ISBA has found that 99% of advertisers are engaging with generative AI, but only 14% report a significant impact on business results.
The industry body’s 2026 Gen AI Survey, based on 200 responses from advertisers, suggests that marketing teams are adopting AI tools rapidly while still working through how those tools improve commercial performance.
The survey found that 65% of individuals are using generative AI regularly, while 28% said it has meaningfully affected their day to day work. Among advertisers prioritising effectiveness as the primary objective for AI adoption, that figure rises to 55%.
At organisational level, advertisers prioritising effectiveness are twice as likely to report a significant impact on business results. ISBA said the findings show a distinction between using AI for efficiency and using it to improve marketing outcomes.
Nick Louisson, director of agency services at ISBA, said: “The pursuit of greater marketing effectiveness is an enduring exercise in innovation and continuous improvement.”
Efficiency gains remain an important starting point, according to ISBA, but advertisers looking at generative AI through the lens of effectiveness are more likely to report meaningful impact and changes to agency relationships, ways of working, scopes, and team structures.
The survey also found that policy comprehension has improved year on year as organisations build governance frameworks for responsible adoption. Agency relationships are evolving as advertisers rethink team design and operating models around new capabilities.
Marketing departments have moved quickly from trial use into practical application. AI tools are now being used for drafting, briefing, content variation, image ideation, transcription, localisation, insight gathering, and workflow support. Those uses can increase speed and reduce manual effort, but output volume does not automatically improve brand performance, customer acquisition, retention, or revenue.
The 14% business impact figure gives the industry a more demanding benchmark than adoption. A campaign team may produce more creative variants, faster research summaries, or cheaper content without improving distinctiveness or conversion. Greater output can also introduce new layers of review, legal sign off, claim checking, brand safety assessment, and data governance.
Measurement remains a recurring weakness in marketing transformation. Gartner has flagged brand measurement failure, warning that marketing leaders often struggle to connect brand health with enterprise growth. ISBA’s survey points to the same discipline problem in AI adoption, where efficiency is easier to evidence than commercial effectiveness.
Agency relationships are also being recast. Routine production, early analysis, localisation, versioning, and workflow tasks may be handled differently as advertisers integrate AI into internal teams. That does not remove the need for agencies, but it changes the kind of value clients are likely to seek. Strategic judgement, creative direction, governance, media expertise, and performance interpretation become more important when basic execution becomes faster and cheaper.
Those changes will alter scopes and remuneration. If AI reduces the time needed for some forms of production, agencies and advertisers will have to agree how value is priced. Time based charging becomes harder to defend when technology changes the labour input, while outcome based or capability based models require stronger agreement on what success means.
Governance is another constraint. Improved policy comprehension suggests advertisers are moving beyond informal experimentation, yet the risk profile rises when AI tools are used across entire departments. Copyright, confidential data, inaccurate outputs, bias, disclosure, model selection, and use of generated imagery all need clear operating rules. Without them, marketing teams may gain speed while increasing legal and reputational exposure.
Earlier technology cycles offer a useful comparison. Programmatic buying, marketing automation, customer data platforms, and attribution tools all promised efficiency, but value depended on data quality, organisational discipline, and the ability to connect tools to decisions. Generative AI is advancing faster, although the management requirement is familiar: technology changes performance only when it is connected to capability, workflow, incentives, and measurement.
Budget scrutiny is likely to increase as experimentation becomes embedded. AI spending may continue, but marketing leaders will be pressed to explain where it improves outcomes rather than simply lowering production costs. The survey suggests that advertisers using AI to pursue effectiveness, not only efficiency, are further ahead in proving value.
Near universal engagement means the competitive advantage will not come from access to the tools. It will come from how advertisers use them: where human judgement remains central, how governance is applied, whether measurement improves, and whether new operating models produce better marketing rather than more marketing.
You must be logged in to post a comment.