Marketing leaders fret over AI brand discovery gaps

Marketing leaders are increasingly concerned with how artificial intelligence systems represent and recommend their brands, as AI agents become more involved in discovery, customer engagement and marketing operations.

Research from Contentful found that 91% of respondents believe the future of marketing increasingly depends on persuading the systems that influence people.

The survey of 350 marketing decision-makers across the US, UK and Australia examined how organisations manage marketing agents, measure their visibility in AI systems and assign responsibility for brand representation.

The research found that AI-powered search and answer engines and AI agents are already major areas of focus. Some 51% of respondents said the two are equally important, while 29% prioritise search and answer engines and 19% prioritise agents.

Agentic commerce is also moving into planning. Some 53% of organisations said they have a defined agentic commerce strategy and are actively building for it. A further 37% are exploring the area without a defined strategy.

Agent autonomy

Marketing agents are already being used across an average of 3.6 functions per organisation.

Search, SEO and answer engine optimisation management are the most common applications, reported by 48% of respondents. Customer service and post-purchase engagement follow at 46%, while 44% use agents for campaign personalisation and segmentation.

Only 2% reported having no agent use anywhere in their organisation.

The degree of autonomy varies. Some 43% of teams allow agents to act across most marketing tasks with periodic review. Another 34% restrict agents to low-stakes decisions, while 21% require human approval before an agent can execute an action.

The report identifies the approval burden created by unnecessary human checks as an "approval tax". It distinguishes this from reviews that are needed to catch errors or prevent incorrect claims from reaching customers.

The research also found differences based on department size. Some 28% of small marketing departments would extend agent autonomy across most tasks, compared with 45% of larger departments.

Brand identity

Brand differentiation is another concern as AI systems increasingly summarise and compare companies.

Some 85% of marketing leaders believe AI-generated summaries will make most brands sound alike. At the same time, 95% believe brands with strong, well-codified identities will widen their lead.

The report argues that AI systems can form their understanding of a company from multiple sources, including websites, product information, metadata, reviews, third-party coverage and citations.

Conflicting information across those sources can make it harder for AI systems to establish an accurate representation.

The research found that 37% of executives see the structure and clarity of content as the main factor distinguishing a brand when AI systems are doing the summarising. A further 20% pointed to third-party validation and authority, while 17% selected the content itself.

Australian respondents placed greater emphasis on structure and clarity. Some 54% identified it as a key differentiator, compared with 35% in the US and 32% in the UK.

AI visibility

Accurate representation in AI systems is becoming a senior marketing priority. Some 83% of respondents ranked it as a top or high priority for the coming year.

Yet measurement remains less widespread. The research found that 55% of marketing teams regularly measure how they appear across the agentic web.

The study describes changes in AI-generated brand descriptions as "representation drift". Tracking this requires organisations to establish a baseline and repeatedly ask AI systems the same buyer questions to identify changes.

Only 33% of respondents said they track recurring buyer questions to identify content gaps. Some 31% have built or acquired a tool to measure agent visibility, while 24% audit brand mentions in AI-generated outputs.

The US and Australia showed a marked difference in measurement. Some 92% of US respondents had measured how AI systems describe their brand at least once, compared with 72% in Australia.

Australian marketers were also more concerned about inaccurate AI representation. Some 40% identified it as their single biggest concern, compared with 22% in the UK.

"For Australian marketers, the question today should be how well-positioned the brand is to be understood, verified and recommended by the systems now standing between us and our customers," said Charlie Bell, Senior Director, Solution Engineering, EMEA & ANZ, Contentful.

"For Australian brands, closing the measurement gap is the first step - our data suggests they may be flying comparatively blind on AI representation at the exact moment they're most worried about getting it wrong," said Bell.

Workforce impact

The growing use of AI is also affecting how marketing teams are structured.

Some 96% of marketing teams reported at least one structural change because of AI. Among them, 46% had incorporated automated systems or digital workers into team workflows, while 43% had added roles focused on AI or automation.

Training has also increased, with 53% reporting higher levels of AI training or enablement.

At the same time, 21% of organisations had eliminated certain roles and 25% had reduced or paused entry-level hiring because of AI.

The research links this trend to the development of professional judgement. Junior marketing roles have traditionally provided experience in areas such as campaign execution, testing, customer exposure and reviewing work.

As those tasks become automated, organisations may need to find other ways for junior employees to develop the judgement required to assess AI-generated work.

Responsibility

Responsibility for AI brand visibility remains divided between functions.

Marketing holds primary responsibility at 45% of organisations. Dedicated AI, data or innovation teams hold it at 21%, while IT or engineering is responsible at 14%.

C-suite leaders outside marketing hold primary responsibility at 11%. Another 6% reported having no clear owner or shared responsibility without a clear owner.

The report argues that responsibility spans brand and content, product information, data and technology, making a single-function approach difficult.

It recommends assigning one accountable marketing executive alongside named owners across content, product and data. The model is intended to establish clear decision rights and a route for tracing AI-related problems back to the source of an error.

"Our research tells us that AI makes the human role in marketing all the more consequential. Organisations navigating this shift must draw clear lines around agent autonomy, protect how their people develop judgement, and track how AI systems represent them over time," said Bell.