Your Buyers Want AI-Era Pricing. You Still Have Pre-AI Costs.

This is the real squeeze hitting mid-sized IT services firms.

If you are running a $50M to $100M services business, the threat is not just that large companies are bidding aggressively.

The bigger threat is that your customers are changing their expectations faster than your delivery model is changing.

They want the same or better outcomes.

They want them faster.

And they are increasingly unwilling to pay old-world prices for them.

That gap is where margins will disappear.

---

What is happening today

Many mid-sized IT services firms are already seeing $10 to $15/hour bids from WITCH companies.

This is not because those companies have suddenly become AI-native delivery machines.

They have not.

This is scale.

Large firms can play this game because they have:

• Massive bench strength
• Aggressive utilization management
• Portfolio-level margin optimization
• The ability to underprice strategic deals
• Large account relationships that allow them to recover margin over time

A mid-sized firm cannot compete with that operating model.

If you try to match them purely on price, you are not being competitive.

You are slowly transferring your margin to the customer.

---

But scale is only the first pressure

There is a second force that is more dangerous.

AI has changed buyer psychology.

Even when customers do not fully understand what AI can or cannot do, they have started believing that software delivery should now be cheaper and faster.

The old conversation was:

Can you deliver this in 12 weeks for $30K?

The new conversation is becoming:

Why does this still cost $30K? Why does this still take 12 weeks? Aren’t your teams using AI? Can this be done with fewer people and a shorter timeline?

That is the real shift.

Customers are compressing budgets and timelines at the same time.

They are not just asking vendors to reduce cost.

They are asking vendors to reduce cost while delivering faster.

Most services firms are not built for that.

---

The old delivery model cannot absorb this

Traditional IT services delivery is built around a simple equation:

More work = more people = more hours = more revenue

That model worked when customers accepted that software delivery was people-heavy, coordination-heavy, and slow.

But now the market is moving in the opposite direction:

Less budget. Less time. Same or better outcome.

That breaks the old model.

If your delivery still depends on large teams, manual coordination, slow QA cycles, repeated handoffs, weak reuse, and senior engineers spending too much time on execution, your cost structure will not fall fast enough.

So when customers push the project from $30K to $20K and expect it in 6 weeks instead of 12, you are stuck.

You either:

• Discount and damage margins
• Overpromise and miss timelines
• Push teams harder and reduce quality
• Move to fixed price and absorb delivery risk
• Lose the deal

None of these is a strategy.

They are symptoms of a delivery model under pressure.

---

Fixed price is not a solution by itself

A lot of firms will respond by moving from T&M to fixed-price projects.

That sounds like adaptation.

It is not enough.

Fixed price only works when your delivery system can compress cost and cycle time faster than the customer compresses budget and timelines.

Otherwise, fixed price simply converts customer price pressure into vendor margin pressure.

You are taking on pricing risk without changing execution capability.

That is how margins collapse.

The issue is not T&M versus fixed price.

The issue is whether your delivery engine can produce outcomes at the new market expectation.

---

Where AI actually matters

AI is not the full cause of this disruption.

AI is first changing customer expectations.

Then it becomes the tool vendors need in order to survive those expectations.

The goal is not to replace engineers.

That is the wrong frame.

The goal is to stop using good engineers as expensive task executors.

Engineers still need to make the important decisions:

• Architecture
• Security
• Data design
• Integration strategy
• Tradeoffs
• System boundaries
• Review gates
• Production readiness
• Failure handling

But repeatable execution should become cheaper and faster.

That means using AI-assisted workflows for things like:

• Boilerplate implementation
• Test generation
• Documentation
• QA automation
• API scaffolding
• Migration scripts
• Codebase understanding
• Repetitive UI work
• Support triage
• Internal delivery tooling

The winning model is not engineer versus AI.

It is engineer-as-architect, with AI-assisted systems handling repeatable execution under strong technical guardrails.

---

This is a delivery architecture problem

Most firms are treating this like a pricing problem.

It is not.

It is a delivery architecture problem.

The firms that survive this shift will redesign how work gets delivered.

They will move from:

• People-heavy delivery to system-assisted delivery
• Manual estimation to repeatable delivery playbooks
• One-off execution to reusable assets
• Late QA to continuous review loops
• Large teams to smaller senior-led pods
• Billing effort to pricing outcomes where possible

This is what AI-native services should actually mean.

Not copilots.

Not chatbots.

Not demos.

A real change in how delivery cost and cycle time are controlled.

---

The real question for CXOs

The question is not:

Are your engineers using AI tools?

That is too shallow.

The real question is:

Can your delivery cost and delivery timeline fall as fast as your customer’s budget and timeline expectations are falling?

If the answer is no, your margin problem is only beginning.

You are currently being squeezed by large firms using scale.

You are being repriced by customers because of AI expectations.

And you will soon compete with firms that have rebuilt delivery around AI-assisted execution.

That is the three-sided squeeze.

---

What needs to change now

Mid-sized services firms need to redesign their operating model before the market forces them to do it under margin pressure.

That means looking hard at:

• How projects are estimated
• Where senior engineering time is being wasted
• Which delivery tasks can be systematized
• Where QA and rework are leaking margin
• What assets can be reused across clients
• Which workflows can be AI-assisted safely
• Which offerings can move away from pure effort-based pricing

This is not about adding AI on top of the old model.

It is about rebuilding the model so you can deliver faster, at lower internal cost, without lowering engineering quality.

---

If this is already showing up in your pipeline

If you are a CXO at a mid-sized IT services firm and you are seeing any of these signs:

• Buyers pushing down budgets
• Buyers asking for faster delivery
• Lower win rates against large vendors
• Fixed-price margins getting worse
• Senior engineers stuck in execution instead of architecture
• Delivery teams using AI tools, but without a real operating model around them

Then the problem is not just sales.

It is not just positioning.

It is your delivery economics.

I help IT services firms redesign their delivery model for this new market: lower customer budgets, faster outcome expectations, and AI-assisted execution.

If this is happening in your business, get in touch with me.

I’ll share the diagnostic framework I use to identify where margin and cycle time are leaking across estimation, engineering workflow, QA, reuse, and pricing.