For a decade, PR measurement meant counting clips, tallying share-of-voice, and slapping an 'ad value equivalency' number on a slide. Finance teams never trusted that number, and they were right not to.
Today, a real shift is underway: PR teams are borrowing measurement tools from performance marketing to show earned media's actual business impact. It is not perfect yet, but it is far more honest than the old scorecard.
Why the old metrics stopped working
Share-of-voice tells you how loud you are compared to competitors. Sentiment tells you whether coverage felt positive or negative. Neither tells a CFO whether a Forbes feature moved a single deal forward.
Progressive brands are now running multi-touch attribution (MTA), a method that assigns partial credit to every touchpoint a buyer interacts with before converting, and applying it to PR. When a prospect reads a press hit, then later searches the brand name and fills out a demo form, MTA can register that earned mention as an assisted touch instead of ignoring it entirely.
That single change, giving PR credit for touches earlier in the funnel, is what separates 2025-era measurement from the ad-value-equivalency era.
A robust earned media effort routinely lifts SEO rankings, branded search volume, and inbound lead flow, all of which can be tracked in existing marketing analytics stacks without any new PR-specific tooling.
Where MTA breaks down for PR
MTA works best when every touch is trackable: clicks, form fills, pixel fires. PR's biggest wins rarely look like that.
A founder profile in a trade publication, an executive quote in a market roundup, or a sponsorship mention on a podcast, none of these generate a clickable link a marketer can tag. For B2B brands especially, this class of coverage does not produce trackable clicks, so MTA misses it almost entirely, even though it may be shaping how a buying committee perceives the company.
This is exactly the gap media mix modeling (MMM) was built to fill.
What media mix modeling adds
MMM is a statistical technique, originally built for TV and offline ad spend, that estimates the relationship between marketing investment and downstream outcomes like revenue, without needing individual-level tracking. It looks at PR volume and timing across weeks or months, then measures whether pipeline or revenue moved in step.
Applied to PR, MMM can credit brand investments, coverage volume, executive visibility, analyst mentions, even when no single reader ever clicked a tracked link. It captures the revenue impact of brand-building activity by measuring the statistical relationship between that investment and downstream business results.
Most sophisticated teams in 2025 do not pick one method. They run a hybrid: MTA for the trackable digital journey, MMM for the broader brand and market-level signal, and controlled experiments, like geo holdouts or staggered announcement timing, to confirm the two are pointing at real causation and not coincidence.
Geo holdouts, delaying or withholding a PR push in one region while running it in another, are the closest thing PR has to an A/B test. If pipeline moves faster in the region that got coverage, that is causal evidence, not correlation.
A worked example: reading a branded-search lift
A B2B software company ran a coordinated push, 6 trade press placements plus 2 podcast mentions, over a 3-week window and tracked branded search volume weekly for the 8 weeks around it. Baseline branded search averaged 4,200 monthly searches. In the week its largest placement ran, branded search jumped to 6,100, a 45% lift, then stayed elevated at roughly 5,300 (26% above baseline) for the next 5 weeks before decaying back toward baseline.
Pipeline created from organic and direct channels, the two sources branded search typically feeds, was 31% higher in the 60 days following the coverage window compared to the prior 60-day period, with no other major campaign running in parallel. No single reader's click was ever tracked, but the correlation between coverage timing and both branded search and pipeline was strong enough that the CFO approved doubling next quarter's PR budget.
What this means for your PR reporting
You do not need a data science team to start moving in this direction. Three practical steps get you most of the way there.
- Tag everything trackable. UTM-tag every link in every press kit, guest byline, and executive quote you place, so MTA tools can at least see the digital portion of the journey.
- Track branded search and direct traffic around coverage dates. A spike in branded search the week a major feature runs is a leading indicator MMM-style analysis can pick up even without perfect attribution.
- Partner with marketing ops, not just comms. MMM and MTA already live inside most marketing analytics stacks; PR's job is to feed them clean, dated coverage data, not to build new infrastructure from scratch.
The goal is not a perfect number. It is a defensible one, something you can bring into a pipeline review without getting laughed out of the room.
Mistakes that undermine PR attribution credibility
- Claiming causation from a single coincidental spike. One good week of pipeline after a placement is not evidence; look for the pattern across multiple coverage windows before drawing conclusions.
- Ignoring confounding campaigns running at the same time. If a paid campaign or product launch overlaps your PR push, any lift could belong to either, and a finance team will find the overlap even if you don't mention it.
- Reintroducing ad value equivalency as a bridge metric. AVE was already discredited before MTA and MMM existed; using it to "translate" PR into dollars undoes the credibility this whole shift is trying to build.
- Reporting MMM results with false precision. MMM produces a statistical range and a confidence interval, not a single exact number, and presenting it as exact invites the same distrust the old metrics earned.
- Waiting for a perfect model before reporting anything. A directionally honest, caveated correlation shared quarterly builds more trust over time than a perfect model that never ships.