
When we wrote about software investors earlier this month, the central question was whether AI makes a business more valuable or erodes what customers pay for. That question has now spread well beyond software, and buyers across sectors are changing the criteria they use to assess acquisitions as a result. For owners of services and other mid-market businesses, AI exposure is becoming part of how buyers set the price.
The businesses facing the hardest questions are those whose revenue grows with the number of people doing the work. That includes outsourced back-office and customer service operations, IT services, staffing, content and marketing services, and parts of professional services billed by the hour. If AI allows a client to get the same result with fewer people, revenue per client is likely to fall, however strong the relationship.
Public markets have already moved. Among India's large listed IT services groups, average EBITDA multiples fell from around 18 times to around 10 times between mid-2025 and mid-2026, as slower growth and uncertainty about AI weighed on valuations. Private buyers are applying the same scrutiny to smaller businesses with similar delivery models.
The same technology can work strongly in a company's favor. A business that uses AI to deliver more with the same team can widen its margins, take on more clients and win work from slower competitors. Businesses with licensed professionals, physical service delivery, regulated processes or proprietary data are often harder to displace, and AI can make each of those assets more productive.
Buyers pay for evidence. A business that can show AI already in use, with measurable effects on productivity, margins or client outcomes, is far more convincing than one describing plans.
Diligence teams are moving beyond asking whether a target uses AI. They want to know whether AI strengthens or weakens the economics of the business, and they tend to look at five areas:
The protections buyers give most weight to are regulatory, compliance or security barriers and deep integration into a client's day-to-day workflow. Proprietary data, specialist expertise and switching costs also count, though views on them vary more from buyer to buyer.
How a business charges can matter as much as what it does. Hourly or headcount-based pricing passes the benefit of automation straight to the client. Fixed-fee, managed service and outcome-based contracts let the provider keep more of the efficiency it creates. Businesses that have started moving clients onto those models before a sale are in a stronger position when the question comes up in diligence.
Buyers will form a view on AI exposure whether or not the seller offers one, so it is better to set out the case first. Before going to market, owners should understand which parts of their revenue are most exposed, what AI is already doing inside the business and what it has delivered, how contracts and pricing are structured, and what retention and client data say about the durability of relationships. That case belongs in the information memorandum and the management presentation, supported by numbers.
Two businesses with similar revenue and margins can now be valued very differently depending on how convincingly they answer the AI question. The gap can be worth several turns of EBITDA. Owners who understand their exposure, can show what AI is already doing for them and can explain why their client relationships will last are best placed to achieve full value, whether in a sale or a financing.
