AI may shift the advantage back to the strongest businesses in established sectors
A lot of the AI commentary is still stuck in fantasy mode. The more interesting shift may be broader: strong businesses in established sectors could use AI to understand customers better, make better decisions, and close part of the gap with newer challengers.
People talk as though a few shiny disruptors are about to remake every sector from scratch. I doubt it. The underlying capability is being industrialised and distributed at speed. That means the real battle is less likely to be disruptors versus everyone else, and more likely to be which incumbents adapt fastest and which stand still.
For years, newer challengers have had a real edge over established businesses. They were often faster to process information, better at analysing data, and quicker to turn insight into action. That edge was never only about product. A lot of it came from better learning loops, better workflows, and better use of technical talent.
That is where AI could start to change the balance.
Weak businesses will still be weak. AI does not fix confusion, poor leadership, or lack of focus. But strong businesses in established sectors may be underestimating how much the balance of power could swing back in their favour if they use it well. Not by trying to become technology companies, but by becoming better at understanding customers, diagnosing problems, and making decisions.
The old advantage was not just product
A lot of newer challengers won because they were simply better at learning. They were faster to connect fragmented information, faster to see patterns, and faster to act on them. They often had stronger product teams, stronger developers, and more technical talent focused on a narrower problem.
In the past, many of these capabilities required meaningful product, data, or engineering resource. For an established business, that often meant a familiar list of reasons to do nothing. Too expensive. Too risky. Too disruptive. Too hard to validate. Too far from the core.
AI changes that equation. It makes it dramatically cheaper to get to a workable answer, whether the issue is analysis, testing, synthesis or internal tooling. That does not eliminate the advantage of newer challengers. It does not give established businesses the same focus or the same appetite for risk. But it can narrow part of the gap that mattered most in practice: speed of learning, quality of diagnosis, and the ability to act on insight without needing a large technical team behind every improvement.
That matters because many established businesses were never weak on customer relationships, domain knowledge, or economics. They were weaker on visibility and response.
“A lot of what looked like a challenger advantage was really a speed-of-learning advantage. AI can now compress that.”
Better customer understanding is the real prize
The most impressive thing AI can help a strong established business do is understand customers far better.
That is the real prize. If a business understands customers more deeply, a great deal follows. Better retention. Better service. Better pricing. Better upsell. Better prioritisation. Better service design. Fewer blind spots.
The old way was often periodic. Meet the customer. Ask what is working. Ask where they need help. Useful, but limited. The customer tells you some of it. The account team knows some of it. Service sees some of it. Finance sees some of it. Product sees some of it. No one sees enough clearly enough, or early enough.
AI can change that. It can surface shifts in behaviour, support burden, hidden demand, signals of dissatisfaction, and commercial opportunities much earlier and much more specifically. In many established businesses, the customer relationship is already there. The missing piece is not access. It is visibility.
That is where the competitive implication starts to matter. Newer challengers often go after markets where customers have settled into quiet acceptance rather than real satisfaction. The service is good enough. The friction is tolerated. Waste becomes normal. A challenger sees that gap and builds around it. A strong incumbent with better visibility can close that gap itself.
The point is not to impose a technology thesis on a business. It is to help a good business become better informed and more responsive.
“In many established businesses, the customer relationship is already there. What is missing is not access. It is proper visibility.”
Better diagnosis changes the quality of decisions
This connects to something I keep seeing in boardrooms. The problem is rarely a lack of ideas. It is that no one has really done the diagnosis properly.
While advising a medical devices business, I saw this clearly. The threat was obvious. Lower-cost competitors from China were putting pressure on part of the market. Everyone in the room knew there was a problem, but no one really understood the shape of it. The discussion dragged into the familiar pattern: instinct, old experience, and the weight carried by the strongest voice in the room. A senior executive presented a segment decline and then a list of possible actions. They all sounded plausible. Focus on higher-spec customers. Defend the premium end. Patch the service. Redesign the offer. But the diagnosis underneath it was still weak.
“The problem in most boardrooms is not that people do not have ideas. It is that no one has done the diagnosis properly.”
The business chose to put more emphasis on customers with higher requirements and lower price sensitivity. That was defensible. It may not have been the best answer.
A better process would have broken the issue down properly. Which customers were truly price-sensitive. Which parts of the offer they actually valued. Which service layers were expensive habits rather than real differentiators. Which accounts were recoverable. Where cost-to-serve could be reduced without damaging retention. AI could have surfaced those patterns earlier and with much more precision.
That would not have guaranteed a better answer. But it would have improved the quality of the discussion. The room would have been debating a much better set of facts and a much better set of options.
We see the same principle in transactions.
“At Zenith Capital, we already use AI extensively in fund management. It is materially improving how we map markets, assess opportunities, enrich data, and ask better questions earlier.”
It helps connect financial statements, notes, and narrative more effectively. It surfaces patterns, inconsistencies, risks, and opportunities earlier. The value is not only speed. It is better judgement before important decisions are made.
That matters to buyers. It matters to owners too. Fewer unpleasant surprises after signing is good for everyone.
AI can improve the whole organisation, not just management
One of the more interesting effects of AI is organisational, not just technical.
Traditional hierarchies often work because judgement is scarce. The further up the ladder you go, the more experience sits there, and the better the judgement tends to be. That is normal. But it also means too many proposals and too many decisions depend on a small number of people having the time and context to think clearly.
AI can help distribute more of that context lower down the organisation.
Not entirely. Humans still matter for validation and decisions. But teams can become better informed, make better proposals earlier, and escalate with more substance. Account teams can see where attention is needed. Service teams can understand customer risk better. Commercial teams can identify where packaging or pricing no longer fits. Finance visibility can become more usable outside the finance function. Managers can spend more time on prioritisation, coaching, and the bigger picture, rather than simply being the only people with enough visibility to interpret what is going on.
That is a meaningful shift. It makes the organisation sharper without necessarily making it bigger. It also improves service without adding overhead.
That should matter to owners. This is not about turning a good business into something else. It is about removing friction from what already makes it valuable.
I am conscious this can start to sound like another sweeping AI article if it is left too abstract, so it is worth being specific about the limit.
AI strengthens strength, it does not create it
This is where restraint matters.
AI does not rescue a weak business. It makes a strong one better. It can improve customer understanding, sharpen decisions, and remove friction from high-value work. It cannot substitute for clear leadership, sound economics, or a business customers actually value.
That is why the right response is not theatre. Keep what works. Do not turn this into theatre.
Start with one real bottleneck. Connect and clean the minimum data needed. Test one improvement with one team and one measurable outcome. Build buy-in through visible wins. Do not confuse motion with progress. Do not start by firing people. AI should help a business grow, not become an excuse for careless cost-cutting.
There are also real limits. AI is less powerful where the data infrastructure is weak, where regulation makes the cost of error too high, or where customer concentration is so high that each decision carries outsized risk. In those environments it can still support judgement, but it should not be mistaken for judgement itself.
The strongest businesses in established sectors are better placed than they think. They already have customers, operating substance, domain knowledge, and real businesses worth improving. If they use AI precisely, they can understand customers better, make better decisions earlier, and improve service more intelligently. That could narrow part of the historic advantage many newer, more focused challengers have had.
“Weak businesses will add AI to the confusion. Strong businesses will use it to widen the gap.”
