Research · August 2026
We talked to nine sales, commercial and product leaders at industrial manufacturers to find out where AI is actually being used in their businesses, and where it is not. This is the short version.
7 of 9
already give their teams access to AI tools
8 of 9
have no standard way of using them
0 of 9
apply AI to directly support the technical sale
The headlines suggest this industry is further along than it is. Very few of these companies have moved past individual experimentation, which means the advantage is still available to whoever goes first.
What we found
01
Access is not the problem. Nearly every company we spoke with already has licenses in the building, whether that is universal ChatGPT seats, subscriptions handed out on request, or Copilot across the business. What they do not have is a standard way of using them. Very little is documented, and almost nothing is measured.
The first-mover advantage in this segment is still on the table.
Where AI does show up, it is drafting outreach emails, summarizing meetings and helping individuals keep track of their own pipeline. Useful work, but not the hard part. Nobody we spoke with had applied it to quoting, configuration or specification work, which is where these deals are actually won and lost.
The highest-value use cases are still wide open.
Four of the nine told us the age of their sales team is the reason they are moving slowly. We would turn that around. The application knowledge that wins specs in this industry is undocumented, and it sits with people who have a known retirement horizon. Worth noting that the heaviest AI user we met described himself as one of the older people at his company.
Capturing company, customer and application knowledge is urgent regardless of how you feel about AI.
The one thing that separated the companies making progress from the companies standing still was whether a single person had been made responsible for the outcome. Company size, budget and average age did not tell us much. The businesses with nobody in that seat were the same ones with licenses sitting unused across the organization.
Naming that person is the first decision to make, and it does not cost anything.
Where companies sit
02
Every company we spoke with had landed somewhere different, and in each case the position was the residue of an unrelated decision. A procurement default. An IT policy. One manager's initiative.
Most of the companies we interviewed sit here
1
Nothing at all
No tools, no evaluation underway, no vendor conversations.
2
Licenses, no direction
Everyone has access. No standard, no shared practice, no measurement.
3
Subscriptions on request
The company pays if you ask. Nothing mandated, nothing coordinated.
4
One approved tool
IT sanctions a single platform and blocks the rest. Access exists, capability does not.
5
A funded platform
A purpose-built system bought for one workflow, with leadership behind it.
→↓
AI-native
Standardized use on a shared data layer that compounds over time.
No one is here yet
Who we spoke with
03
Interviews
Nine
Fieldwork
July to August 2026
Revenue range
$50M to $500M
Basis
Not for attribution
Interviews ran between July and August 2026 with companies in the $50M to $500M annual revenue range. We asked where AI is applied in sales and marketing today, which tools are in use and who owns them, how standardized any of it is, and where leaders think this is heading. Conversations were not for attribution, so nothing here identifies a company or an individual.
How Skylite helps
04
Roadmap
Figuring out what is worth doing, and what is not. Workflow mapping, opportunity prioritization, feasibility testing, and a sequenced roadmap, including where AI is not the answer.
Engineering
Building the data plumbing that makes everything else possible. Connected commercial data, governed machine access, a clear source of truth, and process knowledge captured out of people's heads.
Impact
Custom quoting, decision support, copilots and automation built on that foundation, then rollout, training and refinement until they are genuinely in use.
Most engagements run in this order, but we meet you where you are.
It covers what separated the companies that moved, our six-step framework for getting a first proof inside ninety days, and what an AI-native industrial business actually looks like. Leave your email and we will send it over.
Skylite is an AI strategy and implementation firm working with lower-middle and middle-market companies. We ran these interviews ourselves, one conversation at a time, because we could not find an honest picture of what this industry is actually doing.
If you want to compare notes on where your organization sits relative to this group, give us a shout. No pitch, and we are happy to share what we are hearing.