Hyperlocal Social Ads Need to Be ‘Current, Relevant, and Contextual’

Share this:

twitter-profile-imageRumors swirled in June that the Twitter was developing a location-based advertising product. The reports, which came days after the company snapped up a local discovery service in Spindle, led many to wonder whether the micro-blogging service was about to make a big move in local. With users pushing out over 450 million tweets a day, Twitter is one of the few platforms that has aggregated enough demand to make hyperlocal targeting work at scale for larger brands.

Enter Topsy, one of a handful data analytics startups that are helping brands make sense of the deluge of data flowing through social networks. As a “Certified Product Partner,” Topsy has unlimited access to the full stream of Tweets sent through the platform.  In June, the company bulked up its own geo-targeting capabilities, allowing marketers to identify inferred location from the content of a tweet — say, a user tweets, “headed to Philly!” — in addition to the much smaller percentage of tweets, which included a user’s exact latitude-longitude.

Street Fight recently caught up with Jamie de Guerre, the company’s VP of product, to talk about the intersection social media and location, and how marketers can leverage location data to make social media marketing work.

How “local” is Twitter?
The rate of geo-tagging tends to be very low, around 1-2%. Today 7 million [geotagged] tweets are sent a day, but the total number a day is upwards of 450 million. I think that the trend will grow overtime, but for certain types of tweets. Mobile is increasingly popular on Twitter, so as a larger percentage of tweets are coming from mobile devices, so naturally more of them will become geo-tagged.

Also, we’ve found that when people are at a particular event and want to share that they’re at it, then people are much more apt to include their location. They’re excited to give an update and they’ll include their location since they’re announcing it.

In June, Topsy rolled out functionality that “infers” the location of tweets. How accurate is that data and what’s the value for marketers?
There are definitely a lot of scenarios when brands would want to access location data for a broader set of tweets. And there are lots of other pieces of information that may be an indirect indication of someone’s location. We look at a number of signals: we look at the user’s profile, words in the actual tweet, language, what URLs are being shared, and local website history. We can then use that information to infer and track locations for past tweets.

With all this, Topsy has accurate location data at the country level for about 95% of all tweets. State and province locations can be found for over 50% of tweets, county level locations in the U.S. can be found for over 30%, and city locations (on a global level) can be found for over 25%.

As mobile advertising ramps, more brands are bringing location data into their campaigns. What opportunities exist on Twitter today?
That’s the same kind of scenario that we’re constantly working with, brands and ad agencies, big — and smaller, more local ones, trying to do that type of hyperlocal marketing. However, it’s very difficult to know what people are tweeting in those areas. Even for a large, national brand that operates on a local level, like a retailer, it becomes important to be able to understand the conversation in every local area that you’re marketing in. When you’re doing advertising on social media, it’s very important that you’re current, relevant, and contextual to something that’s going on at that time for your consumer.

With Topsy Pro, we have large brands watching certain locations for any spikes in the conversation is spiking for their topic of interest, and respond to alerts to create a local promotion or campaign related to it. And the same can be done for small advertisers as well.

Have you seen any major differences in social media campaigns that engage users based on location and those that do not?
In working with brands with their social campaigns and advertising, we’ve definitely seen that being contextual to something relevant and local provides for much better results for the ad and social campaigns they’re running. Being able to do that on a local level takes that to an even higher degree of being relevant and contextual to the consumer. I think that it’ll be something more and more brands are trying to do when they do their campaigns through social media.

Max Antonucci is an intern at Street Fight.