PR measurement starts with the question a campaign is meant to answer. Counting releases and clips is straightforward, but those totals alone cannot show whether the right audience understood or acted on the message. A useful metric has a definition, source, timeframe, and limitation that a colleague could check.
Four layers of evidence
01
Activity and outputs
Record pitches, releases, briefings, coverage, owned content, and distribution. Separate independent stories, syndicated copies, and paid placements.
02
Initial audience response
Use qualified visits, event attendance, message recall, questions, or comprehension checks when they match the objective. Report the denominator.
03
Outcomes
Look for a measured change in awareness, attitude, trust, inquiries, adoption, or behavior among the intended audience.
04
Organizational impact
Connect defensible audience changes to a broader goal, while acknowledging that other teams and outside events may contribute.
Metrics that need careful labels
Potential reach is an opportunity to see, not the number of people who read a specific article. Impressions can overlap, and identical release copies can inflate totals. Share of voice depends on which competitors, topics, channels, and period were included. Sentiment coding requires a documented method; it is not a universal fact about an article.
Referral traffic and tagged conversions can show a path from a link, but missing tracking and other promotions limit attribution. Advertising value equivalents put an ad price on editorial space and do not demonstrate PR value. Report the observable measure rather than converting every outcome into a speculative money total.
Build a useful scorecard
Begin with one or two clear objectives and the audience that matters. Record a baseline before activity starts. Select a small set of output, response, and outcome measures, plus a qualitative check on message accuracy. Give each metric an owner and source. A monthly report should explain what changed, the confidence in that finding, and the next decision, not just stack the highest possible counts.
AMEC's Integrated Evaluation Framework separates outputs, out-takes, outcomes, and impact. Its taxonomy warns against using a number measured at one layer to claim a result at a higher layer. That distinction helps a team celebrate genuine progress without pretending that a clip count proves trust or revenue.