A rep walks into an account and heads straight to the cooler to count inventory. Back at the warehouse, a back office of 5, 10, or even 30 people works through tasks that are still largely manual. And cash that could fund growth sits on the floor as extra inventory.
None of this is new. In March, a distributor told VIP CEO Andrew Criezis that the industry has been talking about making sales reps more efficient for 30 years. What’s new is that AI can now take on a real share of this work. The harder question, and the one at the center of VIP’s panel on AI in beverage distribution at the NBWA Annual Convention, is how to make sure the time AI saves turns into money.
Andrew was joined by Brian Kelly, VIP’s Head of Product, and Sahil Puri, a partner with McKinsey & Company’s QuantumBlack division, with VIP Head of Sales Lucinda Ochwat moderating.
Sahil described the same gap inside McKinsey. The firm built its own AI tool on a century of its knowledge, and adoption was strong: more than 90% usage, more than 80% of people coming back, and people reporting 30 to 60 minutes saved each day. The bottom line didn’t move.
“30 to 60 minutes per employee doesn’t actually mean anything for your EBITDA.”
Sahil Puri, Partner, McKinsey & Company
The time went to longer lunches and dog walks, he joked, which is good for morale but invisible at the leadership level. Results came when McKinsey picked five to seven areas, such as proposal writing and performance management, redesigned how that work got done, and retrained people to work the new way.
For distributors, the lesson is to point AI at workflows where time converts into cash, margin, or service.
VIP expects the first returns to show up in service levels and headcount productivity, whatever you implement. The panel saw that happening in four places.
AI applied to order planning and forecasting, trained on a distributor’s own business, can take real dollars out of the warehouse.
“There was a distributor I worked with who saved $5 million in inventory just from implementing one AI tool that cost, around $70,000,” Andrew said.
Forecasting and ordering may not reduce headcount, but they reduce the inventory you need to carry and put the right product in place at the right time, so out-of-stocks stop costing you downstream.
Predictive ordering now brings 80 to 90 percent of an order ready to go. Today that looks like a rep opening the iPad, reviewing an order that is already built, and changing a line or two. The time saved goes to the conversation with the retailer instead of the cooler count. As the system learns more of your context, it can also suggest the next best product to pitch in each store.
On the route, AI helps you keep hitting delivery windows as new accounts come on, so service levels hold as the business grows. With large retailers, a missed window can mean fines.
Distributors are often running back offices of 5 to 30 people. As AI takes on more of the manual work, owners can choose: take the efficiency to the bottom line, or move those people onto the projects that never got staffed.
Two habits separate a return from an experiment.
First, start where you can measure. Pick an area of the business that already has measures, understand its current state, and then apply a tool. Fewer people involved means less friction.
Second, lead it from the top. McKinsey reports fewer than 30% of companies have their CEO directly sponsoring the AI agenda. Asking IT to hand everyone a chatbot license won’t produce gains on its own. Every leader in this industry now needs to give 10% of their thinking to technology.
We don’t treat AI as a side experiment. We build it into the workflows distributors already run every day: the order on the rep’s iPad, the forecast that sets what sits on the floor, the plan that gets trucks out on time. It runs on 50+ years of beverage data and a platform trusted by 550K users across all 50 states.
As Andrew said on stage, buying the software is only the first step. You know your products, your accounts, and your delivery patterns better than anyone. The more of that context the system learns, the better its orders, forecasts, and recommendations get, and you should push us, as your partner, to tailor it further. Through all of it, your data is your data.