What Are You Selling When the Work Gets Faster?

When AI reduces delivery time, clients may challenge pricing that remains tied to hours, headcount, or output volume. The solution is not to move immediately to outcome-based pricing. Rather, make the value of your expertise, methods, evidence, and accountability clear and defensible to prospective buyers, then select a pricing model aligned with what you can control and measure.

Key takeaways

• AI productivity becomes a pricing problem when your offer is communicated in terms of effort.

• Five questions determine which pricing model is most appropriate: What is the outcome? What expertise is required? What level of control do you have? How will performance be measured? And what evidence supports the offer?

• A hybrid model combining a base fee with a modest variable component is a credible next step.

• Pricing cannot be addressed by the finance team alone. The offer, proof and sales narrative must evolve accordingly.
  
  

The renewal conversation is already changing

AI is accelerating research, analysis, content, software delivery, and project coordination. That’s good news for capacity and margin. It’s bad news for revenue if your clients believe they’re mostly paying for labor that just got cheaper.

This trend is most evident among large IT services companies. Forrester’s review of Q2 2026 earnings found that provider revenue increased only slightly, bookings remained largely flat, and revenue per employee barely moved, as companies reprice contracts to deliver more work with fewer employees. Infosys lowered its full-year growth outlook this year, and analysts noted that clients expect AI-related savings to be reflected at contract renewal.

Clients are also setting higher expectations. In Thomson Reuters’ 2026 Future of Professionals research, 78% of corporate buyers reported that AI-driven quality improvements from their providers are very important or essential. Only 6% indicated that most of their providers deliver them.

The question is not whether AI should accelerate delivery. It should. The question is whether buyers can identify what remains worth paying for once it does.

A quick example

Note Illustrative scenario. Replace with an approved client story when one is available.
A 60-person data consultancy sells a 12-week warehouse migration for $180,000, scoped at approximately 1,200 hours. With AI-assisted code conversion and testing, the team now delivers the project in roughly 750 hours.

If the proposal continues to list hours and a blended rate, the client is likely to request a $70,000 discount and would have a reasonable basis for doing so. If, instead, the proposal focuses on the outcomes the migration protects (no downtime during cutover, a validated data model that the client’s team can maintain, and a firm go-live date), the price remains justified. The company retains most of the efficiency gains as margin and frees capacity for its next project.

Same work. Same team. A very different renewal conversation.

Five questions that should decide your pricing model

1. What business result does the client care about?

   Specify the decision, risk, capability, or operational outcome that the work improves. If your answer is a list of activities or deliverables, your offering is still focused on effort.

   
   

2. What expertise remains hard to replace?

   Identify the knowledge, problem framing, proprietary methods, industry context, relationships and accountability for implementation that remain valuable when production gets faster.
   
   

3. How much of the result do you control?

   Accept responsibility for outcome-related risks only when you have sufficient control over the results. Client decisions, data quality, user adoption, timing, and third-party vendors may all affect the outcome.

   
   

4. Can both parties see and measure the outcome?

   Define the baseline, the metric, the timeframe, how results get attributed, and who owns what. An attractive outcome isn’t enough. Both parties must be able to measure it and agree on what it means.

   
   

5. What proof supports the value?

   Connect the offering to previous client outcomes, reduced risk, improved decision-making, time savings, enhanced capabilities, or protected revenue. Proof makes the premium easier to justify.

Match the pricing model to the evidence

Time and materials

Best fit: Scope or outcome is genuinely uncertain.

How to use it: Keep it for discovery, investigation and open-ended advisory work.

Fixed fee

Best fit: Scope is clear and change can be managed. Price the defined work.

How to use it: Protect it with a change-order process.

Subscription or retainer

Best fit: Value comes from ongoing access to expertise or capability.

How to use it: Package access, updates, decision support and agreed response times.

Outcome-based

Best fit: The result is measurable and you have strong influence.

How to use it: Define the baseline, attribution, controls, risk caps and payment trigger.

Hybrid

Best fit: Some work is in your control, some value depends on results.

How to use it: A clear base fee plus a limited variable component tied to a measurable outcome.

Why hybrid is usually the credible next step

Pure outcome-based pricing sounds like perfect alignment. However, it also transfers risk for things beyond your control. A 2026 working paper from the Stanford Digital Economy Lab maps consulting pricing against two questions: how easily can the client see the outcome, and how easily can they see your costs? The paper concludes that AI is likely to encourage companies toward hybrid and asset-based pricing models.

For most tech and consulting companies, this means charging a base fee for the work, expertise, and access within your control, alongside a smaller variable component tied to a result that you and the client can both measure and influence.

Better AI output doesn't mean better people

There’s a second risk that gets less attention. In a 2026 MIT and NBER study, 133 patent lawyers used an AI drafting assistant for three months. Their drafts improved, particularly those produced by junior lawyers. However, when all participants were later assessed without the tool, only the senior lawyers demonstrated a lasting improvement.

If AI performs tasks that junior employees would handle as part of their learning, organizations need a plan to develop and transfer expertise. That expertise will be central to the value they provide three years from now.

What to change now

• Stop using effort as your main proof of value. Explain the decision, risk, outcome, or capability the client gets.

• Package repeatable expertise into well-defined offerings, methodologies, diagnostics, tools, and decision support.

• Separate your productivity from the value delivered to the client. Faster delivery should improve your margins rather than automatically reduce your prices.

• Demonstrate value through outcomes and decision quality, supported by baseline measures and clearly defined client responsibilities.

• Provide executives and sales teams with a consistent explanation of how you create value beyond the tools you use.

This is a question of growth structure

Finance can’t fix this alone. The offering, value proposition, proof, executive narrative, sales language, and delivery model must all communicate a consistent message. If one element continues to sell hours while the others promise outcomes, buyers will notice.

As delivery becomes faster, the companies that continue to grow will be the ones whose expertise is easy to understand, trust, purchase, and repeat.

Questions for the leadership team

• Which offerings still rely on hours, headcount, or deliverable volume to justify the price?
• Where is AI reducing effort without changing what the client actually receives?
• Which outcomes can we measure and influence well enough to support variable pricing?
• What expertise would remain valuable if every competitor had the same tools?
• What evidence helps our buyer justify our pricing to their CFO?

Find out where your offer stands

The AI Value and Pricing Diagnostic identifies where AI is enhancing your offer and where it may be putting your pricing at risk. Ten questions. About three minutes.

Take the Diagnostic

Frequently asked questions

Will AI eliminate hourly consulting?

No. Time and materials still make sense when the scope or outcome can’t be defined reliably. Its role will shrink as repeatable work becomes easier to automate and package.

Should every consulting company adopt outcome-based pricing?

No. Pure outcome-based pricing can be risky when the client controls implementation or external factors influence the results. A hybrid model is usually more practical.

What becomes more valuable as delivery gets faster?

Specialized knowledge, problem framing, proprietary methods, trusted access, accountability for implementation, and evidence that reduces the buyer’s risk.

Should we pass AI savings on to clients?

Some of them, in certain circumstances. Clients expect AI to improve quality and speed. However, this does not mean that every hour saved should be deducted from the invoice. Deliberately determine which gains to share, which to retain as margin, and which to reinvest.

Where should a smaller company start?

With one offer. Define the buyer outcome, name the expertise that sets you apart, document the proof and choose the pricing model that matches what you can control and measure.