Table of Contents
Every profession accumulates tools, and public relations has accumulated more than most: databases of journalists, monitors that never sleep, systems that send a release to a thousand inboxes before breakfast, software that drafts, counts and charts. Together these are sold under one label — PR tech. Strip away the marketing and the PR tech meaning is plain enough: it is the layer of software that sits on top of a judgement profession, doing the parts of the work that are mechanical so that the parts that are not can get more attention. This article is about technology in public relations practice — the tools practitioners themselves use. It is not about public relations for technology companies, which is a different subject, covered elsewhere in this series, with different problems.
The honest starting point is a standing rule that governs everything below: a tool changes the speed and the scale of PR work, never its standard. A weak pitch researched badly and written worse does not improve by being sent to ten thousand people instead of ten; it merely fails at industrial volume, in front of an audience that remembers. Technology in this profession is leverage. Leverage multiplies whatever it is applied to — including error.
What PR tech actually means
It helps to divide the stack by function rather than by brand, because functions endure while products churn. PR technology does five kinds of work. It finds: databases of media contacts, outlets and their subject beats. It listens: monitoring services that catch mentions of a client, a competitor or a subject across news and social channels. It sends: distribution systems that deliver releases and pitches, manage lists and log who was sent what. It drafts: writing assistance, from transcription and summarising software to the current generation of AI tools. And it counts: measurement dashboards that aggregate coverage, reach and engagement into charts. Every product in the field is some bundle of these five, wrapped in a subscription.
Notice what is absent from the list. No category exercises judgement: none knows whether a story is true, whether a journalist will care, whether a claim will survive being checked, or whether this is the week to stay quiet. Those decisions are the profession. The tools serve them the way a good diary, a good filing system and a fast bicycle serve a reporter — they remove friction from the errands so the judgement gets the hours. When a tool is marketed as though it performs the judgement itself, the marketing has crossed the line this article will keep returning to: confusing capacity with outcome.
Media databases
The media database is the oldest PR technology and the most misunderstood. What it genuinely does is clerical and valuable: it gathers names, outlets, roles and subject areas into a searchable whole, so that a practitioner starting a list does not start from a blank page. What it cannot do is know a journalist. Journalists move beats, change outlets, go freelance, develop enthusiasms and aversions, and a database record decays from the day it is entered. The contact who covered consumer technology last spring may cover health by autumn; the listing will not always say so.
The working rule follows: a database is a starting list, not knowledge. Knowledge is what the practitioner adds — reading what the journalist actually writes now, noting what they have covered and what they have sworn off, recording the conversation from the last approach. That is the craft this series covers in its article on contact lists, and the database’s proper role is to feed it, not replace it. A pitch built from a database row alone produces the field’s most recognisable failure: the message addressed to the right job title at the wrong person, about a subject they left two jobs ago. The sender had technology. What they lacked was the ten minutes of reading the technology was supposed to make time for.
Monitoring and listening
Monitoring tools watch the published world so the practitioner does not have to: alerts when a client is named, a competitor covered, a subject ignited. Listening tools extend the watch to social conversation. Their genuine contribution is timeliness and completeness — the mention in the regional title nobody in the office reads, caught the morning it appears rather than the week after.
Their limit is interpretation, and it is absolute. An alert finds a mention; it does not tell you what the mention means — whether this is the story that matters or noise, whether the tone the software assigned bears any resemblance to the tone a reader would perceive, whether the conversation being listened to is a market or a dozen accounts agreeing with each other. Press tracking as a discipline, and the honest measurement of what coverage achieved, is a separate subject covered elsewhere in this series. This article’s point is the narrower one: the tools deliver the raw material of awareness. Awareness itself — knowing which of fifty alerts deserves a phone call — remains a human shift, worked by a human.
Distribution systems
Distribution systems handle the mechanics of sending: the formatted release delivered to lists and wires, the embargo timed, the send logged. The mechanics matter. A release that arrives broken, mistimed or addressed to a list the sender cannot reconstruct afterwards is a professional failure, and systems exist to prevent exactly that. Used well, distribution technology also enforces a basic honesty of record: who was sent what, when, under what embargo — the sort of fact that becomes precious the first time a story appears early.
But distribution must be described as what it is: delivery. Distribution is not coverage. A system can put a release in a thousand inboxes; it cannot make one editor care. The industry habit of reporting sends as though they were results — releases “distributed” counted like stories placed — is a category error with a long pedigree, and software makes it easier by counting the sends so beautifully. The honest ledger has two columns: what we sent, and what was published because a journalist chose it. Technology keeps the first column immaculately. The second column is still earned the old way.
Drafting and AI assistance
The newest layer is drafting assistance, and it deserves a cooler assessment than it usually gets. The current generation of tools is genuinely useful at the clerical end of writing: transcribing the interview accurately, summarising the long document, producing a first draft of the routine announcement from verified material, searching one’s own archive for the figure used last year. In each case the tool works on material the practitioner supplies and can check, and the time saved is real.
The standing risks are equally specific. These tools produce fluent errors: sentences that read beautifully and are wrong — a date shifted, a title invented, a quotation smoothed into something the person never said. They generate detail where none was supplied, because fluency abhors a gap. And they write in the house style of nobody, which in a profession where voice is a brand asset — as this series’ branding article argues — is not a neutral property. The governing rule is the one that has always applied to junior assistance of any kind: anything that goes out carries a named human’s verification. A practitioner who sends AI-assisted text they have not checked line by line has not delegated the drafting. They have delegated their professional responsibility, which was never theirs to delegate.
Measurement dashboards
Measurement dashboards aggregate the countable: mentions, reach estimates, engagement figures, share of voice, sentiment scores assigned by machine. Their virtue is speed — the counts arrive instantly, pre-charted, in the format of a board slide. Their vice is the same speed, because a dashboard answers the questions it can count and quietly trains its users to ask only those. Counts are delivered instantly; meaning is delivered never. No dashboard can tell you whether the coverage described your client accurately, whether the right twelve people read it, or whether the story that ran will still look like an asset in five years — the questions the branding and ethics articles in this series identify as the ones that matter.
The discipline, as with monitoring, is to keep the tool in its place: dashboards as the raw feed for a human assessment, written by someone who has read the coverage itself. The vanity-metric warning that runs through this series’ treatment of measurement applies here in its purest form — the number chosen because it went up, presented as though its going up were the achievement. Software did not invent that habit, but it has made it effortless, and effortlessness is how bad habits become standard practice.
Where technology earns its keep
Against these cautions, the genuine workflow gains deserve to be stated plainly, because they are considerable. Technology earns its keep first in research before pitching: a practitioner’s hour now buys what a day once did — the journalist’s recent output gathered and searchable, the outlet’s coverage of the subject mapped, the competitor’s announcements assembled. The tool does not make the pitch good, but it removes every excuse for pitching blind. Second, in records that survive staff changes: the contact history, the record of who was sent what and what they said, the standing documents — the backgrounder chief among them — maintained in shared, dated, searchable form rather than in a departing employee’s inbox. A practice whose memory lives in its systems can lose a person without losing the decade that person knew. The backgrounder-maintenance discipline this series describes elsewhere is, in large part, a technology habit: versioned, dated, owned.
Third, in logistics across distance and time: the embargo that must lift in three markets at coordinated moments, the announcement sequenced across time zones, the crisis rota that reaches the right person at 3 a.m. their time rather than the office’s. These are unglamorous and they are where errors used to breed — the release sent early, the list from last year, the correction that reached half the recipients. Systems that hold the timing, the list and the record steady are worth their subscriptions on the strength of the disasters that quietly do not happen.
The buying discipline
Because the tools are sold harder than they are used, buying them is itself a PR discipline — the practitioner as sceptical audience. Three questions do most of the work. What work does this replace? A tool should retire a named, tedious task; if the answer is a gesture at “efficiency”, there is no case yet. What does it cost in attention? Every dashboard demands to be watched, every alert stream to be triaged, every system to be fed with maintained data; the subscription price is the smaller half of the cost. What happens to our records if we cancel? The contact histories, the archives, the logs — if they cannot be exported in usable form, the practice is not renting a tool, it is depositing its memory in someone else’s vault, and should price that accordingly.
Two habits support the questions. The first is pilot before commitment: a small team, a real campaign, a defined period, judged against the work the tool was bought to replace — not against the demonstration. The second is scepticism toward the pitch that promises outcomes. A tool can honestly promise capacity: faster research, steadier logistics, fuller records. A tool whose sales case promises coverage, rankings or results is promising things that depend on journalists and publics, and is therefore either confused about how PR works or hoping the buyer is. Feature lists deserve the same treatment: a long one usually describes a product trying to be several mediocre tools instead of one adequate one.
The risks of tooling
The risks specific to tooling are the benefits, continued past their proper stopping point. Automation blasting is the emblematic one: scale applied to outreach is precisely how a practice earns a spam reputation with the very journalists it exists to reach. The journalist who receives the tenth automated follow-up does not think worse of the software. Data decay presented as fact follows: the dashboard and the database both speak in the present tense about a world that has moved, and the practitioner who quotes their figures without checking is manufacturing accuracy. Over-fitting to what the dashboard counts bends the work itself toward the countable — announcements shaped to generate the metric rather than to serve the story. And dependence completes the list: workflows built entirely inside one vendor’s system, records without exports, a practice that would lose its memory the month the subscription lapses or the vendor is acquired. None of these requires the technology to be bad. All of them require only that it be used without a governor.
What technology does not change
End where the standing rule began. The journalist relationship does not change: it is still built from accuracy, usefulness and time, and no platform licenses a shortcut through it. The accuracy standard does not change: a false claim sent by an advanced system is a false claim, with better distribution. The ethics lines do not change — the categories this series sets out in its article on ethics in public relations, deception and manipulation and concealment among them, apply to tool-assisted work exactly as to hand-written work, and “the system sent it” will join “an intern sent it” on the long list of explanations that explain nothing. What technology changes is velocity: a practice’s virtues and its failures both arrive faster, at greater scale, on a more permanent record. That is the whole of the PR tech meaning worth keeping. The tools are excellent servants of a standard they cannot set — so the standard, as ever, is the practitioner’s job.
