Skip to main content
Mentions resolve against a single, hierarchical location taxonomy, the same schema whether it came from a news publisher, a TikTok post, or a Reddit thread.

​
The taxonomy

Location resolves across six levels, each built on an established standard so it interoperates with other systems you use.
LevelCoverage
RegionGlobal
SubregionGlobal
CountryGlobal
State / ProvinceUnited States only
CountyUnited States only
City / MunicipalityGlobal
Each level has exactly one parent, and each mention resolves to exactly one location. Selecting a location automatically includes everything beneath it: filtering on a state returns the cities inside it, and filtering on a country returns every state, county, and city within it. State and county are US-only. Mentions outside the US resolve to a country and, where the source is specific enough, a city.

​
Coverage by channel

Location resolution is best effort, and how often it succeeds depends heavily on the channel.
Because coverage varies by channel, a geographic breakdown of a mixed-channel workspace reflects where location was resolvable, not where the conversation actually happened. Read location distributions as directional, and be careful comparing across channels with different resolution rates.

​
How location is resolved

News. Location reflects the publisher’s headquarters, typically resolved to country level. This is a signal about the outlet, not about where the story took place. A Washington-headquartered outlet covering a flood in Texas resolves to the US, not to Texas. Social. Location comes from the account’s self-reported or detected profile location, coerced into the taxonomy. Free-text entries like “Chi-town” or “NYC” are matched against an alternate-names list and resolved to their canonical entry. Reddit. Reddit has no location field, so location is inferred from geographically named subreddits: a post in a city or country subreddit resolves to that place. Posts outside geographic subreddits carry no location. Unresolvable locations aren’t labeled. The platform does not guess. A mention with no resolvable location is absent from geographic breakdowns.

​
Time coverage

Normalized location is available on mentions published on or after July 1, 2026. Mentions published before that date carry no normalized location.

​
Use cases

Group analytics by country or state to see which markets are driving a spike, then drill to city level to find the specific concentration.
Filter a workspace to a region before pulling narratives and top accounts, so a briefing reflects the market that’s actually affected rather than global volume.
Break the same narrative down across countries to see whether framing differs by market, useful for global brands running region-specific messaging.
Separate home-market coverage from international pickup to understand whether a story has crossed borders.
Coverage that resolves to an unexpectedly narrow set of locations can be worth a closer look, though given the coverage caveats above, treat this as a prompt to investigate rather than a finding.

​
FAQs

Most social platforms only expose location when a user has filled in their profile, and many don’t. Twitter and Reddit resolve for under half of mentions for this reason.
Not currently. State and county are US-only at launch; other countries resolve to country and, where possible, city.
Yes. Selecting any level automatically includes everything beneath it.

API Changelog

Full technical detail on the location fields.

API Introduction

Getting started with the PeakMetrics API.

MCP Server

Querying your workspace conversationally, including by place.

Exporting Data

Taking workspace data into your own analysis.
Questions? Reach out to support@peakmetrics.com.