Eighty-two percent of marketers say artificial intelligence has improved their productivity, according to a new report from Emplifi. The same report puts a sharp qualifier on that headline number: only 35% report major gains.

That 47-point gap is the story. Most marketers using AI feel some benefit; few can point to a step change in output. Emplifi, which sells customer engagement and social media management tools, has an obvious commercial interest in encouraging AI adoption, and the report reflects that posture. What it does not do, at least in the published summary, is explain why the gap exists.

What does the 82% figure actually measure?

The number covers marketers who say AI boosts their productivity. It is a self-reported perception, not a measured change in output, cycle time, or cost per campaign. That distinction matters. A marketer who saves twenty minutes a week on copy drafts and a marketer who cut reporting time in half both count toward the same 82%.

The 35% figure — those seeing major gains — is the more honest indicator of where the technology stands in day-to-day marketing work. Nearly two-thirds of AI users, in other words, are experiencing modest improvements at best.

Who was asked, and what is missing?

Emplifi's published summary does not state the sample size, the survey dates, the markets covered, or how "major gains" was defined to respondents. Until those details are available, treat both percentages as directional rather than definitive. A vendor survey of its own customer base can skew toward teams already committed to AI tooling — and already inclined to justify the spend.

What is also not said: whether the productivity gains translated into revenue, headcount changes, or budget savings. Productivity that never reaches a financial statement is a soft metric.

Why the gap likely persists

The report's own framing hints at the pattern trade readers will recognize from earlier technology cycles: broad adoption, shallow impact. Tools get purchased widely; workflows get redesigned rarely. The 35% who report major gains are plausibly the teams that rebuilt processes around AI rather than bolting it onto existing ones.

That reading is consistent with the numbers, though the report itself does not test it. Any buyer evaluating AI tooling should ask vendors the corollary question directly: what share of your customers see major gains, and what did those customers change to get them?

The practical takeaway

For marketing operators, the Emplifi data sets a reasonable expectation: AI is likely to help, and it is unlikely to transform — at least not without deliberate process change. Budget accordingly. A tool that 82% of users find useful but only 35% find transformative is a marginal-efficiency investment, not a strategic leap.

The next round of survey data will tell more. If the 35% figure climbs while the 82% holds, adoption is maturing into real impact. If it stalls, marketers are buying convenience, not capability.