What changes in a commercial team under AI.

Nearly every AI conversation with directors starts with tools. Which tool should I buy, which tool works for sales, which for marketing. But the conversation that rarely gets had is what happens to the team. Who becomes redundant. Who gets heavier work. Which role that exists now won't exist in its current form in two years.

That isn’t speculation. It’s happening now, in commercial organisations seriously building AI in. And the director who doesn’t have a clear picture of it ends up with a team redesign landing in his lap at a moment he hasn’t been able to steer.

Three patterns are taking shape in commercial teams under AI. One category of roles that disappears. One that fundamentally changes. One that, on the contrary, gets heavier and more important. Anyone who recognises these patterns can prepare their organisation deliberately.

What disappears: the production work AI takes over

The clearest shift. Roles consisting mainly of production work become largely redundant or shrink sharply.

In marketing: junior content marketers writing copy, social media coordinators planning and publishing posts, marketing operations setting up campaigns in tools, lead qualifiers scoring whether a lead is warm enough.

In sales: sales development representatives sending outbound emails and booking first meetings, sales administrators preparing quotes and updating the CRM, account coordinators handling follow-up between account managers and back office.

In customer service: first-line agents answering standard questions, escalation coordinators assigning tickets, complaint handlers responding to a script.

None of these roles disappears completely overnight. But the volume of work they used to handle shrinks dramatically. A marketing team that had four junior content marketers in 2023 can, in 2026, do the work with one director who edits AI output. A sales team that had five SDRs in 2023 can, in 2026, hit the same pipeline volume with two seniors plus AI tooling. A customer service team that had twelve first-line agents in 2023 can, in 2026, serve the same customer base with four specialists and an AI layer underneath.

This isn’t a scenario. It’s happening in organisations we work with closely. The pace varies by sector, but the direction is constant. And the team composition of four years ago no longer fits how you’d design a commercial organisation today.

What fundamentally changes: executor roles become director roles

A second category shifts. Roles that are now executors become directors. Same job title, different content.

A content marketer used to write copy. Now she briefs AI, reads output, picks between variants, edits to publishable level, decides what’s worth pursuing. That’s work closer to an editor-in-chief than a writer. Competencies shift toward judgement, taste, quality assessment under pressure.

A data analyst used to analyse. Now she receives output from AI systems that continuously analyse, and her work becomes translating that output into decisions. What’s signal, what’s noise? Which patterns deserve action, which to ignore? The competency shifts from technical (building dashboards) to interpretive (explaining what the numbers mean to whoever decides).

A salesperson used to spend his day on administration, follow-up, and meeting prep. Now AI does that. His work becomes: more meetings, and deeper ones. The competency shifts from administrative discipline to conversation skill and judgement about which deals deserve strategic attention.

For the people in these roles, this isn’t an easy transition. Anyone who’s spent ten years building skills AI now takes over has to develop different skills or move elsewhere. For the director, this isn’t an abstract HR question. It’s a question about who stays, who gets supported to grow, and who becomes happier somewhere else.

What gets heavier: strategic judgement, customer conversation, quality control

The third category is the most underestimated. There are roles AI doesn’t make redundant or change. AI makes them heavier and more important. Because fewer people do the executive work, more weight falls on the people doing the strategic work.

Whoever delivers strategic judgement gets more weight. The marketing manager choosing which position to take, which segments to walk away from, which stories to tell. That role was always important, but used to be smothered under execution pressure. Now, with AI accelerating execution, the strategic side becomes more prominent. A wrong choice gets executed faster and at larger scale. The choice itself has to be sharper.

Whoever has deep customer conversations gets more weight. An account manager who genuinely understands his customers can, with AI support, have two to three times more customer contact than before. But that only works if his conversational ability is up to it. An AI tool doesn’t make the average account manager better. It makes the difference between a good and a mediocre account manager more visible.

Whoever judges quality gets more weight. AI produces a lot in a short time. What’s acceptable for your brand, what isn’t, and who makes that distinction: that becomes a central function. In some organisations this role is already called “quality controller” or “brand editor”. In others it’s spread across roles without clear ownership. Anyone who doesn’t assign it explicitly ends up with mediocre output at scale within six months.

Three kinds of heavier work: strategic judgement, conversations, quality control. All work AI doesn’t do. All work that was underrepresented in the old balance. All work that, in the new balance, makes the difference.

What the director has to do about this

Three things, none delegable.

One: make a role-redesign plan before deploying AI at scale. Which roles we have now won’t exist in this form in two years? Which people can grow into director roles? What support do they need? Which new roles do we create (quality controller, AI owner, conversation expert)? This is HR work, but under board direction. Otherwise the pattern emerges of adding half-roles for AI everywhere without removing anything.

Two: invest in the people who’ll do qualitatively heavier work. Anyone who’ll deliver strategic judgement needs training in assessment. Anyone who’ll have deeper conversations needs training in questioning and reading customer signals. Anyone who’ll guard quality needs training in taste development. These aren’t typical courses in the standard L&D catalogue. They’re targeted development tracks the director has to facilitate.

Three: set the quality bar explicitly. AI can produce ten things a week instead of two. But ten mediocre things isn’t progress. What’s acceptable for your brand and what isn’t is a board decision. It becomes visible in what you approve and what you send back, in which work goes out and which doesn’t. A board that doesn’t guard that bar explicitly ends up with mediocre output at scale over time.

What this means for team building

A commercial organisation under AI needs three to five years to be fully rebuilt. Not all at once. But the direction is clear.

Fewer people in total. Smaller team. Different distribution across roles. More seniority per discipline. Higher quality demands on every individual contribution. No room for weak links, because they become visible immediately once AI handles the average execution.

For some directors, that’s an uncomfortable prospect. It means difficult conversations with people who’ve done good work for years in roles where the content is now shifting. It means investments in development that don’t all pay off. It means the team of today won’t naturally become the team of the day after tomorrow.

But the alternative is larger. Competitors who do make this transition get three to five years’ lead with a team that can do the work AI can’t. Competitors who don’t stay stuck in a team composition that no longer fits where the market is going. The difference shows up in commercial effectiveness, not productivity. And it’s a difference that, after some years, becomes nearly impossible to close.

---

Further reading

Want to read more about how AI can fundamentally redesign commercial work, with three concrete examples from marketing, sales, and customer success? Our pillar piece works it out: AI as the engine under commercial strategy: how a team of five does what twelve used to.