Why brand, PR and community results are hard to claim
The roles by specialism guide describes brand work as judged on slower measures and social and community roles as judged on reach and engagement plus conversions further along the journey. It stops there. It does not say what to write when your evidence is a survey wave, a lift study, a search trend or a log of coverage.
The guide to stating your contribution is written for any measure. It has you separate your part from the team's, a result from the movement that would have come without the work, and a platform's reported figures from the business's own records, and it gives a ladder of wording that runs from tested work to work left unmeasured. Brand, PR and community work strain those rules in four specific ways, each traceable to what the owner guides say about the measures:
- The evidence is a difference with uncertainty, a behaviour or a count. A tracker or a lift study reports a difference between waves or groups with a range around it, a search trend is a behaviour that stands in for memory, and a count of coverage or replies describes output. None of them is a result without something to compare it with.
- The reading may arrive after the case. A slow measure can still be unread when the case is written, and a verdict given before the reading exists rests on the output alone.
- Outputs look like results. A count of pieces, posts or replies is simple to report, but it describes what was done, not what changed.
- Others made part of the work. Agencies, designers and approvers made assets, and journalists and community members, not you, decided what ran and what was said in reply.
This guide is for the marketer who ran the work and has to write about it for a portfolio, a CV, an interview or a promotion case. It builds no measure: how a tracker, a lift study, a search series or a coverage log is built and read stays with the owner guides, linked in the table below. It says nothing about how a hiring manager, a client or an employer reads brand or PR evidence, because no page the site cites covers that. What you may say about the work at all is settled by your contract, the client's terms and your organisation's policy; employment, confidentiality, data-protection and disclosure questions start with those, with legal and compliance colleagues and, where a question stays open, an outside adviser. Everything beyond what the linked guides state is general practice and is labelled as such.
Name the change the work was meant to make
Before choosing a measure, write one sentence naming the change the work was meant to make, and for whom. For brand work it might be more of the category's buyers recalling the brand; for a PR story, coverage in the titles buyers read that carries the agreed message; for a community, members answering each other's questions. This step is general practice. The sentence fixes what counts as evidence, and it stops a measure that happened to move from becoming the goal after the fact. The KPIA metric tied directly to an objective. the work served belongs in the sentence too.
The contribution guide's use of the KPI tree applies: find the input you moved and follow its branch upward.
Then state the level of each figure. The PR measurement guides separate outputs, outcomes and impact. As general practice the same three apply to brand and community work. A set of assets made, a post count and a reply count are outputs. Recall or consideration among the people the strategy names, and members answering each other, are outcomes. Orders or revenue are impact, which needs its own evidence and the comparison the contribution guide describes. An output can be claimed as an output without apology; it goes wrong when it is written in the language of an outcome.
Name the population as well. The awareness guide in the table below says an awareness figure describes a defined audience and that the brand KPI should say which one. A case that says awarenessThe share of a defined audience who recall a brand unprompted or recognise it when prompted, measured by survey. rose, with no population beside it, leaves the reader unable to tell which figure moved.
The contribution guide also cites the Chartered Institute of Marketing's framing of reporting as accountability: a contribution put in terms that people across the business would recognise. For slow measures, the practical reading is general practice. Say what level a figure sits at, and where the business keeps a record of its own, such as visits, enquiries or orders, place it beside the survey or search figure instead of letting the survey figure stand in for the business.
What each measure can support
Keep this table open while you write. Read across a row: the measure, what it shows, what it cannot show and the kind of wording that suits it. Each row links the guide that owns the measure, so how the measure is built and read is in that guide and not here. The wording column is general practice; the other columns follow the linked guides.
| Measure | What it shows | What it cannot show | The wording it supports |
|---|---|---|---|
| Coverage logThe consistent record of every piece of coverage, by title, tier, story, prominence, message, sentiment and link, from which PR measures are counted. and counts, read as outputs (media monitoring and coverage reporting guide) | Which pieces appeared, in which titles, with what message and prominence, and which story produced each | That anyone remembered or acted on a piece; a title's total audience overstates who read one piece | Describes the work and its output; the pieces the log ties to your story are the ones to claim |
| Quality-scored coverage and PR outcomes such as earned share of voice in target titles (PR KPI guide) | Whether coverage in the titles that matter carried the message, and how the earned share compared with rivals | Business impact from click-based attribution, which undercounts earned media; any value in money | Contributed to or coincided with, depending on whether a baseline and a fixed set of rivals exist |
| TrackerA survey that asks the same questions of a fresh sample of category buyers in repeated waves, so changes in awareness and associations show over time. waves, read with their intervals (brand health report tutorial) | What the strategy's audience recalls or would consider, for the brand and its competitors, wave against wave | A movement inside the interval, which the tutorial marks too early to tell; cause, since a before-and-after reading leaves other changes in the same weeks mixed in | Contributed to, with the interval stated, when the change is clear of it; otherwise no clear movement yet |
| A platform brand liftThe difference in brand measures between people exposed to a campaign and a comparable unexposed group. study (brand awareness and lift guide) | A difference in survey answers between people who could see the ads and a held-back group, with uncertainty | An independent audit, a total across platforms, an effect on sales, or what the campaign would do at a different reach and frequency | An estimated difference for that platform's own ads, in the survey measure, within its range |
| Branded search and share of searchSearches for a brand as a share of searches for all the brands in its category, read as a trend beside the brand tracker and market share. as proxies (share of search tutorial) | Interest people expressed by searching for the brand, read as a share of the category and compared year on year | Awareness itself: existing customers searching, promotions and news move the figure too | Coincided with, with the events of the period named; an early signal read beside the tracker |
| Community health measures and response standards (community management guide) | Whether the team met the standards it set, and whether the community is getting healthier: conversation against broadcast, sampled sentiment and what the community produces | Health from met standards alone, since the two are separate; worth in a raw count of interactions, which scores spam and a real conversation the same | Describes the operation you ran; a health reading set against an earlier period coincided with your work |
Two notes apply across the rows. Where two measures agree, say so: the brand health tutorial treats a change that holds across waves, or that the search measures echo, as stronger than a single reading, and, as general practice, where they disagree, report the disagreement. And do not put a money value on coverage. The PR guide retires advertising value equivalence and reports that the IPA, among other bodies, directs its effectiveness evidence at outcomes and business effects, not at notional media value, so a claim about coverage rests on what changed, not on what the space would have cost.
Time: when a claim can be made
Slow measures set the date of a claim, not the writer. As general practice, an output can be claimed once it exists, a change in the audience once a reading taken after the work exists, and a claim about the business once there is a comparison to read it against. The owner guides say how each measure is read over time: the awareness guide reads branded search with a lag against brand activity, the PR guide treats awareness and impact measures as slow movers, and the community guide reports against the team's own standards. This page states no lag, cadence or interval of its own. Take them from those guides and keep them in the claim.
A claim about change needs an earlier reading to compare with, a baselineAn estimate of what a metric would have done without a given change, campaign or promotion, against which the effect is measured. taken before the work ran. If none was taken, say so and claim less. As general practice, date each figure: the wave or window it covers, the day it was read and what it was compared with. Keep the tracker's own interval in the sentence, and name the events in the period that could have moved the figure, such as promotions, launches and news, in the way the share of search tutorial annotates them.
When the reading arrives after the case, write the case with what has been read, mark the later reading as open, say which measure will settle it, and update the case when it arrives. Do not write an expected result as a result. If you may leave the role first, note in your private record who holds the later reading.
A case can also show that you know the difference between a verdict and a reading. A campaign written off because sales did not move is the situation the page on awareness up but no sales lift takes apart: it separates campaigns that were judged wrongly, such as sales read across the flight weeks alone, from campaigns that were delivered badly. Your write-up gains from showing which of the two you found.
Your part when others make the work
The contribution guide has you list everyone who touched the result, and its list of contributors includes a designer, an agency and a manager. In brand, PR and community work, add to that list: the agency and designer who made the assets, whoever approved them, the journalists who chose what ran and the members who answered each other. Your own line is one entry on it, and the full list is what makes that entry precise.
The roles guide describes brand work as writing or approving a brief, reviewing agency work against it and guiding other teams on using the brand. As general practice, those are the verbs to reach for when an agency made the creative: briefed, chose the message, reviewed against the brief, approved, rejected, redirected. Designed, wrote and produced describe making, and they stay with whoever made the work. The contribution guide's verb test carries over.
In PR, the coverage log has a field for the story or campaign that produced each piece, or for whether it was reactive or unprompted. As general practice, claim the pieces the log ties to your story, and keep the weaker and unprompted pieces in the record. In community work, the community guide counts members answering each other as a sign of health. Those answers are the members' own, so what you can claim is the replies, the rules and the featuring that made room for them.
Where an AI tool made part of the work, the tutorial on describing AI-assisted work splits the tool's part from yours and a reviewer's. The same idea of splitting parts applies when the maker is a person or an agency.
Wording that matches the evidence
The contribution guide's ladder ranks wording by the evidence behind it, from tested work down to a plain description of work that was not measured. It is neutral about measures, so this page keeps that ladder and adds rows for the evidence the table above produces. The hierarchy of evidence explains why the rungs sit where they do: the higher kinds of evidence leave fewer other explanations standing. The quoted lines in the table below show the shape of the wording only; they are not anyone's record.
| Evidence you hold | Where it sits on the contribution guide's ladder | How it reads |
|---|---|---|
| A coverage count and its quality fields | The rung for describing the work, since it measures output | "I pitched the launch story, and the log records coverage in target titles carrying the agreed message; no outcome measure was set, and I would now add one" |
| A tracker wave clear of its interval, against the baseline wave | The rung for contributed to | "As part of the brand team, I briefed the campaign; unaided awareness among category buyers rose between the baseline wave and the next, beyond the tracker's interval" |
| A tracker wave inside its interval | The rung for describing the work, plus what was read | "I briefed the campaign; the wave after it shows no movement beyond the tracker's interval, so the claim stays open until a later wave" |
| A platform lift study, with its range | Beside the top rung: a comparison with a held-back group, but for that platform's own effect on survey answers | "In the video flight I briefed, viewers gave higher ad recall than a held-back group by an estimated amount within the study's range, which is the platform's own estimate" |
| A proxy that moved: branded search or share of search | The rung for coincided with | "Share of search rose, read year on year, after the campaign; a promotion ran in the same period, so I read it as an early signal beside the tracker" |
| A community health reading | The rung for coincided with, or for describing the work if only standards were measured | "Comments per post and members' replies to each other rose while I ran the community, and the team met its own response standards; I did not test whether my work caused the rise" |
Two habits apply to each row, and both are general practice. First, carry the count or interval behind a figure into the sentence, following the small numbers guide on any rate built on a small count: a handful of respondents, pieces or replies can swing a long way by chance, so show the count beside the rate and say too few to tell where that is the finding. Second, name what else changed in the period, using the contribution guide's check, and drop to a lower rung if another event could have produced the same movement.
In a case, a CV line and an interview answer
The portfolio guide structures a case as context, decisions, results and learning, and these claims fit that structure without change. Put the change the work was meant to make in the context. Put your briefs, rejected options and approvals in the decisions, along with which choices the agency's work reflected. Put each figure in the results with its level, its date and wording from the table above, with outputs labelled as outputs. Put the measure you would add, and any reading still open, in the learning.
For a CV line, the CV guide leads each achievement line with the result and asks you to state your part when a team produced the result. A one-line form for slow measures is general practice: the measure at its level with its wording verb, the comparison with its interval, then your verb and the work, the audience and who made what. For example: "[Measure] [wording verb] against [comparison], [inside or beyond] the tracker's interval, after briefing and approving an agency-made campaign for [audience]; [other event] ran in the same period." Keep the real figures in the private record the portfolio guide describes, and share the version your organisation allows.
For an interview, the interview preparation tutorial prepares stories as situation, task, action and result, with what you learned. For slow measures, prepare four short answers: what the figure was compared with, what else was happening, what you decided and what the agency or approver decided, and what the later reading will tell you. A story in which the evidence was thin and you said so, then named the measure you would add, shows judgement, and nothing in it overstates the result. A promotion case follows the same lines.
A communications lead writes up a windmill's open-days campaign (illustrative)
Paglesk's communications lead writes up an open-days campaign
Paglesk is an invented community-owned windmill with a visitor centre. Its part-time communications lead runs press, a supporters' group and the social posts for a campaign promoting six open days, and an agency made a set of 3 posters. Before choosing any measure she writes down the change the campaign was meant to make: more local residents thinking of the windmill as a place to visit on an open day.
She keeps outputs apart from outcomes. The outputs are the posters, 20 social posts and 8 pieces of coverage that the coverage log ties to the campaign's story, 6 of them carrying the open-days message; 3 other pieces in the log were unprompted, and she leaves them out of the claim. The outcome evidence is a small local tracker of about 100 residents per wave. Unaided recall of the windmill as a place to visit is 17% in the first wave and 23% in the second, each carrying an interval of about 8 points either way, so the two waves cannot be told apart yet. The visitor centre's own records show 410 visitors on the open days against 360 on the same days a year earlier, but a council festival ran in the same weeks, so she can say only that the rise coincided with the campaign.
She separates the parts. The agency's designer made the posters; she wrote the brief, chose the open-days message, reviewed the proofs against the brief and approved them with a trustee. Her case reads: "I wrote the brief and chose the message for a six-open-day campaign, reviewed an agency-made poster set against it and approved it with a trustee. I pitched the story to local titles, and the coverage log records 8 pieces tied to it, 6 carrying the open-days message. Open-day visitors were 410 against 360 on the same days a year earlier, in weeks when a council festival also ran, so I describe the rise as coinciding with the campaign. Two tracker waves of about 100 residents each show unaided recall at 17% then 23%, each carrying an interval of about 8 points either way, so I claim no movement in awareness yet; the next wave is the check." Her CV line carries the same facts in one sentence, and for interview she prepares answers on the festival, on what the agency did against what she did, and on what the next wave will decide.
The numbers belong to the invented case; the pattern is what carries over. She names the change first, labels outputs as outputs, keeps the tracker's interval in the sentence, names the other event, states the agency's part apart from her own and leaves the open question open.
The marketing skills hub covers learning, practising and evidencing a skill; this guide belongs to the evidence stage.
Common mistakes
- Reporting outputs as outcomes. Pieces of coverage, posts and replies show what was produced. Label them as outputs and read an outcome beside them.
- Quoting a tracker movement inside its interval. A change the interval cannot separate from no change is not yet a result. Say that it is open.
- Calling a platform's lift the brand's lift. A lift study estimates that platform's own effect on survey answers, and it does not measure sales.
- Reading a search rise as awareness. Branded search and share of search move with promotions, news and existing customers. Name the events and read the proxy beside the tracker.
- Claiming the agency's work or the members' answers. Briefing, approving and setting rules are yours. Making the creative and answering each other are not, unless you did it.
- Writing the later result in advance. An expected reading is a plan. Mark the case open until the reading arrives.
- Leaving out what else ran in the period. A festival, a promotion or a news story can move the same figures. Name it beside the figure so the claim does not ignore it.
- Pricing coverage as if it were advertising. The value of the space says nothing about what changed.
