By Deepak Dhanak, Co-founder & COO, Rocket.new
For much of the past three decades, the global services industry has rested on a fairly stable economic logic: clients paid for specialised expertise delivered through time-bound engagements, and firms scaled by converting talent into billable hours. The model proved especially powerful in countries such as India, where a large pool of skilled professionals, combined with process discipline and cost competitiveness, helped build one of the world’s most formidable services economies.
Artificial intelligence may not dismantle that model overnight. But it is beginning to alter one of its central assumptions: that effort is a reliable proxy for value.
That matters because a growing share of white-collar work is becoming easier, faster and cheaper to produce. Tasks that once required teams of analysts, associates, researchers or developers can now reach a credible first draft in a fraction of the time. Market scans, competitive mapping, financial modelling, technical documentation, product specifications and strategy synthesis can increasingly be generated on demand with AI assistance.
This does not eliminate the need for services. Nor does it make professional judgement obsolete. But it does weaken the traditional link between labour input and commercial value. Once effort becomes easier to compress, hourly pricing becomes harder to defend as the dominant organising principle.
This is not simply a technology story. It is a business-model story.
The firms best positioned for the next phase of services will not be those that merely deploy AI tools to reduce turnaround times. They will be those that rethink how work is structured, priced and delivered. In other words, the shift is less about automating individual tasks and more about redesigning the economics of service delivery.
That distinction is important. Many firms will adopt AI as a layer within their existing operating model. They will use it to improve productivity in research, drafting, documentation or workflow management, while leaving the broader architecture of the business unchanged. Those gains will be real. But they are unlikely, on their own, to produce durable strategic advantage.
The larger opportunity lies in moving from effort-based delivery to outcome-oriented design.
That does not mean every service can or should be sold as a neatly packaged output. Complex transformation work, strategic advisory, legal judgement, enterprise implementation and other high-trust engagements will continue to require iteration, ambiguity and close human involvement. But even in these cases, AI is shifting the baseline expectation. Clients are likely to become less tolerant of paying legacy prices for work whose production cost has fallen sharply. What they will continue to pay for is not raw effort, but context, judgement, accountability, domain depth and dependable execution.
This is where the debate becomes particularly significant for India.
India’s rise in services was built on a combination of human capital, process maturity and delivery scale. Much of its success depended on the ability to mobilise talent efficiently across large volumes of work. If AI reduces the amount of human effort needed for substantial parts of that work, then the next source of competitive advantage cannot simply be additional labour capacity. It will have to come from systematised knowledge, productised workflows, faster execution and stronger alignment between fees and business outcomes.
In that sense, AI may mark a transition from labour leverage to systems leverage.
The implications are considerable. Firms that successfully codify repeatable methods, reduce reinvention across engagements and build AI into the operating core of delivery may begin to exhibit qualities traditionally associated with software businesses: greater consistency, higher throughput, improved margins and less linear dependence on headcount growth. Services, in parts, could start to behave more like products — not because human expertise disappears, but because it becomes more concentrated at the points where judgement truly matters.
That is likely to reshape both competition and valuation. Markets have historically rewarded services businesses differently from software businesses because their economics are different: lower gross margins, heavier dependence on labour and more limited scalability. If AI-enabled firms can move even partially away from those constraints, the distinction may begin to narrow in some segments. The result will not be that all services firms become software companies, but that the most successful among them will operate with far greater structural efficiency than their predecessors.
There are, however, reasons for caution.
Not all AI-generated work is reliable. In regulated or high-stakes environments, the cost of error remains material. Clients may be eager for faster and cheaper delivery, but they will also demand traceability, accountability and defensibility. In many industries, the promise of AI will be moderated by governance requirements, legacy systems and organisational inertia. The transition, therefore, is unlikely to be immediate or uniform.
Yet the direction of travel is increasingly clear.
The principal threat to services firms is not that AI will replace them outright. It is that AI will expose the fragility of models built on monetising effort that is no longer scarce. Firms that treat AI as a marginal productivity aid may improve at the edges while remaining vulnerable at the core. Firms that use it to re-architect delivery, capture institutional knowledge and price more explicitly against outcomes will be in a stronger position.
The future of services is unlikely to be defined by who adopts AI first. It will be defined by who redesigns around it most effectively.
And the question clients will increasingly ask is not how many hours went into the work, but whether the outcome was delivered reliably, quickly and well.