Location search continues to play a crucial role in social media newsgathering. When news breaks, journalists cannot rely on eye-witnesses to use terminology or hashtags which are consistent or easy to search – often making location the only reliable way to identify eye-witnesses.
Despite the promise of GEO-tagged UGC, it can often be unfruitful and frustrating to search. There are multiple reasons for this, one of which being that location searches rely on the uploader enabling location sharing features when uploading their content. One of the bigger reasons, however, is the growing disconnect with how platforms (ours included) have represented location based searching. Most location search tools want users to define a radius or draw a geographical area thinking it will pick up exact locations of content being shared in that area.
The problem is that this is not how uploaders think of sharing location data – nor is it how social networks encourage their users to share sensitive location information. For some time now, Twitter, Instagram, and most social networks guide their users to ‘tag’ or add a specific place or city to their posts instead of having them post exact latitude/longitude.
Instagram, while having access to precise EXIF data (which includes very accurate location information), only uses the data to a perform place location lookup. It then asks its users to tag a place and proceeds to exclude any precise location data once the upload has been completed.
Uploaders have never been very comfortable sharing precise location data with the world – tagging locations, on the other hand, adds just enough ambiguity for users to feel more comfortable. The overall amount of location data shared via social networks is on the rise, so why does it seem like less uploaders are sharing location data? Mainly because drawing a large GEO-fence or search radius pulls in too many ‘places’ and not precise coordinates the way a large radius would suggest. To put it another way, a search like the one below will pull in some GEO-tagged results, but it will consist mainly of scattered results from ‘places’ across the city.
In the case of searching for GEO-tagged UGC during the Paris attacks, searching the entirety of Paris returned mainly noise. The most reliable way to search for UGC during the attacks was, to no one’s surprise, to think like the uploader and search for ‘places’ that would be tagged – Bataclan, Stade de France, Le Petit Cambodge, and so on. In essence we need to think of location search less like actual coordinate based search and more like the traditional hashtag search – but with a hashtag that is much more specific to the location that you are interested in. Unlike the hashtag, tagged ‘places’ are part of a structured and guided tagging system based largely on shared location databases like Foursquare and Google. All of this is great news for newsrooms as we effectively have a more universal tagging structure across disparate social platforms.
What is SAM doing about all this?
Today we’re pleased to announce our redefined approach to social media location search. SAM users will see several changes to the way our tool enables you to find Places (airports, malls, landmarks, stadiums, etc.) and Regions (cities, neighbourhoods, areas, etc.) and more precisely track content being shared from those locations. Here is the full breakdown of new functionality:
Autocomplete: It’s hard to get accurate place information on the spot, especially when details around the emerging story are sparse. Too often journalists have to open Google Maps to find a location then go back to SAM and paste in the exact latitude and longitude. Those days are gone! Now SAM will guide you with autocompleted suggestions of Places and Regions. Keep an eye on key locations within seconds.
Places & Regions: SAM will now power two distinct types of location searches: Places and Regions. Places are designed to track granular spots like airports, malls, buildings, universities, landmarks, and the like. We anticipate that tracking Places will be your new favourite way to search for UGC, as the granularity and specificity cuts out noise and puts you in the mindset of the uploader. Regions, on the other hand, are larger city level searches tailored for tracking content that might not be centered around a specific point (think weather stories). To help users better distinguish between Places and Regions we’ve set them to have flag and pin icons, respectively.
Clusters: Because most social media location data is centred around specific Places, clustering results is important to show which Places have the most content. SAM will now show you clusters for more accurate representation of where content is being shared.
Smart Grouping: What happens if the uploader shares content right beside a Place but not the specific Place you searched for? What happens if a user tags a slightly different Place that should technically be the same (for example, popular Places like an airport might have multiple sub-Places or different variations of the Place)? No problem! SAM has you covered with our Smart Grouping so you don’t miss any precious content.
Freeform Targeting: Ok, so you’re searching for a neighborhood fire; a Region is too big and there is no specific Place to search for. Don’t stress! SAM still lets you pan the map around and target a location search anywhere in the world. To perform a freeform target search simply drag the map anywhere using the crosshair, then click the ‘target search here’ button to search that location.
Unlimited Location Searches: Ok, this one isn’t new. You’ve always been able to run as many location searches as you want, but it’s pretty darn handy to be able to track multiple locations in a single workspace.
All of these features are now live for all SAM users.