HubSpot research finds gap between AI adoption and results

Nearly all marketing teams use artificial intelligence in some form, but far fewer report transformational results, according to HubSpot’s 2026 State of AI in Marketing research. The research puts AI use at 98% of marketing teams, while 10% report transformational outcomes. The contrast highlights a difference between putting AI tools to work and producing measurable business results.

HubSpot describes that difference through the concept of AI maturity. For go-to-market teams, maturity measures how consistently AI contributes to outcomes that can be measured. In marketing, those outcomes can include changes in marketing-qualified leads, customer acquisition cost and revenue, as well as gains in customer experience.

The framework looks beyond the number of tools a team uses or experiments it has launched. A marketing team may have several pilots under way without showing that any are repeatable or connected to business results. HubSpot points to factors such as whether an experiment connects to the customer relationship management system, has an owner and can demonstrate a before-and-after result.

Four stages of maturity

HubSpot’s framework groups teams into four stages: Early, Emerging, Established and Transformational. It also sets out an assessment intended to help teams identify where they stand and consider their next steps. The guide addresses data readiness and responsible AI governance alongside the use of AI across the customer funnel and the measurement of its impact.

Marketing appears further along than other customer-facing functions in the research. HubSpot says 42% of marketers report advanced AI use, defined as the Established or Transformational stages. The comparable figures are 27% for sales and 23% for service. The same broad definition of maturity applies across those functions, although their workflows and metrics differ.

For marketing teams, the more mature approach described by HubSpot involves documented workflows, clean customer data and human oversight. It also requires a connection between AI activity and the measures the business uses to assess performance. On that basis, the research distinguishes the presence of AI in a team’s work from its ability to produce results consistently.

What this means for marketers

Marketers assessing their AI use can start by looking at existing workflows and pilots, rather than counting tools alone. Check whether each initiative has an owner, uses reliable customer data and can be linked to a measurable outcome. Where results are available, a before-and-after comparison can help show whether an experiment has delivered an identifiable change.