How to Find Where People Discuss a Brand: A Discovery Playbook

Brand Visibility · Playbook

Somewhere right now, someone is recommending a competitor over you in a thread, asking Reddit which tool to buy, or getting a one-line verdict on your product from ChatGPT — and you have no idea where. How to find where people discuss a brand is a discovery problem before it's a monitoring problem: you can't track or join a conversation you haven't located. This playbook gives you a repeatable way to surface the exact platforms, communities, and pages where people actually talk about you.

Brand mentions are any references to your company, products, or people across the web — posts, comments, threads, reviews, articles, and AI-generated answers — whether or not anyone tags or links you. Finding where those mentions happen is the step most teams skip: they jump straight to a dashboard and point it everywhere, instead of first learning which two or three rooms host the conversations that matter. Get the discovery right and everything downstream — alerts, sentiment, participation — points at the right place. Get it wrong and you monitor surfaces your buyers never visit.

How to find where people discuss a brand: the five-move method

The fastest way to find where people discuss a brand is to work outward from what you already know and let the search results hand you the venues. You're not trying to read every mention yet — you're mapping the rooms. Run these five moves in order and you'll have a ranked list of real discussion venues in an afternoon.

1. 01

   List your search terms — brand and category

   Write down your brand name and its misspellings, every product and feature name, your key people, and — critically — the problem your product solves in the words a customer would use. The category and problem terms are what surface the threads that never name you. This list is the seed for every search that follows.

2. 02

   Run name searches across the open web

   Search each term on Google, then again scoped to specific surfaces with site: operators (site:reddit.com, site:quora.com, your review platforms). You're looking for which pages already rank and which domains keep reappearing — those repeat domains are your candidate venues.

3. 03

   Follow threads back to their communities

   A single Reddit result isn't the find — the subreddit behind it is. Open the high-signal results and note the community, the review site, or the forum they live in. One good thread usually reveals a whole room where similar conversations happen regularly.

4. 04

   Check the AI answer layer

   Ask ChatGPT, Perplexity, and Gemini the buying-intent questions in your category and see whether you're named and which sources the answer leans on. The cited sources are themselves discovery leads — they point at the pages models treat as authoritative about your space.

5. 05

   Rank venues by signal, not vanity

   Score each venue you found by how often real, on-topic conversations appear and whether the audience matches your buyers. A handful of active rooms beats a long list of places with one stale mention. This ranked list is what you'll actually monitor.

That sequence — terms, name searches, community trace, AI layer, rank — is the whole discovery loop. Everything below is detail on how to run each move well and what to do with what you find.

Manual discovery: the free search recipes that surface venues

You can find most of your hot spots before paying for anything. Manual search is slow but it teaches you the shape of the conversation, which no dashboard does for you. The trick is to search deliberately, surface by surface, instead of typing your brand name into Google once and calling it done.

The reason to run the category and problem terms, not just your brand, is that a large share of discussion never names you. People say "I need a tool that does X" or describe the pain, and your brand only enters the thread later, if at all. Searching the problem is how you find the rooms where buyers gather before they know your name — which is exactly where you want to be present.

The places to check first: a starter map of discussion zones

People talk about brands in more places than any single tool covers, but the venues cluster into a few predictable zones. Use this as a starting checklist, then prune hard to the zones where your specific buyers actually gather — a B2B SaaS lives on Reddit and G2, while a consumer product lives on social and review apps.

You don't instrument all six. You find which two or three carry real conversation for your brand, and you go deep there. If you want a structured way to score coverage across these zones once you've found them, our brand visibility map playbook turns this loose list into a scored grid.

How to find brand mentions on Reddit

Reddit deserves its own pass because it's where the most candid opinions live and where discovery is hardest. Finding brand mentions on Reddit is tricky for one specific reason: a huge share of threads surface without ever tagging or naming the brand, so manual keyword searches and basic alerts miss discussions buried in smaller subreddits and comment chains. People describe the tool or the problem instead of typing your name.

The practical method has three layers. First, search Reddit directly for your brand and product names, then for the problem you solve, and note which subreddits the results cluster in. Second, run the same terms through Google scoped with site:reddit.com, which indexes older and buried threads Reddit's own search drops. Third, read the communities you keep landing in — the subreddit is the find, not the single thread. Our guide to analyzing Reddit audiences covers turning a list of subreddits into a real understanding of who's talking and why.

A note on what to do once you find these threads: read first, participate carefully. Reddit communities spot and punish promotional replies fast, so if you engage, do it as a genuine participant who adds something. The mechanics of useful, non-spammy presence are in how brands use Reddit and how Reddit affects GEO.

When a brand monitoring tool earns its keep

Manual discovery has a ceiling. Once you've proven that conversations are frequent, scattered across several platforms, or moving faster than you can check by hand, it's time to let software do the watching. A brand monitoring tool earns its cost when you need near-real-time alerts, sentiment at scale, or coverage of the untagged mentions a manual search will always miss. The category covers everything from free alerts to a full social listening brand monitoring tool that watches dozens of sources at once.

Names you'll encounter span the free and the paid: Google Alerts and F5Bot at the free end; broad social listening platforms like Brand24, Mention, Awario, Meltwater, and Brandwatch for tracking brand mentions across networks; and a growing set of AI visibility tools for the answer layer. One example is the brand mentions tool Brand24 offers, which scans social, news, blogs, forums, and reviews and distills the volume into sentiment and a word cloud. Pricing and free tiers change constantly, so check current vendor pricing rather than trusting a number from any roundup — and buy the narrowest tool that covers the surfaces you proved matter, not the one with the prettiest dashboard.

The discipline that makes a tool pay off — watch list, alert rules, routing, sentiment, cadence — is the subject of our how to track brand mentions playbook. Finding is the input; tracking is the system you build on top of it.

Confirm a venue is worth watching

Discovery produces a long list. The job now is to cut it to the few rooms that justify your attention, because monitoring everything is how teams end up reading nothing. Sample each candidate venue before you commit a tool or a calendar block to it.

Three quick checks settle most calls: recency (is anyone talking this quarter, or did the conversation die in 2023?), density (do searches return a steady trickle of relevant hits, or one lonely thread?), and fit (is this where your buyers gather, or just where your name happened to appear?). A surface that passes all three goes on the monitoring list. The rest you can re-check by hand once a quarter without guilt.

From finding to monitoring: the handoff

Finding where people discuss a brand is a one-time mapping exercise; keeping up with what people are saying is the ongoing system you hand off to alerts and a weekly review. The handoff is simple: take your ranked, validated venue list and assign each room a way to watch it — a free alert, a tool, or a recurring manual check — so no proven surface goes silently uncovered.

• Load your terms once, everywhere. Brand name and misspellings, product names, key people, competitors, and the problem keywords — the same seed list you built for discovery feeds your alerts and tools.
• Assign every proven venue an owner and a method. Each room from your validated list gets either a tool, a free alert, or a calendar block. If it's worth watching, it's worth a named way of watching it.
• Watch the trend, not the ping. A single mention is an anecdote; the direction over weeks — more conversation, shifting sentiment, a new venue heating up — is the signal. Re-run your discovery searches quarterly, because the rooms move.

That last point matters most: discovery is never truly finished. New subreddits form, review sites rise, and AI answers reshuffle which sources they cite within days. Treat your venue map as a living document. For the wider system this feeds — measuring presence across every surface — see what organic brand visibility is, and for turning raw counts into a competitive read, how to analyze brand share of voice.

Finding where people discuss a brand isn't about reading everything — it's about locating the few rooms that carry real conversation, confirming they're worth your time, and then pointing your attention there. Run the five-move method, search the problem and not just your name, validate each venue with recency, density, and fit, and hand the proven list off to a monitoring rhythm. Do that, and you'll catch the thread you'd otherwise find weeks too late.