Boolean search for media monitoring is the practice of writing a query out of logical operators (AND, OR, NOT, quotes, parentheses, wildcards and proximity) so a monitoring platform returns the media coverage you care about and leaves out the coverage you do not. It is the highest-return half hour a communications team can spend, because every alert, every dashboard and every number in the monthly report starts as a Boolean query.
Get it right and your team reads twenty relevant articles a day. Get it wrong and you either drown in false positives or, worse, you never see the story that mattered. Most PR teams inherit their queries from whoever configured the platform, and nobody has read them since.
This is a working guide for public relations and communications practitioners. Operators first, then Boolean strings you can adapt, then the failure modes we see most often when we audit a client's media monitoring queries.
What is Boolean search for media monitoring?
Boolean search is a way of combining search terms with logic rather than guesswork. The logic comes from George Boole, the English mathematician who set it out in An Investigation of the Laws of Thought in 1854, and the same three operators he described (AND, OR, NOT) still carry most of the weight in a modern monitoring query.
In a media monitoring context, the Boolean query does two jobs at once. It defines what counts as a mention of your brand, and it defines what counts as noise. Both halves matter. A search query assembled from a few obvious keywords and nothing else will bury you; one that over-excludes will quietly hide media coverage from your reports, and nobody will know until a stakeholder forwards an article you missed.
That is why we treat query craft as a communications measurement discipline, not an admin task. AMEC's Barcelona Principles 3.0 puts integrity and transparency at the heart of communication measurement (Principle 7: "Communication measurement and evaluation are rooted in integrity and transparency to drive learning and insights"). A Boolean search string is where that integrity either exists or does not. If you cannot explain why a mention is in your share of voice figure, you cannot defend the figure.
The Boolean operators that do the real work
Most monitoring platforms support the same core set. Syntax varies slightly, so check your platform's documentation before pasting anything in.
Wildcards and proximity are what separate a working media monitoring query from a keyword list. A wildcard extends one keyword into all of its endings; a proximity operator turns two keywords into a relationship. Three things about this table are worth more attention than they usually get.
The first is operator precedence. Most platforms evaluate NOT before AND, and AND before OR. So Summit OR mortgage AND Sydney does not mean what you probably think it means. Bracket everything. A parenthesis costs nothing, and parentheses remove an entire category of silent error.
The second is proximity. AND tells you two terms appear somewhere in the same article, which in a 2,000-word news feature is a weak signal. “...”~10 tells you they appear within ten words of each other, which is usually the difference between a story about your organisation and a story that mentions it in a list. Proximity is the operator that most comms teams underuse and the one that most improves precision.
The third is case. Search terms are almost never case-sensitive, but the operators usually are, so and gets read as an ordinary word while AND gets read as logic. Acronyms are where this bites hardest. A three-letter acronym will happily match surnames, place names and mid-sentence noise, so give it a case-sensitive operator where your platform offers one and keep it inside a context condition rather than letting it run loose in a search query.
Boolean search strings you can adapt
These are patterns, not copy-paste-and-forget. Swap "Summit" for your own brand and adjust each Boolean string to your platform's syntax. We use an invented company here so the logic stays visible, and we have written these for news, broadcast and print rather than for social listening.
1. Core brand mentions, with the obvious homonyms excluded
("Summit" OR "Summit Group" OR "SummitGroup" OR "@SummitGroup")
NOT ("climate summit" OR "leaders' summit" OR "summit talks" OR "summit meeting" OR "G20 summit")
2. Spokesperson coverage & Broadcast Media transcription variation
("Jane Doe" OR "Doe, Jane" OR "J. Doe" OR "Jayne Doe")
NEAR/10 ("Summit" OR "chief executive" OR "CEO" OR "managing director")
3. Issue and crisis watch, tuned for early signal
("Summit" OR "Summit Group")
NEAR/15 (outage OR breach OR recall OR "class action" OR investigation OR "under fire" OR apolog*)
NOT (hire* OR "sponsored content" OR advertorial)
4. Campaign tracking
("Summit" OR "Summit Group")
AND ("Better Banking" OR "#BetterBanking")
AND (launch* OR campaign OR announc* OR partnership)
5. Category share of voice
("Summit" OR "Rival Co" OR "Third Bank")
AND ("interest rate" OR "cash rate" OR mortgage OR "home loan")
6. Policy and regulatory monitoring
("Summit" OR "Summit Group")
NEAR/20 (ASIC OR APRA OR "Senate inquiry" OR "draft legislation" OR submission OR "regulatory review")
Keep each query doing one job. A single query that tries to cover brand, spokespeople, campaigns and crisis at once cannot be tuned, cannot be reported on cleanly, and cannot be handed to a colleague.
Why Boolean search breaks differently across news, broadcast and print
Almost every guide to Boolean search is written for social media monitoring. That is a different problem. Social text is short, tagged, and written by the person you are tracking. News, broadcast and print media coverage is long, edited, and written about your brand by someone else. The social listening playbook does not transfer cleanly, and three consequences follow.
Broadcast is transcript text, so spelling variants are not optional. Radio and television monitoring runs on speech-to-text, and accuracy drops in exactly the conditions broadcast news creates. Published benchmarks put transcription accuracy at roughly 70 to 85 percent in noisy environments and 75 to 90 percent for heavily accented speech (AssemblyAI). If your organisation has an unusual name, the transcript will spell it several ways, and a query that only matches the correct spelling will under-report your broadcast coverage. This is a good reason to review how radio news is tracked in a monitoring platform before you assume the gap is a coverage gap rather than a query gap.
Print is scanned text, so the same tolerance applies. Optical character recognition misreads characters that speech-to-text never would, so print mentions go missing for reasons no tidy keyword list anticipates. Wildcards and a short list of plausible misreadings do more for print recall than any other change.
Online news repeats itself at scale. Syndication means one wire story can appear across dozens of Australian mastheads under slightly different headlines. Boolean logic alone will not deduplicate that, but title: and proximity operators will help you separate the original from the reprints when you need to report on originating coverage.
Channel by channel, the sensible pattern is one query family per channel rather than one query for everything. It is more work up front and considerably less work every month afterwards. Our view on how the channel mix is shifting is set out in media monitoring in 2026.
Seven mistakes that wreck a media monitoring query
Skipping the parentheses. Covered above, and still the most common fault we find. If a query has more than two operators and no brackets, assume it is wrong.
Leaving a phrase out of quotes. Without quotes, cash rate is read as cash AND rate and matches any article containing both words anywhere in the text. Anything two words or longer belongs in quotes, and that includes your own trading name.
Excluding a word that also appears in coverage you want. Excluding "sport" to filter out sponsorship noise will also drop the business story that mentions your sports partnership. Test every exclusion against last month's coverage before you commit it.
Ignoring name variants. Legal name, trading name, acronyms, social handle, common misspellings, transcript errors, and the way journalists actually refer to your brand in second reference. All of them belong in the OR block, and the acronyms belong there behind a case-sensitive guard.
Wildcards that are too greedy. A wildcard on a short stem does more harm than good: comm* will match community, commercial, commuter and commit. Use at least four or five characters before the asterisk.
Treating false positives as free. They are not. A monitoring stream running at a high false positive rate trains the PR team to stop reading alerts, which is the failure mode that lets a real issue through. Precision is a workflow control, not a tidiness preference.
Setting the query once and never returning to it. Product names change, spokespeople change, competitors launch, and news cycles create new homonyms. A quarterly review of every active Boolean string is a reasonable minimum, and a review after any crisis is mandatory.
Does Boolean search still matter now that monitoring platforms use AI?
Yes, and the reason is auditability.
AI in media monitoring is not one thing, so it helps to be specific about which part is doing what. Modern platforms use semantic and vector search to catch mentions that share no keywords with your query, entity resolution to tell your brand apart from a similarly named one, and language models to classify topic and sentiment at volume. These techniques find coverage that Boolean logic misses, and we build them into the platform because recall matters.
What they do not give you is an explanation. You cannot show a CFO an embedding and defend the number that came out of it, and sentiment scores in particular invite the question "how do you know". A Boolean search string, by contrast, is readable, testable and reproducible. You can hand it to an agency, a colleague or an auditor and they can tell you exactly why a piece of coverage was counted.
So the practical division of labour looks like this. Boolean defines the boundary of what you are measuring and keeps the measurement defensible. The AI layer widens recall and does the sorting inside that boundary. PR teams that run both get better numbers than teams that pick one, which is roughly the argument we made about what changes and what does not when AI enters the monitoring workflow.
If you are still choosing a platform, the operator set it supports (particularly proximity, field operators and case sensitivity) is a legitimate evaluation criterion. Our rundown of what to look for in advanced media monitoring tools covers the rest.
Frequently asked questions
What is Boolean search for media monitoring?
Boolean search for media monitoring is a query written with logical operators (AND, OR, NOT, quotation marks, brackets, wildcards and proximity) that tells a monitoring platform which mentions to capture and which to exclude. It defines the boundary of what your organisation counts as coverage of its brand.
Which Boolean operators do media monitoring platforms support?
Nearly all support AND, OR, NOT, exact-phrase quotes and parentheses. Most also support wildcards and proximity operators such as NEAR/n, and many add field operators such as title:, a source filter, or a case-sensitive match for acronyms. Syntax differs between platforms, so check the documentation.
How do I stop my media monitoring query returning irrelevant coverage?
Put every phrase in quotes, bracket your logic, replace weak AND conditions with proximity operators, and build an exclusion block from the actual false positives in last month's media coverage rather than from guesses. Test each change against a fixed historical window so you can see what it adds and what it drops.
Is Boolean search case-sensitive?
In most media monitoring platforms the search terms are not case-sensitive, but the operators are: AND, OR and NOT have to be capitalised or they get treated as ordinary words. Some platforms add a case-sensitive operator, often matchcase:, which is worth using for acronyms and initialisms.
Does Boolean search still matter if my platform uses AI?
Yes. AI features such as semantic search and machine classification improve recall, but a Boolean search string is what makes your measurement auditable. It is the part of the process you can explain to a stakeholder, which matters when the numbers reach the board.
How often should we review our Boolean search strings?
Quarterly as a baseline for most public relations teams, and immediately after any crisis, rebrand, product launch or change of spokesperson. Anything that changes how journalists refer to your organisation changes what your query should match.
Where to take this next
A good Boolean search string is worth more to a PR team than another dashboard. It decides what the team reads, what the report says, and whether anyone can defend the numbers when they are questioned. Write it deliberately, document it, and revisit it on a schedule.
If you want a second pair of eyes on the Boolean strings you are running now, see how Truescope monitors broadcast, print, online and social media coverage for PR and communications teams. Our analysts build and tune these queries with clients across Australia and New Zealand every week, and would be happy to help with whatever your team may need to make your queries more efficient.
See how you can get more signals and insights from your media intelligence platform
See how you can get more signals and insights from your media intelligence platform
.svg)
