AI Will Make Services Firms More Productive. Not Automatically More Profitable.

AI is arriving at exactly the wrong time for many IT services and professional services firms.

Not because the technology is bad. Quite the opposite. AI will make parts of services delivery faster, cheaper, and more scalable. It will help teams write code, analyse logs, prepare documents, generate test cases, summarise meetings, draft proposals, review contracts, map compliance requirements, and automate repetitive delivery work.

The problem is commercial.

Many services firms are not currently enjoying easy growth. Buyers are cautious. Budgets are tighter. Procurement is sharper. Clients are asking what AI means for cost, speed, and pricing. Delivery teams are experimenting faster than leadership can redesign the commercial model around them.

The result is a dangerous gap.

AI may improve productivity inside the firm before the firm knows how to capture that productivity economically.

That is the central issue. AI will make services firms more productive. It will not automatically make them more profitable.

Profitability will depend on whether the firm can convert productivity into better pricing, stronger margin discipline, clearer delivery models, and a business model that still works when hours are no longer the cleanest measure of value.

For many firms, this is not a technology problem. It is a business model problem.

The Weakness Of Lazy T&M

Time and materials has always been a convenient commercial model for services firms.

It is simple to explain. The firm sells hours or days. The client pays for effort. Delivery risk is shared or passed through. If the scope changes, more time is billed. If the work takes longer, revenue often increases.

That model works reasonably well when effort is the best proxy for value.

AI weakens that assumption.

If AI allows a consultant, engineer, analyst, or delivery team to complete work in half the time, the client will eventually ask a simple question:

Why am I still paying for the same number of days?

Some firms will try to defend the old model. They will argue that experience still matters, quality still matters, and AI is only a tool. All true. But that will not fully protect a pricing model built around visible human effort.

Clients will not expect every saving to flow to them. But they will expect the pricing conversation to change.

This is where lazy T&M becomes exposed.

By lazy T&M, I do not mean all time-based work is wrong. There will still be advisory, discovery, support, incident response, and complex delivery work where T&M is appropriate. The issue is T&M used as a substitute for commercial thinking.

If the firm cannot explain the value of the work beyond the effort consumed, AI compresses the foundation of the price.

The firm faces a choice:

• keep selling hours and risk giving away productivity,
• move toward fixed price and take on more delivery risk,
• move toward fixed fee or retainer models and manage capacity tightly,
• or redesign its service mix around value, repeatability, and margin.

That choice cannot be made in a generic strategy deck. It has to be made in the operating model of the firm.

Fixed Price Looks Better. It Also Becomes More Dangerous.

AI will make fixed price and fixed fee models more attractive.

If a firm can use AI to reduce delivery effort while holding the commercial price, margin can improve. A fixed price discovery, migration, audit, implementation, or assessment becomes more profitable if the firm can deliver the same outcome with fewer human hours.

That is the upside.

The risk is that fixed price shifts more delivery risk onto the firm.

If the estimate is wrong, the firm owns the overrun. If scope is loose, the firm absorbs the ambiguity. If senior people need to rescue the work, the blended cost rises. If the client treats the fixed price as unlimited access, the margin collapses.

AI does not remove these risks. In some cases, it increases them.

Why? Because the firm may become overconfident.

A team that believes AI will make delivery faster may price aggressively before it understands which parts of the work are actually compressible. It may assume automation can handle work that still requires senior judgment. It may underestimate review, quality control, client communication, governance, security, or exception handling.

The project is sold on the promise of AI leverage, but delivered with human rescue.

That is not AI productivity. That is margin transfer.

Fixed price and fixed fee models require stronger commercial discipline than T&M, not less. The firm needs to model:

• role mix,
• senior review effort,
• phase structure,
• assumptions,
• delivery risk,
• scope boundaries,
• cost rates,
• target contribution margin,
• and what happens if the work does not compress as expected.

In the AI era, the question is not simply: "Can we deliver this faster?"

It is:

Can we price, staff, govern, and control this work so the productivity gain becomes margin rather than leakage?

The Leverage Model Will Change

Professional services firms have always depended on leverage.

Senior people sell, shape, review, and handle complexity. Mid-level people manage and deliver. Junior people perform more routine work, build capability, and create capacity. The economics of the firm depend on getting that mix right.

AI changes the mix.

Some work traditionally done by junior staff may be automated or accelerated. First drafts, basic analysis, code scaffolding, test generation, documentation, research, data clean-up, and administrative preparation may take fewer hours.

That sounds like a margin opportunity.

But it also creates a question: if AI compresses junior work, what happens to the pyramid?

A firm cannot simply remove junior capacity without consequences. Juniors are not only delivery capacity; they are the future senior team. They learn by doing. If AI removes too much of the routine work without a deliberate talent model, the firm may improve short-term efficiency while weakening its future capability pipeline.

At the same time, the firm may need more senior judgment, not less.

AI output needs review. Clients need advice. Risk needs interpretation. Delivery still needs context. Security, architecture, commercial judgment, stakeholder management, and accountability do not disappear because a model can draft an answer.

So the future leverage model may not be "fewer people." It may be a different shape of people.

Owners will need to reconsider:

• which tasks should be automated,
• which tasks should remain junior development work,
• where senior review is required,
• how grades and roles should evolve,
• how utilisation should be measured,
• how much bench capacity is strategic,
• and how pricing should reflect AI-enabled productivity.

This is a business model question, not a tooling question.

A high-end consulting firm may use AI to increase insight speed while preserving premium pricing. A managed services firm may use AI to reduce repetitive support cost and improve fixed-fee margins. A project delivery firm may use AI to compress implementation effort but needs tighter scope control. A growth-stage hybrid firm may need to decide which parts of the business deserve investment and which are becoming structurally weaker.

There is no single AI services model. There are only business model choices.

Productivity Gains Can Become Discounts

The most important commercial question in the AI era is simple:

Who captures the productivity gain?

If a task used to take ten days and now takes six, does the firm keep the price and improve margin? Does the client demand a lower price? Does sales discount pre-emptively to win the work? Does delivery absorb extra scope because "AI made it easier"? Does the firm reduce effort but fail to adjust the contract model?

Without pricing discipline, AI productivity becomes a client discount.

This can happen explicitly or quietly.

Explicitly, the client asks for a lower price because the work should now be faster.

Quietly, the firm gives away more. The team includes extra analysis. The project absorbs more iterations. The client expects faster turnaround for the same fee. Sales uses AI productivity as a reason to cut price. Delivery uses AI to keep up with scope that should have been charged.

In all cases, the firm works faster but does not earn more.

This is why pricing discipline becomes more important, not less.

The firm needs to know where AI creates value, where it reduces cost, where it changes risk, and where the benefit should be shared with the client versus retained as margin.

Some productivity gains should absolutely benefit the client. Better speed, better quality, and better economics can be part of a stronger value proposition. But if every efficiency gain is handed away through lower prices, the firm has funded AI transformation for the client without improving its own business.

That is not strategy. That is leakage.

AI Can Make Spreadsheet Sprawl Worse

Many services firms already run on disconnected spreadsheets.

There is a pricing spreadsheet, a resourcing spreadsheet, a project margin spreadsheet, a utilisation spreadsheet, a finance forecast, a partner view, and a delivery view. Each one may be useful. Each one may be maintained by someone competent. But together they create fragmented truth.

AI may make this worse.

It will become easier to create and maintain local models. A practice lead can ask AI to build a pricing sheet. A project manager can generate a delivery tracker. Sales can create margin scenarios. Finance can create a forecast model. Each sheet becomes smarter. Each sheet becomes easier to update. Each sheet becomes more convincing.

But if the models are disconnected, the firm does not have better control. It has faster fragmentation.

This is the danger of isolated AI-maintained sheets.

They may reduce administrative effort locally while increasing reconciliation effort centrally. They may give each leader a more polished version of their own truth. They may make it easier to defend assumptions that are not connected to the firm's real cost, capacity, and commercial model.

AI does not solve the source-of-truth problem. It can accelerate it.

The firm still needs one operating model where commercial assumptions, delivery reality, people economics, and forward financial consequences connect.

Otherwise, the leadership team will spend the AI era arguing over which automated spreadsheet is right.

The Operating Layer Matters More Than The AI Tool

Most firms will adopt AI tools. That is not the differentiator.

The differentiator is whether the firm redesigns its operating layer around the new economics.

If AI changes delivery effort, that should affect deal modelling. If deal modelling changes, that should affect target margin. If target margin changes, that should affect pricing and contract structure. If delivery effort changes, that should affect roles, grades, utilisation, and capacity. If scope expands, that should affect extensions and change requests. If modelled pipeline changes, leadership should see the effect on future profit before the close.

These things cannot sit in separate files.

The operating layer needs to answer:

• What work have we already contracted?
• What extensions or change requests could change the economics?
• What pipeline has a defined deal model?
• What role mix and cost basis are assumed?
• Where is AI expected to reduce effort?
• Where does senior judgment still dominate?
• Which future reality is forming?
• What lever can still be changed?

That is the level at which AI becomes a profitability issue.

Not because the firm has adopted AI, but because the firm can see whether AI is improving the economics of the work.

The Profitdrive Connection: From AI Productivity To Forward Profit

This is where Profitdrive fits.

Profitdrive is not an AI tool for generating another spreadsheet. It is the operating layer that helps services firms connect business model economics, deal shape, delivery reality, people cost, and forward profit.

The first principle is one source of truth.

A deal model should not live in Excel beside the opportunity. Project economics should not be rebuilt later by finance. People cost should not be manually interpreted each time a grade, role, or contractor mix changes. The operational fact should be entered once, where the work happens, and then flow through the financial model.

The second principle is shift financial insight left.

When a role is added to a deal, margin should move. When the commercial mode changes from T&M to fixed price or fixed fee, the risk should be visible. When staffing changes, project economics should move. When an extension or change request appears, the forward view should change. Financial insight should appear when the operating decision is made, not after the month closes.

The third principle is preserve commercial intent.

If the deal was sold with a target margin, role mix, phase structure, and delivery assumption, that intent should survive into delivery. The project should be judged against the economics it was won on. AI makes this more important because assumptions about effort compression need to be tested against delivery reality.

The fourth principle is separate future realities.

Leaders should not look at one blended forecast. They need to see contracted work, extensions and change requests, and modelled pipeline as separate lenses. In an uncertain AI market, this matters. It lets leadership distinguish what is already committed, what could expand, and what future work would contribute if converted.

This is how AI productivity becomes financial control.

Not through another dashboard. Not through isolated AI-maintained spreadsheets. Through an operating model where the economic consequence of work is visible as the work changes.

The Strategic Question For Owners

AI will change services delivery. That is already happening.

The open question is whether it will improve services profitability.

For some firms, AI will increase margin. They will redesign pricing, tighten scope, adjust leverage, protect commercial intent, and capture productivity gains deliberately.

For others, AI will create more pressure. Clients will demand faster work for lower fees. Sales will discount. Delivery will absorb more scope. Senior people will still rescue projects. Spreadsheets will multiply. The firm will feel more productive but not more profitable.

The difference will not be technology adoption alone.

It will be commercial discipline.

Owners should be asking:

• Which parts of our work will AI compress?
• Which parts still require senior judgment?
• Are we pricing effort, outcome, access, or value?
• Where should we move from T&M to fixed price or fixed fee?
• Which productivity gains should we keep, and which should clients share?
• Do our deal models reflect current cost and AI-enabled delivery assumptions?
• Can we see the financial effect of delivery changes before month-end?
• Are we building one operating model, or just smarter disconnected spreadsheets?

AI changes the work.

Profit will come from redesigning the economics.