A journalist calls for comment on a story you have not seen. A customer complaint has been shared in three Telegram channels since breakfast. Your CEO's interview ran yesterday, and nobody on the team knows how it landed.
Media monitoring exists so these moments stop being surprises. This is a beginner's guide: the words you will hear, how the software works, what to track, how to write your first query and how to pick a tool.
What is media monitoring?
Media monitoring is the habit, and the software, that tells you what is being said about you, where it is being said, and whether it matters. The topics are usually your brand, products, executives, competitors and industry.
Two things separate it from a quick search:
- It is continuous. The tool keeps collecting while you sleep, so you see the picture over time, not a snapshot.
- It is analyzed. Each mention gets labels, so you can count, compare and report.
You will also hear "social listening." The two overlap heavily. Monitoring tends to focus on finding and reporting mentions; listening leans toward patterns in audience conversation. Most modern tools do both, and we compare them in social listening vs media monitoring.
Words you will hear
Vendors use slightly different names for the same things. These ten terms cover most demos and dashboards.
- Mention
- One item that matches your query: an article, post, comment, video or review.
- Theme
- A saved topic you monitor, such as your brand or one competitor. Other tools call it a project, search or topic.
- Boolean query
- The search rule behind a theme, built from keywords with AND, OR, NOT, quotes and brackets.
- Exclusion
- A NOT rule that removes predictable false positives, such as a place or product that shares your name.
- Deduplication
- Grouping reprints and near-copies of one story, so fifty republished copies count as one story with its spread.
- Sentiment
- A positive, neutral or negative label on each mention, usually set by an AI model.
- Reach
- An estimate of how many people could have seen a mention, based on the source's audience.
- Share of voice
- Your mentions as a percentage of your mentions plus your competitors' mentions.
- Spike alert
- A notification when volume or negativity jumps above normal for a theme.
- OCR and speech-to-text
- Technology that reads text inside images and video frames, and transcribes spoken words, so mentions nobody typed can still be found.
Where mentions come from
A good setup covers every place your audience reads, talks and shares. In 2026 that is a much longer list than newspapers.
| Channel | What you find there | Why it matters |
|---|---|---|
| News sites | Articles, interviews, op-eds, press release pickups | Earned coverage and reputation |
| Blogs | Reviews, industry commentary, long-form opinion | Niche experts and early signals |
| Forums | Threads, questions, complaints | Honest customer views |
| Social networks | Posts, comments, short video | Volume, virality, influencers |
| Messengers | Public channels and groups, such as Telegram | Fast-moving news in many markets |
| Reviews | Ratings and written reviews on review sites and app stores | Product and service feedback |
| Images and video | Text in images, speech in videos and stories | Mentions nobody typed as text |
The last row is the one most teams miss. A brand name on a poster, said aloud in a video or shown in an Instagram story is a mention, and a text-only search will not see it. Our guide to image and video monitoring explains how tools catch these.
How it works, step by step
Every tool follows roughly the same pipeline. Knowing it helps you ask vendors sharper questions, because each step is a place where mentions can be gained or lost.
- Sources. It starts with a list of news sites, social accounts, forums, channels and review pages. Fit matters more than size: a million sources mean little if the regional outlets your audience reads are missing.
- Collection. Crawlers and platform connections fetch new content on a schedule. Some tools refresh within minutes; cheaper plans may refresh every few hours.
- Matching. Each new item is checked against your query. This is where most quality problems start: a loose query floods you with noise, a tight one misses real mentions.
- Deduplication. Republished and lightly edited copies are grouped, so you see one story with its spread instead of fifty identical cards, and reprints do not inflate your numbers.
- Enrichment. Each mention gets labels: language, sentiment, topic tags and source data such as author and estimated reach. Sentiment models trained on each language do much better than one English model applied to everything.
- Alerts and reports. Spike alerts flag jumps as they happen, a daily digest sums up the last day, and reports turn weeks of data into charts for a manager or client.
What to track
Start with five themes. Most teams need all five, though the weight differs by role. Keep each as its own theme: mixing them makes every number harder to read.
Your brand
Company and product names, slogans, campaign hashtags and official handles, plus common misspellings and transliterations.
Your executives
The CEO and spokespeople. Their quotes shape your reputation, and they are often targeted personally in a crisis.
Competitors
Two to five direct rivals, for benchmarks, share of voice and early warning on their launches.
Industry topics
Regulation, market trends and the debates in your sector. This is where you find stories to join and journalists to brief.
Crisis triggers
Words like recall, lawsuit, outage, fraud, strike or boycott next to your brand, with tighter alerts. See the crisis monitoring playbook.
Metrics that matter
You can measure dozens of things. These five answer most of the questions a manager or client will ask.
| Metric | What it tells you | How to use it |
|---|---|---|
| Volume | How many mentions you got in a period | Spot spikes and trends, compare campaigns |
| Reach | Estimated audience of the sources that mentioned you | Separate a small post from a front-page story |
| Sentiment | Share of positive, neutral and negative mentions | Track reputation, catch problems early |
| Share of voice | Your mentions as a percentage of you plus competitors | Benchmark visibility against rivals |
| Top sources and authors | Which outlets and people talk about you most | Find allies, critics and journalists to brief |
Read volume together with sentiment: a spike is good or bad news, and you only know which once you open the top stories. Treat reach as an estimate, because every vendor calculates it differently, so compare periods within one tool, not across tools.
Writing your first query
Your query decides what you see. Most tools support Boolean search: AND, OR, NOT, quotes for exact phrases and brackets for grouping. Syntax varies by vendor, but the logic is the same. Let's build one for an imaginary coffee chain called Nova Coffee.
1. List every name
Start with every way people refer to the brand:
- Full name and short forms:
"Nova Coffee" OR "NovaCoffee" OR "Nova Café" - Handles and hashtags:
@novacoffee OR #novacoffee - Product names unique enough to stand alone:
"Nova Cold Brew" - Common misspellings:
"Novva Coffee" OR "Nova Cofee"
2. Add context to short names
"Nova" alone matches astronomy, cars and a dozen other brands. Pair the short name with context words:
"Nova Coffee" OR (Nova AND (coffee OR café OR latte OR barista OR cafe))
3. Exclude the noise
Exclusions remove predictable false positives. Add them as you find them in your first week of results:
NOT ("Nova Scotia" OR supernova OR "Bossa Nova" OR "Chevrolet Nova")
4. Cover every language and script
Your audience does not all write in English. Add the name in each language and script your markets use:
- Turkish:
"Nova Kahve" - Russian (Cyrillic):
"Нова Кофе" OR "Нова кофе" - Arabic:
"نوفا كوفي"
5. Test against a week you know
Run each theme against a past week you remember well. If an article you know about is missing, find out why and adjust. Review exclusions monthly, because new noise appears as your brand grows.
Choosing a tool
Prices and features vary widely, so judge every tool on your own data during a trial. Tick these off before you sign:
- Coverage in your markets. It finds the regional outlets, messengers and networks your audience uses, in their languages.
- Images and video. It catches your name inside images, video frames and spoken audio, not only in captions.
- Sentiment per language. It uses models trained on your languages, and your 20-mention spot check agrees.
- Speed. You know how quickly new mentions appear, and whether the refresh rate depends on your plan.
- Query control. You can write full Boolean queries with exclusions, not only simple keywords.
- Deduplication. Reprints are grouped and do not inflate your counts.
- Alerts. You get spike alerts and a daily digest by email or the messenger your team uses.
- Reports. You can rebuild your monthly report and export it in the formats your managers expect.
- Limits. You know the topic, mention and user limits, and what happens at the cap: upgrade, top-up or overage bill.
- Contract terms. You know the minimum term, the renewal notice period and whether prices are published or quote-only.
Comparing the big suites? Our guides to Meltwater alternatives and Cision alternatives put features and published prices side by side.
How Scano does it
Scano is an AI media monitoring and social listening platform. It follows the pipeline above, with a focus on markets and formats that English-first tools cover thinly.
- Sources: 1M+ news sites, social networks, messengers, blogs, forums and review sources in 20+ languages. Coverage is strongest in Türkiye, Central Asia, the Caucasus and the Gulf, and includes Telegram channels and VK alongside Instagram, TikTok and YouTube.
- Images and video: OCR reads text in images and video frames; speech-to-text transcribes what people say in videos and stories, including Telegram video notes. See Visual Insights.
- Sentiment: dedicated AI models for English, Turkish, Arabic, Russian, Kazakh, Persian, Hebrew and French.
- Smart tags classify mentions by topic automatically.
- AI Analyst: ask questions about your mentions in plain language; answers link back to the source posts.
- Alerts: a daily digest each morning and spike signals when volume or negativity jumps, sent to email and Telegram.
- Reports: 13 analytics reports, theme comparison and share of voice, with exports to Excel, Word and PDF and scheduled reports on your template.
- Unlimited users on every plan, plus iOS and Android apps.
Prices are public on the pricing page. Starter costs $490 a month, or $417 a month billed annually, for 5 themes and 10,000 mentions a month. Mention caps are hard limits: you top up or upgrade, so there are no surprise overage bills.
Want to see what monitoring output looks like? The monthly Media Pulse reports are built from Scano data.
Conclusion
Media monitoring is simple to describe and easy to do badly. The difference comes from covering the channels your audience uses, writing queries that catch real mentions without the noise, and reporting a few metrics consistently.
Start small: one theme each for your brand, executives, competitors, industry and crisis triggers. Test against a week you know, then tune every month.
Book a Scano demo and we'll set up a trial workspace on your own brand after a short demo, so you can see what your current setup is missing.