AI is raising the minimum standard. What happens to your business model when better, faster and cheaper become expected?

AI gives businesses four fundamental opportunities: to become more innovative, to produce better work, to operate faster and to deliver services more cheaply.

Businesses can pursue one, several or all four of these opportunities. The combinations will be different depending on the sector, the work, the customer and the risks involved.

For some, AI will primarily reduce costs and increase operational efficiency. For others, it will improve quality or allow much more work to be completed in the same time. And for the more ambitious, it will make entirely new services, products and ways of solving problems commercially possible.

But there is an uncomfortable economic question behind all this progress.

If everyone becomes more productive, who ultimately gets to keep the value that AI creates?

Will businesses become more profitable as their productivity improves? Will clients expect lower prices for the same work? Will higher output become the new minimum standard across entire industries? And what happens to businesses whose revenues have historically depended on the time required to do the work?

The early research suggests these are no longer hypothetical questions.

1. Productivity improvements are real, although the benefits are far from uniform

A study involving 758 Boston Consulting Group consultants found that participants using GPT-4 completed 12.2% more tasks and worked 25.1% faster on average across tasks within AI's capabilities. The research also found improvements in the quality of the work produced.

However, the same research identified what it called a jagged technological frontier: AI enhanced performance on some tasks but reduced it on others, including tasks that appeared superficially similar.

Another study, published in The Quarterly Journal of Economics in 2025, examined 5,172 customer-support agents. AI assistance increased the number of issues resolved per hour by an average of 15%, with particularly significant benefits for less experienced workers.

These findings are significant, although they should not be treated as universal productivity forecasts applicable to every business.

They demonstrate that AI can change the relationship between labour, time and output. They also demonstrate why more output does not automatically mean a better result.

In professional services, particularly where mistakes carry legal, financial or reputational consequences, judgement, checking and accountability remain essential.

Nevertheless, the direction of travel is becoming clearer as the technology develops and adoption increases.

Clients are likely to expect substantially more output, quality or value from the same amount of professional time.

Those changing expectations have significant consequences for how businesses price their services.

2. The hourly rate is facing a fundamental productivity problem

For decades, many professional-services businesses have sold their expertise primarily through the time required to deliver it.

Lawyers, consultants, accountants, agencies and other specialists have built substantial businesses around hourly rates, chargeable time and utilisation targets.

But what happens when AI changes how much work can reasonably be completed in an hour?

Suppose a professional charges £200 per hour and historically requires ten hours to complete a project.

Under that traditional pricing arrangement, the client would pay £2,000 for the completed work.

Now suppose AI enables the professional to complete the same work, to the same standard, in four hours.

Should the client pay £800 for the four hours actually worked? Should the hourly rate rise to £500 to preserve the previous fee? Or should the professional still charge £2,000 because the client is buying the outcome, not the time?

There is no single correct answer to these questions because the commercial circumstances will differ. But the underlying economics of delivering the work have fundamentally changed.

Simply raising hourly rates to compensate for fewer hours may be difficult to justify where the underlying work has become easier to deliver.

On the other hand, clients may reasonably pay more per hour for people whose expertise, judgement and effective use of technology generate better outcomes.

This points towards a more nuanced future for professional-services pricing models.

Hourly rates may survive, but the amount and quality of work an hour is expected to deliver will change.

We can already see evidence that professional-services pricing practices are evolving.

Clio's 2025 Legal Trends research reports that 59% of surveyed law firms billed using flat fees, either exclusively or alongside hourly fees, in 2024. Among firms using AI more extensively, 45% reported making pricing adjustments in response to AI.

That does not mean hourly billing is disappearing from professional services. In complex, uncertain or contentious matters, the scope of work may be difficult to predict and time-based arrangements may remain appropriate.

But charging for the time a task used to take is unlikely to be a sustainable strategy.

3. Fixed-fee businesses could benefit significantly, at least during the initial stages of adoption

For firms charging fixed or project-based fees, AI creates a potentially substantial opportunity to improve profitability.

If a business can deliver the same £10,000 project in half the time, without sacrificing quality, its economics may improve dramatically.

Consider a simplified example involving a fixed-fee professional-services project.

These hypothetical figures illustrate the potential economics rather than observed industry results. Delivery costs include assumed relevant project costs, and contribution is not equivalent to net profit.

At first, this appears to be an extremely attractive commercial proposition for the business delivering the work.

But there is an important complication that businesses cannot afford to ignore.

Their competitors will also have access to increasingly capable AI technology.

Over time, some of the productivity benefits are likely to be competed away through lower prices, greater scope, faster turnaround or higher quality.

Imagine the market price of that £10,000 project eventually falls to £6,000, while delivery costs remain at £4,000.

The provider now makes £2,000 per project rather than the original £3,000.

Its percentage contribution margin has improved from 30% to approximately 33%.

But its absolute contribution has fallen by a third compared with the original business model.

A business can become more efficient, improve its margin percentage and still make less money from the work it delivers.

This distinction between percentage margins and absolute profitability is critically important.

Not every business will be able to retain the same revenue or absolute profit per engagement. Even percentage margins may come under pressure as expectations rise, technology costs increase and competition intensifies.

Higher margins are not guaranteed simply because work becomes cheaper to produce using AI.

4. The productivity gains will not all remain with the businesses providing services

Thomson Reuters' 2025 Future of Professionals report estimated that legal professionals could free up almost 240 hours annually through AI, corresponding to an estimated $19,000 in annual value per professional.

Those figures were projections based on survey research, rather than audited savings. But they illustrate the potential scale of the economic opportunity associated with increasing professional productivity.

The more interesting question is what happens to that value as the market adjusts.

Some could become additional profit for businesses that adopt AI successfully. Some could fund better services, research, training, risk management or innovation. Some could be passed to clients through lower fees and faster turnaround.

And some could simply disappear into a rising competitive standard that everyone is expected to meet.

We have seen this pattern with the adoption of other technologies across different industries.

Capabilities that initially distinguish leading businesses can eventually become normal expectations shared across an entire market.

AI may accelerate that process, particularly where the tools are accessible and the work is relatively standardised.

Today's competitive advantage can quickly become tomorrow's minimum acceptable standard of quality, innovation and efficiency.

The strategic challenge is not merely to capture the first round of savings from adopting AI.

It is to create an advantage that survives once those savings are widely available to competitors.

5. Client expectations are already shifting towards higher standards of delivery

There is emerging evidence that clients are no longer treating AI adoption as an optional improvement.

The 2026 Thomson Reuters Future of Professionals report, drawing on responses from 1,816 professionals across 62 countries, found that 78% of corporate clients considered AI-enabled quality improvements very important or essential.

Yet only 6% believed most or all of their professional-services providers were delivering them.

Perhaps more significantly, 32% reported having reconsidered, or planning to reconsider, relationships with providers they believed were falling behind.

That represents a substantial gap between what clients increasingly expect and what they believe their advisers deliver.

It also raises an important distinction between the price of a service and the value received.

Clients are not necessarily demanding that everything becomes cheaper simply because AI is being used. They may want better advice, more thorough analysis, faster responses, improved risk management or more commercially useful outcomes.

The opportunity, therefore, is not exclusively about cutting the cost of existing work.

It is about increasing the value delivered for the money being spent by the client.

The new commercial benchmark may be less about how many hours you worked and more about what you achieved with the resources available.

6. Businesses will have different ways to win, and not all of them involve protecting existing margins

Businesses will need to make deliberate choices about where and how they intend to compete.

Winning through efficiency means producing more at a lower unit cost while maintaining appropriate quality standards. This may work particularly well for repeatable, high-volume services, although competitive pricing pressure could be intense.

Winning through scale means serving more clients or delivering more projects without a proportional increase in people or costs. Lower profit per engagement may be offset by greater volume, provided demand increases sufficiently and the business can manage that growth.

Winning through quality and trust means using AI to improve analysis, consistency and responsiveness while retaining specialist expertise and human accountability. This may be particularly relevant in environments where the consequences of mistakes are significant and clients value experience, reliability and sound judgement.

Winning through innovation means creating new products, services and solutions that were previously too expensive, slow or complex to deliver. This may offer greater differentiation than efficiency alone because the business is creating additional value rather than simply reducing the cost of existing services.

None of these approaches is automatically superior to the others. Businesses may combine them, and the right choice will depend on their market, capabilities and commercial objectives.

However, innovation deserves particular attention as a source of potentially lasting competitive advantage.

If every competitor can eventually produce the same report in one hour rather than five, speed may cease to differentiate anyone.

But a business that uses the additional capability to solve a problem its competitors cannot solve may create an entirely different source of value.

That is a much more interesting opportunity than simply automating existing tasks.

7. Complexity might become one of the most important opportunities for meaningful innovation

AI is exceptionally useful for handling information, generating possibilities and helping people explore complicated questions.

But processing complexity is not the same as understanding it, particularly when important decisions carry real consequences.

A problem involving intellectual property, contractual obligations, confidential information, AI systems, employees, regulatory exposure and commercial relationships cannot necessarily be solved by treating each issue separately.

A technically correct answer to one question may still lead to a poor overall commercial decision.

There may be no established process to follow in an unfamiliar or rapidly developing situation. Relevant facts may be incomplete, contested or changing as new information emerges. There may be several viable options, each carrying different legal, financial, practical or reputational consequences.

These are precisely the environments in which combining AI with human expertise could become particularly valuable.

AI can help people examine more information, test more scenarios, identify patterns and explore options more quickly.

Humans bring context, experience, ethical judgement, commercial understanding, accountability and the ability to challenge the questions being asked.

And there is research supporting the need for care when using AI for complex professional work.

The BCG consultant experiment found that AI could worsen performance when tasks fell outside its effective capabilities. Meanwhile, the 2026 Thomson Reuters report found that 41% of professionals lacked access to AI tools designed for the standards of professional work.

For New, this is a particularly exciting direction in the development of our services and approach.

We are interested in situations where new technology can work alongside human expertise to unravel difficult legal, commercial, technological and people problems.

The opportunity is not simply to complete familiar professional work more quickly than before.

It is to discover new ways to understand problems, assess risk and develop solutions that would previously have been difficult, impractical or uneconomical.

New solutions to new and old problems, made possible through the combination of human judgement and emerging technology.

That is where the innovation opportunity becomes far bigger than efficiency improvements alone.

8. Businesses need to rethink their economics, not simply introduce AI into existing processes

I think the starting point is to reconsider the economics of the entire business, rather than simply asking staff to use AI more.

Businesses need to understand their actual cost of delivering work, including technology, human review, quality assurance and the work that still cannot sensibly be automated.

They need to think about how much of their productivity gain they can retain, how much may eventually pass to customers, and how their competitors will respond.

For businesses selling time, the question is what an hour of expertise should now achieve.

For fixed-fee businesses, it is how to improve project economics without assuming current fees will last indefinitely.

For businesses operating in highly competitive markets, it may be whether greater volume can compensate for lower revenue or profit per transaction.

And for businesses operating in complex, higher-value markets, the greatest opportunity may be to develop entirely new services and capabilities.

The commercial model matters just as much as the technology being introduced into the business.

A business that doubles productivity but fails to adapt its pricing, pipeline, proposition or operating model may never realise the financial benefit.

Equally, a business that becomes dramatically more efficient but competes only on price may find itself working harder to maintain the same absolute profit.

The objective should not simply be to become more efficient, but to build a better business with that efficiency.

9. The minimum acceptable standard is increasing, potentially by considerably more than many businesses expect

The evidence doesn't establish exactly how quickly prices will change, which industries will experience the greatest disruption or who will capture the financial gains.

But it does establish that meaningful productivity improvements are already possible, that performance varies by task, and that professional-services clients increasingly expect providers to benefit from AI.

My expectation is that the market will gradually reset around a much higher minimum standard of innovation, quality and efficiency.

Some services will become cheaper as the cost of delivering them falls. Some will remain profitable but generate less absolute revenue for each engagement. Others will command higher prices because they deliver greater value and better outcomes.

Some businesses will protect or improve percentage margins but still need substantially more customers to maintain their overall profits.

And others will use AI to create offerings that barely resemble the services they sell today.

The most interesting businesses may not be those achieving the greatest reduction in delivery time.

They may be those using the additional capacity to tackle previously unsolved problems.

That requires a different question from the one many businesses are currently asking.

Rather than simply asking how AI can help us complete our existing work more efficiently, perhaps we should be asking something much more fundamental.

What would we offer, how would we deliver it, and how would we charge for it if we were designing our business from scratch today?

The technology is remarkable, and the opportunities for innovation extend well beyond straightforward automation.

The economic consequences could be equally significant for businesses, professionals and their clients.

And I suspect the businesses that benefit most will be those prepared to rethink what they do, rather than just how quickly they do it.

How are you applying AI to improve the way your business creates and delivers value?

Are you trying to maintain your existing margins and business model as the market changes? Are you preparing to deliver substantially more value for the same price? Or are you using AI and human expertise to create something genuinely new?

And perhaps the bigger question is how your business intends to compete when today's innovations become tomorrow's minimum standards.

How are you planning to win when today's innovation becomes tomorrow's minimum standard?