Track Brand Mentions and Sentiment
By the end of this recipe you'll have an automation that searches the web every morning for people talking about your brand, tags each mention as positive, neutral, or negative, logs it to a board, pings you the moment something negative shows up, and posts a plain-language read on the day's mood to your team. Reputation moves fast, and the difference between a small issue and a public one is often how quickly you noticed — so this pipeline turns scattered, manual "did anyone mention us?" checking into a standing watch that runs on its own.
Readiness: Available now (built as one Automation, plus an optional second one for a weekly trend report). The version where a single named teammate owns the watch and posts the mood on its own schedule is rolling out in early access.
What it does
Every time it runs — say, each weekday at 8 AM — the automation does five things in a row:
- Searches the web for recent mentions of your brand, products, and key people using an AI web-research step.
- Pulls each mention apart into a tidy record: where it appeared, the link, the exact snippet, and when.
- Reads the tone of every mention and tags it Positive, Neutral, or Negative.
- Logs each one as a row on a Brand Mentions board — your searchable, filterable archive of everything said about you.
- Reports the mood by summarizing the batch into a short read and posting it to your team channel — and if any single mention is negative, it messages you right away instead of waiting for the digest.
The board becomes the memory (every mention, every tone, forever), and the daily message becomes the pulse. Because sentiment is stored per row, you can slice it any way you like later: "show me every Negative mention from last month," or "which topic gets the worst reception?"
Build it yourself
One-time setup
Make a board with a table for mentions. Add columns for Source (where it appeared — a publication, a social platform, a forum), Link (URL), Snippet (the quoted text, as rich text), Sentiment (a single-select with options Positive, Neutral, Negative), Topic (a select like Product, Support, Pricing, PR), and Date. This board is both your live queue and your long-term archive.
List what the watch should look for: your brand name and common misspellings, product names, key executives, and your main tagline or handle. You'll paste these into the research step. Specific terms find better mentions than a single vague one.
The daily mention scanner
This is the core automation.
Add a Schedule trigger set to when you want the sweep — for example, every weekday at 8:00 AM. Run it more often (a few times a day, or hourly) if your brand moves fast and you want negatives surfaced sooner.
Add an AI → Web research step and give it your search terms. It gathers current, cited material from across the web — news, blogs, forums, reviews, and indexed social posts — and returns what it found.
Add an AI → Extract step to turn that research into a clean, structured list — one entry per mention, each with its source, link, snippet, and date. Extract pulls those fields out of the free-form research text so the later steps have something tidy to work with.
Add a For each step over that list so the following steps run once per mention found.
Inside the loop, add an AI → Sentiment step on the mention's snippet. It reads the tone and returns Positive, Neutral, or Negative. Add an AI → Classify step too if you want each mention bucketed by Topic (Product, Support, Pricing, PR) at the same time.
Add a Create row action on the Brand Mentions table. Map the source, link, snippet, and date into their columns, set Sentiment to the score you just got, and set Topic if you classified it. Every mention now has a permanent, filterable home.
Still inside the loop, add a Condition that compares Sentiment — continue only when it equals Negative. On the true branch, add a Send direct message (to you) or a Post channel message (to a dedicated #brand-watch channel) with the snippet and link. This is what turns a log into an early-warning system — you hear about the bad ones within one cycle, not at the weekly review.
After the loop, add an AI → Summarize step over the batch to write a short read on the day — how many mentions, the overall lean, and anything notable. Then add a Post channel message action to drop that summary into your marketing or comms channel. That daily message is your team's pulse on how the brand is landing.
Prefer email over a channel post for the daily read? Swap the final step for a Send or draft email action and send the mood summary to your comms list. Use draft mode if you'd like a human to glance at the wording before it goes out to a wider audience.
Optional: a weekly trend report
The daily scanner already reports each day. To also get a "how are we trending?" view, add a second, weekly automation that reads back the board.
Add a Schedule trigger — for example, Friday at 4:00 PM.
Add a Find rows action on the Brand Mentions table, filtered to rows with a Date in the last seven days.
Add an AI → Summarize step over those rows: the sentiment split (how many Positive vs. Negative), which topics drove each, and how the mood shifted from last week.
Add a Post channel message action for the team, or a Create document action to keep a running weekly log, or a Send or draft email to leadership. Pick wherever your team already looks.
Or just ask Copera AI
You don't have to assemble the graph by hand. Open Ask Copera AI and describe the outcome:
"Create a Brand Mentions board with columns for source, link, snippet, sentiment, topic, and date. Then build an automation that every weekday at 8 AM searches the web for mentions of Acme and our products, tags each mention's sentiment, logs it as a row, sends me a DM whenever a mention is negative, and posts a summary of the day's overall mood to the Marketing channel."
Ask Copera AI can create the board and its columns, then plan, draft, and assemble the automation for you. With the default Ask before acting permission mode, it shows a review card for each change before anything is created — so you approve the board and the automation before they exist. See What Copera AI Can Do for the full list of what it can build.
Make it a standing teammate
Standing AI teammates are rolling out gradually. When one is available in your workspace, you can hand this whole watch to a single named teammate — a "Brand Analyst" — that runs the sweep on its own schedule, logs each mention to the board, and posts the day's mood to your channel under its own identity. If you don't see teammates in your workspace yet, they're on the way — everything in this guide can be built today as an Automation.
Tips
Feed it good search terms. Web research finds what you point it at. List your brand name and common misspellings, product names, key people, and your handle or tagline — vague terms return vague mentions. Revisit the list whenever you launch something new.
Route negatives to the right person, fast. The escalation branch is the most valuable part of this recipe. Point the negative-mention DM at whoever owns reputation — your PR lead, community manager, or founder — so a brewing issue reaches a human within one scan cycle.
Go deeper than the open web when you need to. The web-research step covers what search engines index. For the full stream of a specific platform, add an authenticated HTTP action that calls a social-listening API you already use, then run its results through the same sentiment and logging steps. See Connectors for how external calls work.
Set a spend cap on the scanner. Because it researches and scores in a loop, give the automation a per-automation spend cap so a noisy news day can never run past a budget you choose. The usage dashboard shows exactly where credits go.
Frequently Asked Questions
Can Copera track brand mentions and sentiment automatically?
Yes. A scheduled automation runs an AI web-research step over your brand terms, extracts each mention, tags it Positive, Neutral, or Negative with an AI sentiment step, and logs it as a row on a Brand Mentions board. It then summarizes the day's mood to your team channel, and can DM you the instant a negative mention appears.
How does Copera detect the sentiment of a mention?
An AI sentiment step reads each mention's text and classifies its tone as Positive, Neutral, or Negative. The score is written to a Sentiment column on the row, so you can filter, sort, and report on how people feel about your brand over time.
Can it alert me when someone posts something negative?
Yes. Add a condition that checks whether a mention's sentiment is Negative; on the true branch, send yourself a direct message or an in-app notification, or post to a dedicated channel. Negative mentions reach you within one scan cycle instead of at the weekly review.
Does the tracker monitor social media?
The web-research step searches the public web, which includes many social posts, forums, reviews, and news articles that search engines index. For the full firehose of a specific platform, add an authenticated HTTP action that pulls from a social-listening API you already use, then feed those results through the same sentiment and logging steps.
How often does Copera check for new mentions?
On whatever schedule you set with the Schedule trigger — every weekday morning, a few times a day, or hourly. Each run searches for recent mentions, so your Brand Mentions board and mood summary stay current without anyone checking manually.
Related
- Monitor Competitors and the Market — the flip side of this watch: track rivals and market moves instead of yourself
- Automations — the full trigger, condition, action, and AI-step reference
- What Copera AI Can Do — everything the assistant can build and act on
- Deep Research — run a deeper, cited investigation into a mention or a reputation question on demand
- Connectors — connect external apps and APIs to your automations
- What You Can Build with Copera AI — the full playbook