How technology services firms get paid is changing. For most of the industry's history, services revenue was a function of effort (hours billed at a rate). That arrangement is expected to give way to project-based pricing, and increasingly to models that price the outcome itself, often delivered through a combination of software, data, and people. The progression matters because the pricing structure ultimately determines the margin profile, scalability, and the market's valuation of the business.
The original services model is time-and-materials (T&M). A firm staffs a team, bills the hours, and earns a margin on the spread between the bill rate and fully loaded cost. The model is simple, low-risk for the vendor, and easy to sell, but it has a structural ceiling. Revenue scales only with headcount. Margin is fixed by the spread between rates. There is no operating leverage, and the vendor is paid more for being slower than for being efficient. From an investor's perspective, T&M revenue is the lowest-quality revenue a services firm can generate: not recurring, not differentiated, and structurally exposed to wage inflation and competition from lower-cost geographies. It is also the model most directly threatened by AI, because the unit being sold (an hour of human work) is precisely what AI is making cheaper.
Fixed-price projects would be the first real evolution. The vendor takes on delivery risk, committing to a defined scope at a defined price, and in exchange captures the upside if it can deliver more efficiently than estimated. This forces scoping discipline, project management rigor, and a degree of productization in how delivery is organized. Margin profiles improve for firms that can deliver consistently, and the most sophisticated practitioners begin investing in accelerators, frameworks, and reusable assets that compound across engagements. But fixed-price work is still bound by the project: it has a beginning and an end, and revenue does not compound across periods. Buyers may also remain skeptical because the vendor's incentive could be to scope tightly rather than to drive outcomes.
Outcome-based pricing is the next frontier, and it is increasingly inseparable from software. Instead of paying for hours or for a project, the buyer pays for a result (i.e. tickets resolved, claims processed, leads converted, costs reduced, revenue generated, etc). The offering is rarely pure services in this model. It typically combines a software platform, data assets, and a service team, sold together as an end-to-end capability. Pricing can take many forms, such as per-transaction fees, a percentage of the value created, gain-share arrangements, or subscriptions tied to defined service levels, but what unites these structures is that the vendor is paid for what the work produces, not for the effort to produce it.
This shift has historically been difficult to execute because outcomes were delivered by humans whose costs scaled linearly with the work. AI changes that math. When agents and software absorb a meaningful share of delivery, the vendor's cost to produce an outcome decouples from the price the buyer is willing to pay. The result is a margin profile that could resemble software: recurring revenue, declining unit costs, and operating leverage that compounds. We have written elsewhere about how this convergence is reshaping the line between services and software (https://alten.capital/blog/when-services-revenue-starts-to-look-like-software).
Alten Capital invests in technology services businesses. Reach out to explore potential partnerships.