Distributors face many challenges in managing working capital – inventory, cash and accounts receivable – including:
- Determining the right inventory levels so that they can serve their customers, but not tie up cash in inventory they don’t need
- Minimizing late payments by customers while providing the credit they need
- Accurately forecasting demand in an uncertain market
Distribution Strategy Group’s Ian Heller recently spoke with technology leaders for in-the-field examples of how distributors can use AI to improve their working capital. The panelists included:
- Graham Smith, Business Development Manager, Esker
- Chandra Subramanian, CEO, ORS Group
- Greg Hartunian, Director, Epicor
How are distributors using these tools to improve cash on hand?
Did you miss the webinar? Watch it on-demand.
AI’s Impact on Working Capital
Smith offered this example of how AI can improve working capital:
Better managing Days Sales Outstanding (DSO) is critical to optimizing cash flow for distributors. Distributors are leveraging AI to improve collections by identifying and addressing payment delays before they become a problem. That minimizes overdue receivables and unnecessary write-offs.
Another big working capital challenge facing distributors and draining cash: inventory management. “There are two kinds of inventory forecasts: wrong ones and lucky ones,” Heller said.
This sentiment captures distributors’ frustration when trying to forecast demand in an unpredictable market. Then there’s the tension between sales and finance — the former wants to extend credit lines to keep customers happy, while the latter seeks to protect the company’s bottom line. Striking this balance is a challenge. Even a small improvement in inventory availability, from 96% to 97%, can have a significant financial impact by reducing both holding costs and lost sales.
More precise forecasts minimize stockouts and reduce excess inventory collecting dust on the shelf. Distributors can leverage AI to make more informed purchasing decisions, balancing inventory investment against forecast demand to lower risk.
Smith said the challenges we’re facing aren’t that different from when he started his career 15 years ago. “High customer expectations have always been a driving force,” he said. “Customers expect same-day shipping, which requires having the right inventory in stock, processing orders quickly and assessing customer creditworthiness quickly.”
Panelists provided other examples of leveraging AI to improve working capital:
Assessing credit risk
Understanding customer reliability and knowing when payments will come in is just as important as making the sale —after all, “cash is king.” Distributors need to reduce their exposure to unpaid invoices.
Reducing cost to serve
AI can automate or streamline order entry, invoice processing and other customer service tasks to lower operational costs and improve efficiency. This is especially important for high-cost-to-serve customers.
Data Quality and AI
A big part of optimizing working capital is optimizing your forecasts. Data is the foundation. But without reliable data, forecasts are harder to trust.
“It’s easier to get data now because many companies have already digitally transformed and migrated to an ERP system,” Hartunian said. “So, the data is now stored and structured in an organized way. It’s easy to extract. However, it’s just as difficult to convert that data into an accurate demand signal, because many companies don’t have good data governance around entering a proper sales order.”
But companies that don’t maintain rigorous standards often face challenges in demand planning.
“I don’t love the term ‘garbage in/garbage out’ because it tends to make buyers think ‘I can’t get started because my data is garbage.’ I can tell you 100% of the companies that started a project still haven’t gotten their data right several years later.”
Hartunian said adopting an AI platform, no matter the state of your data, “puts a stake in the ground that says: ‘Let’s invest in inventory optimization and forecasting software — you have a reason to fix that data to become a better inventory decision-maker.”
Some companies are emphasizing “data-light” forecasting, which minimizes the number of assumptions and leverages existing clean data to simulate realistic outcomes.
While advanced forecasting techniques may seem out of reach for some, distributors shouldn’t let data quality challenges deter them from investing in technology that can improve decision-making around working capital over time.
“If you go back many years, you’d spend millions on implementing a good forecasting solution. You don’t have to do that anymore,” said Subramanian. “It’s available and you can try it in small prototypes and that expand to the rest of the organization.”
Getting Started
Subramanian said that distributors need to first focus on their core business challenge and determine the best tools to solve it – AI or otherwise. “AI is a means to an end,” he said. “I think the real question is, what’s the business challenge? What’s the use case you’re trying to solve? Fundamentally, it’s about helping manage risk in a fast-moving environment with quantitative techniques and models achieved at scale. We need to figure out how to use techniques with the data available now to help decision-makers.”
Success with AI, like with any technology, often comes down to user buy-in and data quality. If employees don’t adopt the new solution, or if there’s turnover in the organization, the initiative is at risk.
“I recall conversations when I was more involved on the invoice-to-cash side of the house where somebody would say, ‘Hey, I just need AR automation.’ There are six different solutions for that,” Smith said. “Let’s better understand the challenges and the initiatives. What are the expectations of your leaders? Then we can work our way back to say, okay, this is a logical starting point because of these issues that you’re facing or these outcomes that you’re looking for, and then we can expand from there.”
Identifying the right starting point ensures that AI adoption leads to meaningful outcomes, with room to expand into other areas.
Hartunian looks for solid change management as a bigger precursor for successful AI adoption over good data. “It’s the need to change. There must be an acknowledgment by the business that we must change our current practices because they are just not good enough to allow us to be competitive in the marketplace. If they come to the table with that, you’re going to overcome the organizational inertia to implement change. You can’t change unless there’s a need accepted by top-down, middle management and end-users.”
Get more insights in our on-demand webinar. Watch it here.
