Skip to main content
Explore gives you a fast, analytical read on any topic in seconds. Type a Boolean query, run it, and PeakMetrics instantly returns a dashboard showing how big the conversation is, where it’s happening, and who’s driving it, all without creating a Workspace. Workspaces are built for sustained, ongoing monitoring. But many questions don’t need that commitment. When you’ve just spotted a potential issue, been asked about something you weren’t tracking, or want to size up a topic before building it out, Explore lets you answer the question right now and decide in minutes whether it deserves ongoing monitoring.
Running a query in Explore doesn’t create a Workspace and doesn’t count against your quota. You only commit when you choose to save a query as a Workspace.

When to use Explore vs. a Workspace

A good rule of thumb: Explore first, then save as a Workspace if it matters.

Getting to Explore

From the Workspaces homepage, find the Explore a Topic card at the top of the page. Enter your query in the search bar and click Explore. Not sure where to start? Click one of the Try an example queries below the search bar to see Explore in action. Inline Image (1)

Writing an Explore query

Explore uses a streamlined version of PeakMetrics Boolean syntax. The following operators are supported:
Explore supports a reduced field set compared to Workspace queries. Advanced fields and operators from the Advanced Guide to Boolean Searches are available once you save a query as a Workspace.
You can run up to 5 queries per minute. If you hit that limit, you’ll see a message asking you to wait a minute before trying again.

Reading the Explore dashboard

Once your query runs, the dashboard loads with your total mention count in the Filters panel and a set of analytics widgets.
An AI-generated summary of the conversation, grounded in a sample of real mentions that matched your query. The sample is drawn proportionally from each channel based on its share of total volume, so the summary reflects where the conversation is actually happening. Linked sources let you jump straight to the underlying coverage. Click Show more to expand the full summary and key takeaways.
A line chart of daily mention volume across your selected date range. Use the Total / By Channel toggle to see the overall trend or how each channel contributes. Spikes are your cue to dig in: they usually line up with a specific story, incident, or viral post.
Ranks each channel by mention count and share of total volume, so you can see at a glance whether a conversation is news-driven, social-driven, or both.
A donut chart showing the English vs. non-English split across all channels. Drill in for a full per-language breakdown. X (Twitter) is excluded from the detailed breakdown, since only the English/non-English split is available for that channel.
Ranks the websites and platforms publishing the most mentions. Click into a news domain for additional publisher-level detail.
News only. A ranked list of the journalists who’ve written the most about your topic, along with their outlet. Useful for identifying who’s covering a story and who to watch or engage.
Social channels only. Breaks down volume by type of post:
  • X (Twitter): original posts, reposts, replies, and quote posts
  • Reddit: posts and comments
A conversation dominated by reposts reads very differently from one driven by original posts.
A short sample of individual mentions that match your query, giving you a qualitative feel for what’s being captured. It’s a quick gut-check that your query is pulling in the right content, not a full mention feed.

Filtering and drilling in

Use the Filters panel on the left to narrow your results:
  • Date: Explore covers up to the trailing 12 months and defaults to the past 90 days. Set a custom start and end date to focus on a specific period.
  • Channel: Toggle channels on or off.
  • Domain: Limit results to specific websites.
  • Language: Limit results to specific languages.
After changing filters, click Apply Filters to update the dashboard. Applied filters appear as chips at the top of the panel. You can also click a value directly in a widget, for example a channel in Channel Distribution, a language, or a domain, to filter the whole dashboard to that value. This works the same way as filtering on a Workspace dashboard.

Data sources and coverage

Explore searches the trailing 12 months of published content across the following channels: Explore only shows channels your account already has access to elsewhere in PeakMetrics.
What does the asterisk (*) mean? For sampled channels (Reddit, VK, and Weibo), PeakMetrics extrapolates metrics from a random sample up to 100%, so the figures represent an estimate of the full conversation on that channel. These values are marked with an asterisk in the dashboard.

Saving a query as a Workspace

If your Explore results show a conversation worth tracking, turn it into a Workspace with one click.
1

Click Save as Workspace

Find the Save as Workspace button in the top right of the Explore dashboard.
2

Review your results

You’ll see a preview of matching mentions so you can confirm the query is capturing the right content before you commit.
3

Create the Workspace

The new Workspace is seeded with the same query, date range, and channels you used in Explore. The mention count you saw in Explore reflects the quota impact of saving it.
Creating Workspaces requires permission on your account. If the Save as Workspace button is disabled, contact support@peakmetrics.com or your Customer Experience Manager and we’ll create the Workspace for you.
Once saved, your topic gets the full Workspace toolkit, including the mention feed, Landscapes, Smart Categories, Reach, Bot Detection, and reporting.

Common workflows

You’ve just seen a concerning post or headline. Run it through Explore to see how large and widespread the conversation already is. Check Volume Over Time for momentum and Channel Distribution for spread. If it’s growing or crossing from social into news, save it as a Workspace and start monitoring.
When someone asks “how much are people talking about X?”, Explore gives you volume, trend, top outlets, and an AI summary in seconds, enough for a quick, data-backed answer without spinning up a new Workspace.
Use Explore as a sandbox to refine your Boolean query. Check the Mention Preview and Domains widget to confirm you’re capturing the right content, adjust the query, and re-run until it’s right. Then save it as a Workspace with confidence.
Use the Top Journalists widget to see who’s writing about a topic and which outlets they write for. Pair it with the Domains widget to understand which publications are leading coverage.
Run a quick query on a competitor, product launch, or industry trend to benchmark conversation volume and channel mix before deciding whether to add it to your ongoing monitoring.

FAQs

No. Explore queries are free to run and don’t create a Workspace. Quota is only affected if you save a query as a Workspace.
Up to the trailing 12 months. Results default to the past 90 days, and you can expand the date range in the Filters panel.
Explore is designed for fast, aggregate analytics. AI classifications like favorability, enrichments like Reach and Bot Detection, and Landscapes are applied once a query is saved as a Workspace and its data is processed.
No. Explore shows a small preview of matching mentions to help you gauge relevance. To browse, filter, and export the full set of mentions, save your query as a Workspace.
Instagram, TikTok, Facebook, LinkedIn, Telegram, YouTube, Threads, Bluesky, and Podcasts are not currently available in Explore. They remain available in Workspaces, depending on your account.
Reddit, VK, and Weibo figures are estimates extrapolated from a random sample of each platform’s content. See Data sources and coverage for details.
Explore supports a streamlined syntax: AND, OR, NOT, parentheses, exact phrases, and the title:, author:, and domain: fields. The full Workspace syntax is available once you save as a Workspace.
Explore allows up to 5 queries per minute. Wait one minute and run your query again.

Questions or feedback on Explore? Reach out to your Customer Experience Manager or support@peakmetrics.com.