Why most marketing teams don’t get results from AI – and what the 6% do differently

Nearly every team uses AI. Only 6% see real results.

Bain’s 2026 survey of 1,397 CMOs, CFOs and senior marketing and finance executives found a wide gap between AI adoption and AI impact. Leaders – companies growing revenue more than 11% a year and gaining market share – are pulling away from everyone else.

6%

of marketing organisations say AI delivers significant performance impact today (Bain)

2×

Leaders are twice as likely to attribute double-digit revenue growth or cost savings to AI (Bain)

47%

of leaders describe AI as a core capability, against 30% of laggards (Bain)

3.7×

Leaders are 3.7 times more likely to fully redesign workflows around AI (Bain)

Separately, Ahrefs’ 2026 survey of 301 marketers found only 2.7% trust AI-generated content without a person reviewing it first – a reminder that tools are only as good as the expertise behind them.

AI in marketing: three reasons teams stall, three things the 6% do

1. Why teams stall: too many tools, no strategy
2. Why teams stall: pilots that never scale
3. Why teams stall: no workflow redesign
4. What the 6% do: centralise strategy
5. What the 6% do: embed AI in existing platforms
6. What the 6% do: test at scale

Why teams stall: too many tools, no strategy

AI has spread team by team, subscription by subscription, with no central plan. Bain found leaders are 1.8× more likely to follow a centralised AI roadmap with priorities set from the top.

Without one, teams run duplicate tools and can’t connect what AI produces to business results.

Ask: Who owns AI strategy in our marketing team?

Why teams stall: pilots that never scale

A chatbot demo is easy. Connecting AI to your CRM, marketing automation platform and customer data – securely, without breaking what already works – is where most projects stall.

Bain found 16% of laggards rely on general-purpose tools such as ChatGPT, Copilot or Gemini, against just 6% of leaders.

Ask: Which of our AI tools are connected to our real data and platforms?

Why teams stall: no workflow redesign

Layering AI over old processes produces marginal gains. Bain found leaders are 3.7× more likely to fully redesign workflows, and twice as likely to restructure teams and job descriptions around AI.

That means changing who does what, which skills matter and how performance is measured – not just adding a tool to the stack.

Ask: Have we redesigned any workflow around AI, or only added AI to old ones?

What the 6% do: centralise strategy

In leading organisations, AI priorities are set at executive level, not left to individual teams. Bain found more than 40% of leaders put 11% or more of their budget into AI, compared with only a quarter of laggards.

The goal is alignment: every AI initiative connects to a business outcome. Tonic’s AI consulting starts with an audit of where AI touches your marketing today and where the gaps are.

Ask: Does every AI project we run have an owner and a business measure?

What the 6% do: embed AI in existing platforms

Bain found leaders are 2.4× more likely to embed AI directly into their marketing tools – CRM, marketing automation, analytics – rather than running it separately.

For most teams that means using the AI already inside platforms such as HubSpot, Salesforce Marketing Cloud, Marketo and AEM, connected to first-party data. Larger changes are often part of a wider digital transformation.

Ask: Which AI features in our existing platforms are we not using yet?

What the 6% do: test at scale

Bain found leaders are 8.5× more likely to run 100 or more marketing experiments a month. AI embedded in real workflows, tested against real outcomes, is what moves results.

Testing also keeps quality in check: Ahrefs found 89% of marketers now manually check AI content, up from 80% in 2025.

Ask: What would good look like in 12 months, and how would we measure it?