Artificial intelligence is reshaping the economics of systems integration, pushing technology services providers to rethink how they deliver projects, price their offerings, and demonstrate value as enterprise customers demand measurable business outcomes.
For Endava, a global technology services company combining consulting, digital engineering, and systems integration, those pressures are driving changes to its delivery methodology and go-to-market strategy as traditional implementation models give way to AI-enabled services.
Speaking with Channel Insider, Keith O’Neill, Endava’s chief growth officer for commercial services, described an increasingly competitive market in which customers are reassessing their technology investments and demanding greater accountability from their service providers.
“It’s a buyer’s market right now in a lot of ways, because competition is high,” O’Neill said.
Endava shifts technology delivery toward agentic AI
Endava’s services span AI transformation, digital product engineering, cloud modernization, data, consulting, and managed services. Rather than emphasizing proprietary AI products, the company is positioning itself to help enterprises integrate technologies from across the vendor ecosystem.
Central to that strategy is Dava.Flow, an AI-native delivery methodology designed to move beyond traditional agile development processes.
O’Neill argued that simply inserting AI tools into existing workflows misses much of the technology’s potential.
“Agentic by nature doesn’t need all those agile rhythms and methodology,” he said, referring to the processes and ceremonies traditionally associated with software development.
Dava.Flow instead incorporates agentic automation, governance, and human oversight throughout the delivery lifecycle. The company says the approach can help organizations accelerate development while maintaining visibility into performance and outcomes.
The larger challenge, however, is helping customers move beyond experimentation.
According to O’Neill, many enterprises have invested in proofs of concept and pilot programs without successfully scaling AI or realizing the efficiencies they anticipated.
“The biggest thing is change management,” he said. “If you go too far, people push back.”
Endava’s approach includes working alongside customer teams, helping employees adopt new technologies, and gradually changing established processes.
AI changes pricing and economics for systems integrators
The transformation extends beyond engineering methodology into the commercial relationships between enterprises and their technology partners.
O’Neill said customers are increasingly seeking outcome-based engagements, fixed-fee arrangements, and greater financial accountability from services providers.
For traditional systems integrators and resellers, those dynamics suggest that implementation expertise and established vendor relationships alone may become less effective differentiators.
As AI reduces the time required for some development and implementation tasks, providers whose revenue depends heavily on billable hours could face pressure to rethink their pricing models. Fixed-fee and outcome-based contracts may offer alternatives, but they also shift more responsibility for delivery efficiency and project profitability onto the provider.
Specialized AI services create new opportunities for IT partners
Looking toward 2027, O’Neill expects broader enterprise adoption to follow today’s early movers, alongside the emergence of more specialized agentic solutions targeting specific industries and business functions.
That evolution could create opportunities for service providers capable of combining industry knowledge, technology integration, and ongoing optimization.
It also reinforces a larger change in the channel’s value proposition: As AI technologies become more accessible, differentiation may depend less on providing the underlying tools and more on helping customers deploy them effectively.
For O’Neill, the pace of technological change makes adaptability particularly important. Customers cannot eliminate uncertainty, but they can select partners and delivery models designed to evolve alongside the technology.