Every time I look at the way we organise our businesses, especially now with agentics coming in, I think: are we really still using org charts as our main tool?
You know the kind: CEO at the top, rows of senior leaders, managers, more managers and finally the people at the bottom who actually do the roles, day in and day out.
That structure worked for a long time because humans needed hierarchy.
Managers could only properly manage a handful of people, five, eight, ten, maybe twenty, because it was all about one to one interaction.
You had to sit down with someone, have a proper chat, and keep up with everyone’s work.
We needed clear lines, clear roles, and a way to know who reported to who, but it is time to look at whether that is still fit for what is now possible, especially by late 2025.
Let us call this out clearly, our organisation charts were designed around human limits, not business outcomes.
We all know that politics often plays just as much a role in who sits where as what the work actually is.
The manager could handle only so many direct reports before it broke down, so to keep things humming you built a pyramid of control.
Everyone knew their place, supposedly, and what their job was, at least on paper.
Work happened through layers, with plenty of time needed to form teams, brief them up, align them, and fix misunderstandings when they happened.
That was the price of doing business in a world where every handoff, every change of course, needed a person to mediate it.
Strategists are now pointing out that as AI becomes native to the organisation, the fundamentals of that pyramid are shifting.
As one major analysis put it,
"The agentic organization will be built around five pillars of the enterprise: business model; operating model; governance; workforce, people, and culture; and technology stack."
That is a very different starting point to a simple tree of names and job titles.
Here is where things are starting to creak for leaders.
You look at the classic pyramid and it still looks clean.
But if you follow an actual piece of work, an onboarding journey, a complex customer issue, a cross border project, it bounces all over the place.
It jumps functions, tools, regions and time zones.
Every jump adds delay, confusion and meetings.
At the same time, AI and agentic systems are slipping into the gaps.
They summarise, route, draft, monitor and even coordinate.
One business journalist captured it bluntly:
"AI is quietly changing the traditional corporate hierarchy, flattening structures and reshaping job roles from the bottom up."
That quiet shift is exactly what makes the old chart feel more like a museum piece and less like a control panel.
The more your work is digital, the more your real organisation looks like a network of flows, not a tidy stack of boxes.
When you still manage purely through the stack, you get drag.
People wait for approvals, context is lost between handoffs, and the best cross functional talent is trapped in neat rectangles on a slide.
Here is the good news and the discomfort.
As agentic workers, Agentics for short, sometimes I call them gentics when I am in a hurry, become more common, the old style hierarchy does not simply get faster, it starts to break.
Agentics do not care about diagrams of power, or who has the most impressive title.
They operate through the work that actually needs to be done.
If you are trying to get a job done across four systems, multiple teams, and different knowledge bases, agentics form connections directly where they are needed, not by waiting for permission to flow down the chart.
Shared knowledge is surfaced between people and digital workers almost instantly.
If a problem shows up, a team can fire up and solve it in minutes, then disperse.
Compare that to days or weeks of coordination for humans.
There is no longer a need to wait for the right line manager to be available, or for someone to add a meeting to the calendar.
The work almost self organises.
Consultants are already talking about intelligent virtual assistants that roam across systems, stitching together tasks and even assembling the right mix of capabilities for a specific outcome.
Put simply, the operating system of work is getting more agentic, even if your chart has not caught up yet.
So what structure makes sense next.
I think we are moving away from rigid org charts and towards something I call strategics, teams grouped by outcome, not hierarchy.
You start with the goal.
What are we trying to achieve in this particular area.
Then you ask, which roles do we actually need to get there.
Human roles and gentic roles.
Agentics make this possible because you are not constraining yourself by headcount, reporting lines, or HR process.
Agents can be on the bench for months with no cost, ready to be activated for once a year issues without anyone worrying about their utilisation.
Teams can form fast for a mission and dissolve just as quickly.
It is all about what needs to get done, not how do we keep everyone busy.
One practitioner summed the shift neatly:
"In The Age of AI, Goals Will Define the Org Chart."
That is the essence of strategics.
You treat the outcome as the primary object, and structure everything else around it.
To lead in that model, you spend less energy on span of control and more on clarity of goals, quality of decisions, and how humans and agents complement each other.
Now let us talk about the bit most leadership teams feel straight away, the economics.
Traditionally, every person on payroll is a cost, whether they are being used daily or sitting in reserve for emergencies.
Carrying specialist talent just in case was a luxury only big firms could afford.
You could not justify a forensic pricing expert, a niche regulator, or a top tier data scientist for a problem that might appear twice a year.
Agentics change that logic completely.
A digital agent can sit idle for months and cost you nothing.
Need someone with an ultra rare skill for an odd situation.
No issue.
Spin them up for that moment and bench them again.
There is no friction to scaling up or down, and no penalty for keeping talent just in case.
Consultants working with agentic enterprise platforms are already reporting cases where virtual specialists are dropped into complex workflows in minutes, then retired when the job is done.
For a leader, this shifts the question from what roles can we afford to carry all year to what outcomes are worth engineering once and then calling on whenever needed.
You can literally have everyone you want, for every role that matters, at exactly the moment you need them.
The cost shifts from being about people hours to being about outcomes delivered.
Your rare talent becomes a capability, not a headcount debate.
Here is the thought experiment I suggest:
If money were no object, how would you structure your company.
What would your dream team look like.
What roles would you fill that you have always felt were a luxury, data experts, rare problem solvers, creative visionaries, forensic analysts, behavioural scientists, whoever you quietly wish you had.
Now add the gentics.
Which repeatable roles could be held by digital workers who never sleep, never forget and can be cloned on demand.
What would you delegate to them first, and how would that change the work humans do every day.
And now, ask yourself, what is stopping you from creating that dream team with agentics.
Previously, the constraint was budget and physical span of control.
Now, with digital workers that can come on and off the bench for free or near free, you can start to design the dream structure first and work backwards.
You organise not by who you can afford to stick around, but by what outcomes you care about.
The only limit is your imagination and your willingness to shift your management mindset.
Forward looking leaders are already mapping their most important outcomes and asking a simple question for each.
If we had no organisational baggage, what is the ideal mix of humans, agents and partners to deliver this outcome at world class level.
That is the starting point for your 2027 organisation chart, or more accurately, your 2027 outcome map.
If you lead a team, division or whole company, what should you actually do with this.
Here is a simple, leadership level sequence.
Map work as it is done today
Take one or two core journeys, like quote to cash or claims handling, and sketch how work truly flows, across teams, tools and approvals. Notice how little it resembles your slideware org chart.
Name 3 to 5 strategics that matter most
These are the outcomes that define your success in the next 18 to 24 months, for example, new market entry, churn reduction, regulatory compliance, or a margin target in a critical unit.
Design the dream team for one strategic
Ignore budget for a moment. List every human role and gentic role that would make this outcome almost embarrassingly effective.
Mark what can be agentic by 2027
Use your judgement and your advisors to highlight which tasks could realistically be carried by agents within two years, and which must stay with humans because they need judgement, trust or complex negotiation.
Launch one small, agentic pilot
Do not boil the ocean. Wrap one strategic with a pilot agentic workflow, something like always on monitoring and triage, automated summarising and routing, or a gentic project manager that keeps tasks flowing.
Strengthen your governance in parallel
As one leading analysis stresses, the agentic organisation rests not only on technology, but also on governance, workforce, people and culture. Make sure risk, ethics, data quality and accountability grow with the pilots, not after them.
If you do this deliberately through 2025 and 2026, by 2027 your organisation will already feel more like a network of strategics than a fragile pyramid.
You will be used to seeing teams that form and disperse as needed, with humans spending more time on judgement, relationships and story, and less time chasing status updates.
Let us bring it all together.
Our old organisation charts were built for a world of human limitations, span of control, communication lag, and the steady beat of politicking.
Agentics break those assumptions, making it possible, and increasingly necessary, to design your company around outcomes.
External observers are already documenting how AI is flattening structures and reshaping roles from the bottom up, and how agentic organisations lean on new pillars of model, governance, culture and stack.
At the same time, practitioners are reminding us that in the age of AI, goals will define the org chart.
Your job as a leader is to turn that from an article you read into a design choice you make.
So, start with the question.
If money were no object, what is the team you would build for your most important strategic outcome.
With digital workers and agentic systems, you can make that team real, assembling and disbanding teams in minutes.
By 2025 and beyond, leaders who reimagine structure with agentics first, designing for flexibility, specialisation and outcome, unlock an edge everyone else will have to race to catch up to.
Have a go at the thought experiment with your team.
Sketch your dream outcome map.
Then ask a simple, uncomfortable question.
Why can you not start building it now.
Links:
- The agentic organization: A new operating model for AI, https://www.mckinsey.com/capabilities/people-and-organizational-performance/our-insights/the-agentic-organization-contours-of-the-next-paradigm-for-the-ai-era, Trust rating: high, Reason: Defines the agentic organisation model across business model, operating model, governance, workforce and technology, giving leadership level structure to the shift away from traditional org charts, Date written: 2024-05-06
- AI is already changing the corporate org chart, https://fortune.com/2025/08/07/ai-corporate-org-chart-workplace-agents-flattening/, Trust rating: high, Reason: Shows real world examples of AI flattening hierarchies and reshaping roles, grounding the blog’s argument in current corporate practice, Date written: 2025-08-07
- The new org chart: Unlocking value with AI-native roles in the agentic era, https://www.cio.com/article/4060162/the-new-org-chart-unlocking-value-with-ai-native-roles-in-the-agentic-era.html, Trust rating: high, Reason: Explores AI native roles and new organisational models, supporting the move from static charts to dynamic, outcome based structures, Date written: 2024-04-20
- How Agentic AI is Transforming Enterprise Platforms, https://www.bcg.com/publications/2025/how-agentic-ai-is-transforming-enterprise-platforms, Trust rating: high, Reason: Provides case studies of agentic AI in enterprise platforms, illustrating dynamic team building and measurable value from agentic workflows, Date written: 2025-01-31
- In The Age of AI, Goals Will Define the Org Chart, https://www.workpath.com/en/magazine/in-the-age-of-ai-goals-will-define-the-org-chart, Trust rating: medium, Reason: Practitioner piece explaining how goals and outcomes are starting to shape team structures in AI enabled organisations, Date written: 2024-03-28
Quotes:
- "The agentic organization will be built around five pillars of the enterprise: business model; operating model; governance; workforce, people, and culture; and technology stack.", McKinsey & Company, Source URL: https://www.mckinsey.com/capabilities/people-and-organizational-performance/our-insights/the-agentic-organization-contours-of-the-next-paradigm-for-the-ai-era, Trust rating: high, Reason: Frames the breadth of change required beyond the org chart, Date written: 2024-05-06
- "AI is quietly changing the traditional corporate hierarchy, flattening structures and reshaping job roles from the bottom up.", Lila MacLellan, Source URL: https://fortune.com/2025/08/07/ai-corporate-org-chart-workplace-agents-flattening/, Trust rating: high, Reason: Captures the lived reality of hierarchy change that underpins the blog’s urgency, Date written: 2025-08-07
- "In The Age of AI, Goals Will Define the Org Chart.", Johannes Müller, Source URL: https://www.workpath.com/en/magazine/in-the-age-of-ai-goals-will-define-the-org-chart, Trust rating: medium, Reason: Expresses the core shift to outcome based structures in a single memorable line, Date written: 2024-03-28