Cosentino + Celonis

Premium surfaces manufacturer Cosentino is optimizing and automating its global business operations with artificial intelligence (AI)

As a test case, it worked with Celonis to build and deploy an AI agent to streamline the management of credit blocked orders. With the agent increasing employee productivity, improving customer experience, and optimizing working capital, Cosentino now plans to introduce an entire digital workforce of AI agents. These agents’ aim? Growth. Sustainable business growth.

Context: Sustainable growth through AI innovation

The Cosentino Group is a Spanish, family-owned company, operating globally and employing around 6,000 people. It produces innovative, sustainable, high-value surfaces for architecture and design, and distributes them in more than 80 countries worldwide.

Cosentino is growing rapidly. But with highly complex sales, logistics, and production processes that require a vast workforce to perform manual tasks, that growth could be hard to sustain.

As a result, Cosentino is looking to AI to automate and optimize business operations. It wants to introduce a digital workforce of AI agents to supplement its existing employees. As Global CIO Rafael Domene explains, “We realize that one way to improve – and improve really fast – is not only automating and changing the systems, but automating what people are doing.”

Challenge: Too many blocked orders to handle

Following initial process discovery, Cosentino realized a large number of sales orders were being blocked because they violated the company’s credit policy. For each blocked order, a credit manager needed to manually review the issue and liaise with finance to analyze multiple factors – including the customer relationship, payment history, and risk – to determine whether the order should be unblocked.

This manual process could delay orders by up to a month, sometimes beyond the expected delivery date. As Cosentino products are used in building projects, which have a lot of dependencies, customers need their orders delivered on time. Any delay can have a significant impact on the rest of the project. Not only did delays have a detrimental impact on customer experience, they also hit revenue and working capital.

Solution: An AI agent for order management

Cosentino worked with Celonis to build an AI agent to solve this considerable challenge. “Our goal was to make an agent that could take care of a process on behalf of, as good as, or even better than humans,” says Domene.

In just seven weeks, Celonis was able to gather the necessary insights and requirements from the Cosentino team, pull information from multiple IT systems, structure the data, and build and deploy the agent using AI Annotation Builder.

The AI agent analyzes each blocked order, using all relevant data including order value, customer credit information, and process-driven KPIs such as block removal rates. In seconds it provides a recommendation to either unblock the order or to maintain the block and investigate further, and it supplies credit managers with detailed, fact-based reasoning they can review with ease. Some orders can be unblocked automatically. For others, credit managers can choose to accept the decision and release the order with the click of a button in the Credit Block App, or to reject the decision and provide feedback for the agent to learn from.

The AI agent now analyzes around 180 blocked orders each day and Cosentino is seeing three key benefits:

1. Increased productivity: Credit managers can handle at least five times as many blocked orders as they did before, without any additional risk. They have more time to focus on value-add tasks, such as reviewing and taking action on high-risk orders.
2. Better customer experience: A considerable reduction in the delivery delays that were caused by unnecessary order blocks is resulting in a better experience for Cosentino customers.
3. Improved working capital: A decrease in the time taken to review, unblock, and deliver the order, means invoices are sent and payments received more quickly, which positively impacts working capital.

Vision: A digital workforce of AI agents

The agent for credit block management is just the tip of the AI iceberg for Cosentino’s ambitious plans. It aims to introduce an entire digital workforce of AI agents to perform tasks humans cannot, enable its existing employees to be more productive, and support sustainable growth. It’s already working with Celonis to implement additional use cases within order management, and expanding to accounts receivable and inventory management.

Cosentino will use Celonis to identify gaps that need to be filled, or manual processes that can be effectively automated. It will use the Process Intelligence Platform to build and deploy AI agents, to provide those agents with the data and business context they need to function effectively, and to monitor their activities. Domene outlines: “We need to build all these AI agents but the key here is to give them the right visibility, give them the right data, explain the processes and, more importantly, be able to monitor what they are doing. And that is what Celonis provides us.”