Element uses Dynamics 365 Customer Insights and AI to discover selling opportunities and drive customer retention

No way to track churn

Businesses in competitive markets understand the need to keep their customers engaged, anticipate their next order, and work hard to retain them. This was especially true of Element, which had grown quickly through acquisitions. Some customers worked with Element every month, while others were occasional, and project-based. “A large percent of our revenue is quoted, and the rest is rolling revenue from contracts,” says Rob Slater, Commercial Program Director at Element. “This added to the challenge of identifying which customers may need additional testing support and when to contact them.” In fact, an analysis revealed that a percentage of customers were falling out of regular engagement with the company. This potentially jeopardized sales forecasts, which would ultimately affect revenues.

With nearly 60,000 customers worldwide annually, the 500-person sales team was doing a bang-up job bringing business in. However, it lacked an easy way to identify which customers were dropping off over time. Large accounts had dedicated account managers, but churn risk and trends of smaller and ad hoc customers were harder to track. Typically, problems weren’t discovered until quarterly revenue reports were produced.

With no possibility of dramatically expanding the size of its sales team, leadership began searching for software solutions that use machine learning and automation to find at-risk accounts and upselling opportunities. Their secondary goals were to gain a complete view of the company’s data and better understand its customers’ needs.

A model for customer retention

Element was already using Dynamics 365 Sales, Customer Service, and Human Resources when leadership learned about Dynamics 365 Customer Insights. At the time, the product was a new standalone solution for unifying transactional, demographic, and behavioral data with customer personas, related information and AI to identify at-risk accounts and timely upselling opportunities—exactly what Element needed.

In 2021, Element decided to move ahead with the customer data platform (CDP) capabilities in Dynamics 365 Customer Insights and selected Edit as a solution integration partner. Edit specializes in innovative, data-first approaches to personalizing Dynamics 365 Customer Insights for its clients.

Customer Insights would do much of the heavy lifting, but with such a diverse range of customers and industries, Element needed highly specialized features that weren’t available out of the box. Edit proposed a bespoke model based on Azure Machine Learning, which its data scientists would build. The model would look for opportunities and flag accounts that met a predefined churn risk threshold using a combined score based on industry, past behavior, and other key factors. A machine learning algorithm would identify various risks based on the score, buying patterns, won revenue history, and timeline deviations.

Working in the background, Customer Insights would act as Element’s “unification engine” to uncover data and the places where it could be used most effectively. This information would be mapped to the right salesperson, and even added to their calendar as daily action items. Ideally, this would help the sales team be more efficient and turn more calls into deals. As Rob Slater explains, “We want to improve the efficiency and effectiveness of our sales teams. With Dynamics 365 we can provide guided sales actions – automated marketing is then the force-multiplier that allows us to engage with relevance and at scale. Any time you save on admin you can use for meaningful conversations.”

The sales team knew it would have to shift from Excel spreadsheets to a whole new world of workflows, reports, and dashboards. Although the old ways were inefficient and yielded inconsistent results, everyone had a rhythm and knew how to use the tools. However, after watching some demonstrations, they could see the potential for their business unit. 

A fit-for-purpose solution

With a plan in hand, the Element and Edit teams got to work gathering and documenting customer knowledge and requirements, developing machine learning algorithms, preparing for data migration, testing, and refining. Given the amount of customer data manipulation that’s required to discover trends and identify at-risk customers, all data was migrated to an Azure Data Lake. From there, it was pushed to Azure Machine Learning, Customer Insights and Dynamics 365 Sales.

The model developed by Edit’s data scientists was put into production using native capabilities in Dynamics 365 to connect Customer Insights with Azure Machine Learning. Results would appear on familiar dashboards in Dynamics 365 Sales to make training easier and facilitate adoption. "Unlike the previous 'needle in a haystack' situation at Element, the new solution was designed to provide actionable information that is identifiable, preventable, and valuable," explains Alex Holt, Edit's Strategic Consulting Director.

After four months of collaboration and regular communications, proof of concept, testing, a churn risk pilot program with one group, refining the data model, promotion, and internal training, Element cut over to customer data platform in Dynamics 365 Customer Insights in October 2022 and has never looked back.

Certifiably successful

The sales team showed some natural resistance to change in the beginning. After sales reps got a bit of training and started receiving automated notifications, they quickly understood how game-changing the new system would be. Churn risk activities informed them about accounts to follow up on and included specific insights to investigate further, saving them hours of time and effort.

"Dynamics 365 Customer Insights is our unification engine and helps identify trends and reduce churn," said Renae Leary, Element's CCO. "It's early days, but the investment is already paying off and improving sales outcomes."

By building on its existing Microsoft investments, Element established a single platform for all company data, reporting, actions, and insights. The integration was straightforward. Although the company isn’t using all the system’s capabilities yet, the benefits have been evident since go-live. These include:

• Fifty percent of automated daily assignments lead to immediate or potential work. Salespeople contact customers in their geographic region and industry-specific testing area and are now confident they’ll get a positive response when they pick up the phone. By comparison, past cold call technique yields hovered around 9 percent.
  
  
• Churn is easier to identify and mitigate. Customer Insights models customer behavior and value histories along with typical buying patterns by lab, test type, or market identified by seven pretrained factors. When deviations are detected, the sales team is notified and can take action to proactively re-engage.
  
  
• The model can learn and improve. Salespeople can dispose and validate the identified churn risks, providing constant feedback and data to the model so that its forecasts get more refined and accurate over time.

Before Dynamics 365 Customer Insights, the Element sales team often couldn’t see the forest for the trees. The data modeling algorithms of Customer Insights have given them a deeper, game-changing understanding of their accounts. Two members of the global sales team sum up how the team felt about the new solution after go-live: “The churn risk data really focuses the sales team’s attention to customers we should be calling,” and, “I like that they pop up in your daily call diary, and it’s so much easier to get through to the correct decision maker and have productive conversations.”

Intelligence for the future

In parallel with these positive results, Element is moving from its various on-premises enterprise resource planning solutions to Dynamics 365 Finance. The project is expected to go into production in 2024 and will surface more details for Customer Insights enabling an even clearer view of customer accounts.

The next phase of sales support will come from adding real-time journeys to Dynamics 365 Customer Insights. Automated emails will be sent to smaller accounts to prompt engagement, complementing existing touchpoints such as newsletters and webinar announcements. This automation will free salespeople to focus on the right customers at the right time. The hope is that these efforts will yield more long-term, higher-revenue accounts while providing better customer service to every type of Element customer.

To better support upselling activities, soon insights that are revealed by the new algorithms will point to other products a customer has that may also require testing. Customers aren’t always aware that Element has facilities in other locations to handle their testing needs. A call from a sales rep can quickly lead to new work.

Element is also investigating using Dynamics 365 Copilot to uncover even more actionable insights across all business applications and enable their teams to work more efficiently. "Given the benefits we're already receiving from our models, we're excited to learn how new AI capabilities in Microsoft Copilot can take some of the repetitive tasks off of our salespeople," said Slater. 

For Element, the future is “up and to the right” and Dynamics 365 will be key to supporting that trajectory.