Lists of public relations best practices are everywhere, and they all share a shape: a handful of confident instructions, each presented as a lever — pull this one and coverage, trust or visibility improves. That shape deserves scrutiny before the list does. A best practice is a claim about cause and effect: do X, and Y gets better. Some of those claims are true in a limited way. Some are true about the work itself and false about the outcomes. And some are false in a way that matters, because a practitioner who believes them will promise clients results that were never theirs to give. This article takes the standard list — the seven practices nearly every version of it contains — and corrects each one in turn. The corrections are not new inventions. Each belongs to a longer argument made elsewhere on this blog, and each practice below points to the place where its correction is argued in full. What this list adds is the test itself: for every practice, what can it honestly promise, and what can it not?

Read the list that follows as two lists interleaved: the work a practitioner actually controls, and the outcomes other people decide. The first list is real and worth doing well. The second is where the genre’s promises go to be believed. Keeping the two apart is the whole discipline — and a best-practices list that promises outcomes is, in the end, a sales document in list form.

The Definition, Corrected First

Before the practices, the definition they serve. The standard version of this article opens by saying that the key goal of public relations is to influence how people perceive a brand. That sentence should be refused, plainly. Perception is not a material a practitioner works on. It is a conclusion that readers, customers, journalists and regulators reach from evidence — from what an organisation has done, said and left on the record. A practitioner cannot reach into that process and adjust it. What the practitioner can do is govern the evidence: the accuracy of the facts issued, the consistency of the record over time, the speed and honesty of corrections. The work’s object is the record and the conduct behind it. Perception is what the record earns, favourably or not.

This is not wordplay. A field that defines its goal as influencing perception will measure itself by perception, promise perception, and eventually start treating the evidence as raw material for the conclusion it wants — which is how accurate records quietly become optional. The branding discussion on this blog makes the same point from the identity side: a brand is an authored artefact; its reputation is not. The authored parts can be designed. The conclusion others draw cannot. Every practice below is therefore stated in the corrected sense: as work on the record, with outcomes named as other people’s decisions.

Practice One: A Consistent Record, Not a Consistent Slogan

The first practice in every list is a clear, consistent message. Corrected, it reads: consistency of facts. A message is wording, and wording may — must — adapt: to the audience, the format, the moment. Facts may not adapt. The date the product launched, the figure in the results, the terms of the offer, the name of the person responsible: these must be identical in the release, the advertisement, the landing page, the interview and the annual report, this year and in five years’ time, because the checking layer compares them. Contradiction between channels is not a branding blemish. It is a factual failure, and it is found.

The coordination doctrine behind this is argued in full in the piece on media and PR inside integrated marketing: integration honestly means one factual record underneath every channel, not one slogan repeated by all of them. The practice also keeps the original’s soundest small advice — avoid jargon — and sharpens it. Jargon in an outward document is usually a fact wearing a costume: a phrase like ‘leveraging synergies’ where a plain sentence would have to name what actually happened. The test proposed in the jargon glossary applies to every draft: any phrase that cannot be translated into plain words without losing its promise was carrying the promise, not the meaning. A consistent record is written in language a checker can cash.

Practice Two: Know Your Publics and Their Stakes

The second practice is usually ‘know your target audience’, supported by research into demographics, preferences and behaviour. The correction is in the noun. The people an organisation’s communication must answer to are not an audience — a crowd that watches — but publics: distinct groups, each with a stake in the organisation and some power to act on it. Customers can leave. Regulators can compel. Staff can resign, testify or speak to a journalist. A local community can object, organise and be believed. Each public reads the same record for a different risk, and knowing them means knowing the stake, not the demographic profile.

The framework is set out in the article on aligning internal and external publics: list the publics, name what each stands to gain or lose, and check that the record serves each one’s legitimate need to know. Research, in the corrected practice, is reading the record of what those publics have actually done and said — the questions journalists on the beat keep asking, the objections raised at the last consultation, the complaints that recur — rather than assembling persona documents that invent a representative person’s tastes. Personas have their place in product and service design; they are signposted here only as the neighbouring discipline. In public relations, an invented reader is a poor substitute for a real public with a real stake, because the real one will check the work against its own experience and the invented one never will.

Practice Three: Media Relationships Buy a Hearing, Not Coverage

The third practice — develop strong media relationships — is the one where the genre’s promise is most often cashed dishonestly. In its standard form it slides, within a paragraph, from building relationships with journalists to gaining favourable coverage and controlling the narrative. Both halves of that slide are false. Coverage is an editor’s decision, taken for the editor’s readers, and it is not transferred by familiarity. The narrative of a published story belongs to the publication that published it. A practitioner who promises either is promising someone else’s property.

What relationships honestly buy is a fair reading: the pitch opened rather than deleted, the claim checked against the sender’s past accuracy before it is discounted, the phone call returned when the story turns difficult. That is worth having, and it is earned in a specific currency — accuracy over time, availability when the news is bad as well as good, corrections issued before they are demanded. The seeker’s side of this discipline is set out in the article on getting media coverage honestly, and the underlying principle — that coverage is an editor’s output, never a product anyone sells — is the premise the whole correction stands on. Stated as a practice: be accurate, be available, be correctable. The outcome stays where it belongs, with the desk.

Practice Four: Digital and Social Channels Are Authored Work

The fourth practice urges the reader to embrace digital PR and social media, and in its standard form it commits a quiet category error: it treats the organisation’s own channels as a route to coverage. They are not. A post on the organisation’s own account is authored work — written, scheduled and published by the sender — and its reach, however large, is owned reach. Presenting it as media coverage, in a report or in the practitioner’s own mind, confuses a publication the sender controls with a judgement someone else made.

Run honestly, the owned channels are valuable precisely as authored work: the fastest place to publish the record, correct it and link to it. The sorting that governs this — authored, bought, decided elsewhere — is laid out in the communications tactics article, and it gives the practice its rules. Authored channels must be attributable, so the reader knows whose voice this is; labelled wherever any payment is involved; and consistent with the record everywhere else, because the screenshot outlives the edit. What they cannot do is substitute for the decided-elsewhere family: no quantity of posting replaces a desk’s decision, and a plan that buys more owned output to compensate for coverage that did not arrive is producing noise around a silence.

Practice Five: Content Is the Record’s Working Documents

Create compelling, shareable content, says the fifth practice, usually with storytelling as the method. Corrected, the practice is duller and stronger: content is the record’s working documents. The release, the briefing note, the data page, the biography, the image with its rights cleared — these are the artefacts a checker actually uses, and their virtues are not charm but checkability: dated, sourced, consistent with everything previously issued, and correctable in public when they turn out wrong.

‘Shareable’ is demoted accordingly, from goal to occasional property. Material that deserves sharing gets shared sometimes; material engineered for sharing — the emotional frame first, the facts fitted inside it — inverts the discipline, and expert readers can tell. Storytelling framing is refused for the same reason: the facts are the story’s skeleton or there is no story, only an arrangement. The working documents themselves are covered where they belong: the release as a document of record, formatted so a desk can use it without repair, and the online press room as the shelf where the standing facts live, findable at the moment a journalist, a committee or a regulator comes looking. A practice that maintains the shelf will look unspectacular in every monthly report and decisive on the one day the record is checked under pressure.

Practice Six: Monitor the Record; Do Not ‘Manage’ Reputation

The sixth practice arrives as a pair — monitor and manage your brand’s reputation — and the correction is to split it. Monitoring is a genuine practice and stays. Managing reputation, as a named activity, goes. Reputation is not a substance held in the practitioner’s hands; it is the sum of other people’s conclusions, and it cannot be operated on directly. What can be operated on is the record and, behind it, the conduct. The honest practice name is the longer one: monitor the record, correct it, and where the record is accurate and the conclusion still adverse, take the finding to the conduct, not to the messaging.

Monitoring itself, done properly, is the listening layer: a running record of decisions and mentions — what was covered, what was said, what was got wrong — kept so that errors surface while they are still correctable. The discipline around corrections is the sharp end of it: a correction made early and openly becomes part of the record, while one extracted later becomes part of the story — a principle the event release article applies to the day a mistake is most likely to be made in public. Sentiment scores and reputation indices, the dashboards this practice usually buys, measure neither the record’s accuracy nor the conduct’s quality; they measure a model’s guess at tone. An organisation that wants to know its standing should read what is actually said about it, in full, and check it against the facts — slower, and the only version that tells it anything it can act on.

Practice Seven: Measure the Work, Not the Weather

The final practice — measure and evaluate — survives, with its object replaced. The standard version reaches for key performance indicators, engagement rates and sentiment: measurements of other people’s responses, presented as the practitioner’s results. The corrected version measures the work, because the work is what the practitioner controls and the only thing measurement can honestly grade. Was the record accurate? Was it consistent across channels and across the year? Was it ready when the predictable moments arrived? Were corrections issued, and honoured, at the speed the error required?

This is not a lowering of ambition; it is a refusal of false precision. The fuller argument — that coverage counts, reach estimates and sentiment scores convert other institutions’ decisions into the sender’s achievement, and that a measurement regime built on them will eventually start managing the numbers instead of the record — is made in the article on PR measurement. The tools matter here only as categories: listening tools, clipping services, analytics packages. Each collects something real; none of them turns a decision someone else made into a result the practitioner produced. Measure what you did. Report what others decided, labelled as theirs.

What the Corrected List Adds Up To

Read back, the seven corrected practices reduce to one discipline stated seven ways: make the record true, keep it consistent, and keep it findable and correctable by the people who will check it. Consistency of facts; publics known by their stakes; relationships that buy a hearing; owned channels run as authored work; documents built to be checked; monitoring that corrects rather than cosmetics; measurement that grades the work. Each is the same commitment at a different station of the job.

And the honest promise of a best-practices article can now be stated, because it is smaller than the genre’s usual one. None of these practices will make a desk decide in your favour, a public trust you, or a market move. Outcomes of that kind belong to editors, readers, customers and events, and an article that promises them is a sales document wearing a list’s clothing. What the practices do is narrower and real: they make an organisation the kind whose claims survive checking. In a field where every important reader — the journalist, the committee, the regulator, the citizen — is a checker by profession or by interest, that is not a consolation prize. It is the only advantage public relations can actually sell, and it is enough to build a practice on.