Certinia study finds integration key to AI success

Certinia has published a global study on AI performance and operational integration in professional services and IT organisations, based on a survey of 1,000 decision-makers across five markets.

The findings suggest operational integration is a stronger indicator of AI success and financial performance than spending on AI or talent alone. Organisations reporting successful AI outcomes were more than three times as likely to be fully integrated across systems and workflows as those with mixed results: 28% versus 8%.

Businesses with the highest profitability showed a similar pattern. Firms with margins above 40% were three times more likely to have fully integrated operations than barely profitable peers. Meanwhile, 68% of organisations achieving net revenue expansion above 100% were aligned or fully integrated, compared with a sector-wide full integration rate of 22%.

The report covered professional services and IT and technology leaders in the US, Canada, the UK, Australia, New Zealand, and Singapore. Sapio Research carried out the study.

The results indicate that fragmented systems remain a barrier for services companies trying to apply AI effectively. Organisations with stronger links between sales, delivery, finance, and customer success were more likely to report better AI adoption, wider account expansion, and stronger margins.

"You cannot scale a modern services business on fragmented infrastructure," said DJ Paoni, Chief Executive Officer, Certinia.

"Connection is the true differentiator for the organisations pulling ahead this year. This data shows that organisations fully integrated across sales, delivery, finance, and customer success are consistently outperforming on AI adoption and profitability. Removing that fragmentation frees people to focus on what matters most: delivering exceptional value to customers," Paoni said.

Perception gap

Another theme in the study was a divide between senior leaders and operational teams. Executives rated their organisations more favourably than practitioners on several measures, including forecasting confidence, AI success, and the ability to grow without adding headcount.

Leaders rated forecasting confidence 24 percentage points higher than practitioners did, while the gap on AI success was 16 points. On growth without additional headcount, executives were 10 points more optimistic than delivery teams.

The research also suggested customer retention goals are not consistently understood across organisations. While 62% of executive leadership teams said they had defined net revenue retention goals, only 45% of services and delivery teams said the same, despite often being closest to customer relationships.

Operational risks

The study found weaker confidence among companies that were only partly integrated. Just 38% of partially aligned organisations reported high confidence in their resource, demand, and revenue forecasts, compared with 75% of fully siloed companies and 78% of fully integrated businesses.

That suggests partial efforts to connect systems may add complexity without delivering a shared operational view. The report described this as a weak point for organisations caught between legacy structures and full integration.

Growth through account expansion also appeared limited. Only 6% of organisations globally reported net revenue expansion above 100%, with a marked regional divide: 9% in North America compared with 2% in Asia-Pacific.

AI performance also varied sharply by sector. Among organisations that had deployed AI, the share reporting moderate or significant success ranged from 43% in accounting, tax, and audit firms to 74% among IT service providers.

Business model changes

The report also pointed to shifts in how services firms expect to price work and hire staff. Three-quarters of respondents said they expected to increase outcome-based pricing over the next 12 months, even though nearly one-third already reported difficulty managing hybrid or complex billing models.

Hiring plans reflected the same pressure to adapt. Eighty-two per cent of leaders said AI and data specialists were their top hiring priority, while the same share expected revenue to grow without a proportional increase in billable headcount.

The findings add to a broader debate over whether AI projects fail because of the models themselves or because of the data and processes around them. For services companies, where work often depends on coordination across finance, project delivery, and customer management, the study suggests those operational foundations remain unsettled.

An external view included in the study added brief context. "Organisations that move directly to AI deployment without first establishing a clean, unified data environment consistently encounter the same problems: AI outputs that are unreliable, workflows that break under edge cases, and a rapid deterioration of user trust," said Mickey North Rizza, Group Vice President, Enterprise Software and Agents, IDC.

"The failure is not the AI - it is the data infrastructure beneath it," Rizza said.