Personalization in B2B commerce is not controversial. Research shows 80% of buyers are more likely to purchase from companies offering personalized experiences. The question is whether artificial intelligence is the right tool to deliver it, or whether simpler approaches may work better for most B2B companies today.
The state of AI in B2B
One-third of digital leaders believe AI will be a game-changer for their organizations. Applications include customer segmentation, product recommendations, keyword search, product categorization, fraud detection, and marketing automation. Two-thirds of B2B buyers now prefer remote human interactions or digital self-service over traditional sales engagement.
Expert predictions suggest that by 2033, AI and automation will be able to perform 40 to 50% of all work. If that projection is accurate, many B2B companies are significantly behind in their AI investments.
The debate format
Team AI is Here featured Brooke Logan from Genuine Parts Company (NAPA) and Jason Hein from Bloomreach, a founding member of the Amazon Business team. Team AI Not Ready included Faisal Hussain from MSC Industrial and Doug Topken from Pool Corp. Both teams brought practitioner experience from major distributors navigating AI decisions.
Round 1: Is AI producing measurable results?
Faisal Hussain argued that before discussing results, companies must acknowledge that most lack the data infrastructure needed for AI. Customer behavior, clickstream data, transactional history, and single source of truth remain fragmented across systems. Without that foundation, AI cannot deliver on its promise.
Doug Topken agreed, noting that many companies are lacking fundamentals. He compared it to children who cannot tell time on analog clocks without their phones: companies need to learn basics before reaching for advanced tools.
So many companies are lacking in the fundamentals. Get the fundamentals down. Learn the basics. We need good salespeople. Machines cannot add value the way we can yet. Someday maybe, not today.
Doug Topken, Pool Corp
Jason Hein from Bloomreach countered that the tools available today are significantly more advanced than even three to five years ago. Companies deploying AI are seeing real results: HD Supply achieved 16% increase in revenue from search and 4% increase in add-to-cart rates. The organizational benefits extend beyond financial metrics to faster problem identification and resolution.
Brooke Logan added that results span the entire funnel: return visitors, add-to-cart rates, average order value, and conversion. The data is there for companies willing to start collecting and using it. Waiting means falling so far behind that catching up becomes impossible.
If you do not get started now collecting that data and using that data, you are going to get so far behind that you cannot ever catch up. Start getting the data and incrementally learn. Train the AI over time.
Brooke Logan, Genuine Parts Company
The audience voted 68% that AI is real rather than hype.
Round 2: Will AI replace salespeople?
Jason argued that the salespeople who will be replaced are those who do not add value: order takers and those handling operational tasks like proof of delivery. An AI can handle replacement orders and add relevant recommendations at scale. If companies want to hire a million salespeople to match that degree of targeting, they are welcome to try.
Doug Topken pushed back, noting that order tracking and simple automation are not AI. Those capabilities existed in 1996. True AI-driven salesforce replacement remains hype. The most valuable thing in B2B is a live salesperson who can add value through human interaction.
Faisal Hussain argued for a hybrid approach. B2B will always require human trust built through personal interaction. AI should enable salespeople with data and insights rather than replace them. The customer relationship depends on that trust.
Brooke Logan shared examples of building tools that arm salespeople with data. No salesperson wants to dig through historical trends to identify where a customer is underpenetrated or why purchasing patterns changed. AI surfaces those insights so salespeople can have valuable conversations instead of doing analysis.
The audience voted that AI will not replace salespeople.
Round 3: Does AI have a sweet spot?
Faisal Hussain acknowledged that AI is playing a role in logistics, where prediction of inventory and delivery timing is well-established. The challenge is applying that to complex B2B scenarios like freight forwarding where service providers are not sophisticated enough to integrate.
Doug Topken maintained that AI is a shiny tool that companies have not learned to use. The fundamentals with existing tools are not mastered. Saving data for future AI use makes sense, but the investment should wait.
Jason Hein argued that what customers want from B2B sellers is to talk to someone who knows what they are doing. The only way to create an authentic expert-driven digital experience is to deploy AI across customer selection and application bases. Manual recommendations lead to obvious failures like recommending more big screen TVs to someone who just purchased one.
Brooke Logan confirmed that AI is perfect for long-tail products that businesses, salespeople, and customers may not even know exist. Walking through distribution centers reveals products no one realized the company sold. AI surfaces those discoveries digitally at scale.
How many times have you bought a big screen TV and then the next week the recommendations give you more big screen TVs? That level of thoughtlessness about how you deploy recommendations results in customers developing distrust of your expertise. AI helps you overcome that at scale.
Jason Hein, Bloomreach
The audience voted that AI will achieve breakthrough status within five years.
The verdict: AI is real, but fundamentals matter
Team AI is Here won two of three rounds. The debate revealed important nuances despite the overall verdict.
First, companies must realistically assess their digital maturity before investing heavily in AI. Many B2B companies remain at early stages of e-commerce without the data infrastructure needed to make AI effective.
Second, AI at scale is most applicable to distributors with large catalogs where manual approaches cannot cover the breadth of products and customer applications. Manufacturers with smaller, more complex product lines may find other approaches sufficient.
Third, the training data advantage means companies that delay may never catch up. Unlike traditional software purchases, AI systems coupled with learning data create competitive moats that cannot be purchased later. The comparison to Google search is instructive: Microsoft invested heavily but could not close the gap once Google had the learning advantage.
The closing argument emphasized that people tend to overestimate what will happen in three years and underestimate what will happen in seven. Given the pace of innovation since the iPhone launched, five years may bring more change than B2B companies expect.