At AHTD Association for High Technology Distribution Spring Meeting, a session from Marcus Sheridan posed a simple question to the room: which is the future of online search? Option A: we are shown dozens of blue links to choose from. Option B: we receive an immediate answer, a justification of that answer, and the ability to take action on that answer.
The room knew the answer. So does every buyer who has used ChatGPT in the last year to research a purchase instead of opening ten browser tabs.
But for industrial distributors, the real implication of that slide goes deeper than search behavior. It goes to the heart of how technical buyers want to interact with your company at different stages of their journey, and what that means for the people inside your organization who are responsible for answering them.
The self-service shift is not new. The scale of it is.
B2B buyers have been moving toward self-service for years. But the numbers have reached a point where they are hard to ignore. According to Gartner, 75% of B2B buyers now prefer to navigate purchases without interacting with sales representatives, at least in the early stages of their research. Buyers are 80% through their buying process before engaging with a sales rep, per Forrester. And around 89% of B2B buyers now use generative AI as a key information source. This number is in the same ballpark (94%) as the one presented in Marcus Sheridan's presentation.
This is not a generational quirk. 37% of B2B buyers now consult AI before Google when researching purchases; documented in multiple 2025 and 2026 industry studies, not a projection. The buyer who used to call your inside sales team to ask a quick product question is increasingly asking an AI first.
There is a specific reason buyers prefer AI for certain questions
The self-service preference is well established. What is less discussed, but equally important, is why buyers prefer AI specifically for exploratory, early-stage technical questions.
Research from Ohio State University and the University of Notre Dame found a consistent pattern: people preferred interacting with chatbots when they felt that what they were asking might expose them to judgment. "Consumers feel less embarrassed because chatbots don't have the level of consciousness and ability to judge them," the researchers noted. The University of Kansas confirmed the same dynamic: when dealing with sensitive or uncertain situations, people prefer the anonymity and nonjudgmental nature of AI chatbots.
This is not a criticism of inside sales teams. It is a description of human nature in the early stages of any learning process. Think about what it actually feels like to be a newer engineer, a buyer who has just changed industries, or even a seasoned professional walking into unfamiliar product territory. Before they are ready to have a real conversation with an expert, they need to orient themselves. They need to ask the questions that feel too basic.
An AI agent creates that space. You can ask the same question five different ways. You can admit you do not fully understand the first answer. You can backtrack, explore, and refine without any sense that you are being evaluated. The practical consequence is significant: buyers who would not call your inside sales team to ask a basic configuration question at 10pm on a Tuesday will absolutely type that same question into an AI agent on your website. And when they do find a clear, accurate answer, they show up to their eventual conversation with your rep better informed, more specific about what they need, and further along in their decision process. The AI does not replace the rep. It prepares the buyer for the rep.
What this means for your inside sales team
Here is the tension every distributor with a strong inside sales team already knows. Your best technical reps are not spending all their time on your most complex, highest-value interactions. They are also answering the quick spec checks, the "what is the difference between these two part numbers" questions, the routine compatibility queries that come in throughout the day. That work matters. Customers deserve accurate answers regardless of how simple the question is. But it is not the highest and best use of someone who has spent years building deep product and application knowledge.
AI-powered tools can quickly deliver relevant product information, answer common questions, and guide prospects through complex offerings without the need to wait for a human representative. That capability does not diminish the rep. It liberates them. When an AI agent handles the exploratory, routine, and early-stage queries accurately and at scale, your reps get to spend more of their time on the interactions that genuinely require them: the complex application questions, the accounts that need relationship investment, the situations where judgment, experience, and trust are what close the deal.
This also works in the other direction. Reps themselves can use AI agents as a tool, not just a customer-facing channel. A technical question that would have required a ten-minute search through a product catalog or a call to a product manager can be answered in seconds through a well-grounded agent. That speed compounds across dozens of interactions a day.
The immediate answer is the new minimum
Marcus' slide framed the future of search as: immediate answer, justification, ability to act. That is not just a description of what AI search does. It is a description of what buyers now expect from every vendor interaction at the moment they need information.
Buyers want to configure products, get answers, and move forward without phone calls or demos. Content needs to answer the complex questions that buyers previously asked sales teams. Brands that fail to offer self-service options risk losing deals to competitors who do.
For industrial distribution, this creates a specific gap. The questions buyers most need answered: compatibility, configuration, specifications, lead time for specific part combinations, are exactly the questions that are hardest to answer without either a knowledgeable human or a very well-grounded AI on the other end. A generic chatbot cannot answer them. A static FAQ page cannot answer them. But a well-built AI agent, grounded in structured knowledge of your actual product lines, can handle them accurately and instantly, at any hour, and route anything that genuinely needs a human directly to the right person.
The goal is not fewer reps. It is better conversations.
The distributors who will get the most out of AI are not the ones who use it to reduce headcount. They are the ones who use it to raise the quality of every human interaction that remains. When buyers can get quick, accurate answers to their early-stage questions without waiting for a rep, the conversations that do involve a rep become richer, more specific, and more likely to lead somewhere.
One important caveat: not all AI is built for this. A generic chatbot or simply using MS Copilot will not hold up under real industrial product queries. If you want to understand why accuracy is the hardest part of this problem — and what architecture actually solves it — I covered that in depth in the first article in this series: "Hallucinations Are Not a Bug. They're How LLMs Work."
At ReshapeX, we build AI agents designed for exactly this model. Agents that handle the routine and exploratory technical queries accurately and at scale, and that give inside sales teams both the time and the information they need to do their best work. The rep who used to spend ten minutes searching a catalog before answering a customer now gets that answer in seconds, and brings that speed on top of the expertise and relationship that no AI will replace. The buyer who would never call your inside sales line will talk to your AI agent. The only question is whether your agent is worth talking to.