Every communications team knows the slide. Fourteen million impressions this quarter, up eight per cent, a tasteful upward arrow. Heads nod around the table. And nobody in the room can say what any of those fourteen million people actually read, whether the campaign’s central claim survived contact with a single journalist, or whether the coverage said what the organisation needed it to say.
Impressions measure opportunity, not attention. They say nothing at all about message.
That gap is what message pull-through exists to close. The term describes whether a campaign’s key messages actually appear in the earned coverage it generates, in the organisation’s own language or a recognisable paraphrase of it. A health charity that spends a year saying “screening saves lives” and finds the phrase, or its substance, running through its coverage has achieved something an impressions count cannot see. One that finds its coverage dominated by a rival framing has learned something more useful still.
What the Industry Already Agreed
None of this is a new argument. AMEC, the international measurement body, published the first Barcelona Principles in 2010 and updated them to version 3.0 a decade later. The principles are blunt where it counts: measurement should identify outputs, outcomes and potential impact rather than stopping at volume, it should mix qualitative and quantitative analysis, and advertising value equivalents are not the value of communication. Ben Levine, an AMEC board director, put the qualitative case in terms most dashboards still ignore, urging teams to “use the full suite of methods to measure those outcomes” so they understand how messages are received and believed, not merely counted.
The industry agreed, then largely carried on. One recent review of PR measurement practice notes that three out of four teams still report impressions, though fewer than half track what audiences do afterwards. Impressions persist for an unglamorous reason. They are easy, they are big, and they always go up somewhere. They are also, increasingly, numbers that nobody around the table quite believes, which may be the strangest part of the ritual: a metric reported quarterly precisely because it commits no one to anything.
How Pull-Through Is Actually Measured
The mechanics are less mysterious than the jargon suggests. It starts with messages defined tightly enough to be findable: not “we want to be seen as innovative” but the three or four specific claims the campaign was built to land. Coverage is then coded against them, piece by piece, for presence, accuracy and prominence. Did the message appear? Was it rendered correctly? Did it lead the piece or trail in paragraph nine? Monitoring platforms automate the searching; the judgement calls, particularly on accuracy and tone, still benefit from human coding alongside the software.
Pull-through also gains meaning in company. Measured alongside share of voice and sentiment, it answers a sharper question than either metric alone: of the coverage the organisation earned in the conversations that matter, how much carried the organisation’s actual argument, and how much merely mentioned its name while a competitor’s framing did the talking.
Health communications raises the stakes on one of those dimensions. When the message is regulated wording — a screening eligibility criterion, a vaccination schedule, an indication statement — accuracy is not a quality measure. A disease-awareness campaign whose careful “speak to your GP about your risk” becomes “new wonder test” in the retelling has technically achieved coverage. It has also achieved a problem. A message that pulls through garbled can be worse than one that never pulls through at all.
The final move is pairing pull-through with behaviour. Coverage that carries the right message and coincides with branded search lift, referral traffic or other observable signals starts to look like evidence rather than activity.
What does a good number look like? Honest answer: it depends, and anyone offering a universal benchmark is selling something. Pull-through rates vary with how many messages a campaign carries, how contested the topic is and how tightly the messages were defined in the first place. The useful comparison is almost always internal — this quarter against last, this message against that one — rather than against an industry average of doubtful provenance.
And there is a limit worth stating plainly. Pull-through proves the message reached the page. It does not prove the message changed a mind, moved a behaviour or shifted a prescription pattern. That requires outcome measurement — surveys, tracking studies, behavioural data — which is where the Barcelona Principles were pointing all along.
Which returns us to the slide. The fourteen million impressions are not wrong, exactly. They are just an answer to the least interesting question available. The better slide, the one worth the room’s attention, shows which messages survived the journey from press release to published piece intact, which were rewritten into something else, and which never arrived. That is a slide a communications director can act on. The arrow is optional.
