MobileMobile & Local Join the Big Data Movement

Mobile & Local Join the Big Data Movement

A new batch of companies are bringing more data into the picture by layering in historical behavior, demographics, or users' movements. The goal is to bring more context to a given device, or to a spot on the map. Call it big data meets local.

As mobile and local continue to collide, we’re seeing lots of evidence that localized content boosts user engagement. That also translates to higher performance and monetization for mobile ads. But one thing often ignored is what exactly we mean when we say “localized content”?

Some define it as geotargeting via IP address. That can tell you what part of the country someone is in — hardly a level of precision that anyone should get excited about. Meanwhile, in mobile search (just as it is online) geographic modifiers in search queries can be used to determine user intent and thus more relevant results.

There’s also display ad placement where users’ local intent is inferred from their presence within a location-centric app like Yelp. That can involve local ad copy and calls to action (i.e. click-to-call, maps, etc.). It can also involve geofenced ad placement which matches a device’s position to an advertiser’s defined service radius.

The latter utilizes GPS but is at the mercy of whatever app publishers are collecting and passing along to ad networks or ad servers. This often requires an opt-in (“This app would like to use your location”), and doesn’t contain deeper data to pinpoint location-related variables like user profiles, intent and predicted behavior.

Just Like Regular Data, But Bigger

That’s where things are starting to get interesting. A new batch of companies are bringing more data into the picture by layering in things like historical behavior, demographics, or users’ movements from place to place. The goal is to bring more context to a given device, or to a spot on the map. Call it big data meets local.

I’ve spent some time talking with this new batch of data providers including PlaceIQ and Sense Networks. These join mobile local ad networks that are building similar systems in-house, including xAd and JiWire. The latter is the latest to make waves in this sub-sector with this week’s launch of its “Location Graph”.

In short, this is a tool to draw correlations between location and behavior. It does this by profiling users based on the patterns of where they are and where they’ve been. Those profiles are used for predictive modeling around future behavior, and thus what contextual and demographic ad targeting will most resonate.

“There are audience solutions out there, and location solutions from a geotargeting perspective,” JiWire CEO David Staas told me in a pre-brief last week. “Both fail to take into account historical usage patterns. Just like social graphs look at interconnectedness of people [our] patterns show linkages between places.”

For instance, Jiwire’s data has indicated that people who visit the Zoo have a high likelihood of also visiting family restaurants. Separately, sixty percent of women eat at the same three restaurants each month. The list goes on and on (and more can be seen in the graphics below).

The company accomplishes this by utilizing all the data it’s been compiling for years as a location based ad network. This includes profiling 500 million devices and 3 billion location tags. Combined with other data sets, this all comes together to form these individual user profiles such as moms, students, or business travelers.

connections-between-locations

Reach vs. Targeting

One argument that local data providers make is that geofencing alone falls short. Why target all users that step within a defined geographic boundary, when these emerging data sets can further segment the demographics that specific advertisers want to reach.

However a classic trade off in mobile local advertising is that the more you divide audiences, the more you segment yourself out of impressions. That’s especially a problem for brand advertisers that place a lot of value in sheer reach. In fact, that’s one of the reasons advertiser demand in mobile remains somewhat depressed.

Staas answers this with the assertion that the Location Graph’s audience profiling speaks the language that large advertisers know: demographic targeting. This frames it as buying certain audiences — which they’ve been doing for years — under the premise that location defines those audiences.

He contends with the scale challenge (resulting from audience segmentation) with a similar argument: Once the location graph identifies user profiles, they can then be messaged wherever they are, thus achieving the coveted campaign reach.

“Ten percent of mobile advertising has true location targeting,” Staas said. “We can use that to power the location graph and create user profiles, then use it to reach the additional 90 percent of ad inventory. Once I know an individual is a mom, I can reach [her] any time any place.”

The result of all of this: a 30-40 percent lift in ad performance over geofencing alone, according to Staas. Of course all of this raises privacy flags. JiWire, Sense Networks, PlaceIQ and others claim responsible use of data, such as anonymized pattern tracking. Or in the case of PlaceIQ the location is profiled, not the user.

Privacy concerns are sometimes overblown when legitimate companies are self-interested in maintaining responsible practices, such as IAB and MMA guidelines. But it’s still an important topic that must be kept in mind in order to keep ad networks in check. That’s an entirely separate discussion (and its own full column).

wheres-mom-look-in-these-places

Resources

The 2023 B2B Superpowers Index

whitepaper | Analytics The 2023 B2B Superpowers Index

8m
Data Analytics in Marketing

whitepaper | Analytics Data Analytics in Marketing

10m
The Third-Party Data Deprecation Playbook

whitepaper | Digital Marketing The Third-Party Data Deprecation Playbook

1y
Utilizing Email To Stop Fraud-eCommerce Client Fraud Case Study

whitepaper | Digital Marketing Utilizing Email To Stop Fraud-eCommerce Client Fraud Case Study

1y