial intelligence in solving everyday problems in restaurant chains.