
Muck Rack finds that PR teams are beginning to measure how brands appear in AI-generated answers, while many still question the metrics they have to present to clients. Its new research puts those pressures in the same frame: visibility tracking is expanding faster than some teams’ ability to explain what the numbers mean.
The company’s 2026 State of PR Measurement announcement, published November 18, documents broader adoption of LLM tracking alongside concerns about accuracy, inconsistent testing and client-driven reporting. The research is notable because it treats measurement credibility as a management issue, not just a problem for analytics software.
Key Takeaways
- Muck Rack’s new survey finds growing interest in tracking AI-answer visibility alongside ongoing pressure to report weak metrics.
- Self-selected survey results describe reported practices, but do not prove that any tracking method improves commercial outcomes.
- PR leaders should assign a measurement owner, repeat the same AI-visibility tests and disclose the limits of each metric.
Table of contents
Jump to each section:
- What the new survey actually measured
- AI visibility is becoming a PR reporting responsibility
- PR teams still report metrics they do not trust
- AI adoption is ahead of repeatable measurement
- What communications leaders should change
What the new survey actually measured
Muck Rack surveyed PR professionals from September 22 to October 19, 2026, then removed spam, large outliers, identifiable mistakes and low-effort responses. The company recruited participants primarily through its own database and email contacts.
571 usable survey responses remained from 697 collected, according to Muck Rack’s November 18 announcement.
That makes this a useful view of how people in Muck Rack’s reach work, rather than a probability sample of every public relations practitioner worldwide. The survey also asks respondents whether they have taken steps toward measuring LLM visibility, a broader standard than maintaining a robust, independently audited program.
For an agency presenting these figures to a client, the distinction matters. The results can help identify questions to ask in a measurement review, but they cannot establish that the same percentages hold across every market or that particular tracking methods improve commercial outcomes.
AI visibility is becoming a PR reporting responsibility
LLM visibility is now part of the reporting conversation for many respondents. That means testing whether a brand appears in AI-generated answers, examining the sources those answers cite, and comparing the findings with the communications team’s existing coverage work.
55% of respondents had taken steps to measure LLM visibility in 2026, compared with 28% in the previous year’s Muck Rack survey, according to the company’s November 18 announcement. Another 23% said they intended to begin.
| Survey year | Respondents who had taken steps to measure LLM visibility |
|---|---|
| 2025 | 28% |
| 2026 | 55% |
37% of teams already measuring LLM visibility assigned primary responsibility to PR or communications, compared with 19% assigning it to SEO or digital, according to Muck Rack.
Those figures describe reported ownership arrangements, not evidence that one department has a superior method. The practical issue for clients is who has authority to define the prompts, check the citations and explain changes between reporting periods.
The underlying question goes beyond a dashboard. A brand may appear frequently in one set of AI prompts yet disappear when an audience changes its wording, location or product category. Communications teams need a stable test design before turning visibility counts into a performance trend.
PR teams still report metrics they do not trust
The emerging AI measures sit alongside a more familiar difficulty: practitioners are being asked to supply figures they do not consider useful. The tension is not necessarily that the numbers are fabricated. It is that a technically available metric, such as raw reach or mentions, may have little relationship with the outcome a client actually wants to assess.
63% of respondents said they sometimes had to report metrics they regarded as weak or unreliable. Muck Rack identified client expectations as the most frequently cited reason at 38%, followed by executive expectations at 28%.
Independent trade outlet PRNEWS highlighted this reporting tension in its coverage of the study. The research gives PR managers an opening to separate what stakeholders request from what the team can reliably interpret. A coverage volume total may still belong in an operational appendix, but it should not automatically stand in for brand trust, lead quality or a shift in audience understanding.
This is particularly relevant for agencies working with multiple client teams. If a marketing leader wants an impression count and a corporate communications leader wants message pull-through, a single summary score may obscure the disagreement instead of resolving it. Both goals need separate definitions and evidence.
AI adoption is ahead of repeatable measurement
Using an AI tool does not guarantee that an AI-related result is sufficiently stable to report. Muck Rack’s findings expose that distinction, especially when teams attempt to compare answers across different prompts or reporting periods.
54% of respondents said they used AI or AI-powered tools for measurement or reporting, while 72% cited accuracy of AI results as a concern, according to Muck Rack.Only 31% of teams already measuring LLM visibility repeated tests on a regular schedule, and fewer than half used consistent prompts, the company reported.
Greg Galant, Muck Rack’s cofounder and CEO, argued that LLM visibility should be held to the same standard as other PR metrics. In the company’s announcement, he said: “That means measuring it consistently, testing it regularly and tying it to outcomes leadership cares about.”
That advice is operationally significant. If a measurement vendor changes its model, the prompt set changes, or tests run from a new country, a change in the score may reflect the test environment rather than a new reputation outcome. Teams should log those conditions before claiming that a campaign caused movement.
What communications leaders should change
The new report is more useful as a checklist for measurement governance than as a league table for AI-visibility performance. A communications leader can apply its findings without buying a particular product or accepting every vendor-defined score.
- Assign one accountable owner. Name the PR or communications lead who approves the LLM prompt set, alongside an SEO or analytics partner who audits technical changes. Record which function signs off on conclusions.
- Freeze a repeatable test set. Use the same core prompts, model versions where controllable, countries, dates and query categories over successive periods. Log any unavoidable changes before comparing results.
- Separate outputs from outcomes. Report press coverage, AI citations and answer visibility as observed outputs. Use available website, sales or audience evidence to test whether they correspond to a business outcome.
- Challenge a weak required metric in writing. When a client still wants a reach total, provide it with its limitations and a stronger companion indicator, such as relevant coverage quality or message accuracy.
There is also a competitive issue for PR firms. Measurement vendors and consultancies can all supply a growing range of visibility figures. Buyers should ask whether the agency can reproduce a finding and explain the uncertainty, rather than assess capability by the number of numbers on a dashboard.
Muck Rack’s November 18 report documents rising attention to AI visibility while showing how much measurement discipline remains unsettled. It offers a more credible starting point for better reporting, not proof that the profession has agreed on an industry standard.