Geography-based bid adjustments are powerful tools that allow advertisers to optimize bids based on a number of different factors related to a user’s location. Whether using
Whether using proximity bidding, average household income (HHI) levers, or standard geographic location targets, advertisers now have more ways than ever before to update bids for users that meet different geographic criteria.
However, it’s important to note that these tools are only as effective as the location tracking capabilities of the search engines allow for. This is because the share of total traffic that can be properly attributed to a location becomes smaller as the type of location becomes more granular.
To this end, it appears that Google has been making big gains over the past year in terms of how much traffic it can track at more granular levels, which has implications for a number of its initiatives, including mobile and in-store tracking.
One method of gauging Google’s ability to track user location is to compare what share of U.S. traffic it can roll up to the ZIP code level in the Most Specific Location column of its Geographic reports (found in the Dimensions tab of the UI) to the share of traffic that is reported at the city level in the City column of the same report. This data is also available through the API.
As a ZIP code is the most specific location that will currently populate in the Most Specific Location column, the difference gives us an idea of how many users would not be targeted if a campaign targeted, say, all of the ZIP codes in a city compared to the city itself.
Here’s how the share of U.S. traffic rolled up to these levels of location granularity shake out across Merkle|RKG’s advertiser base:
In Q1 of 2015, Google was able to track 71% of desktop and tablet traffic and 67% of mobile traffic to a zip code, compared to tracking 97% of desktop and tablet traffic and 93% of mobile traffic to a city. The remaining traffic gets rolled up to less granular location types such as state or country.
There are a few interesting things to note about this data:
Google’s ability to track users as granularly as possible is important to advertisers and to the search engine giant itself for a few reasons.
Advertisers using geographic location modifiers to adjust bids based on location want to be able to influence bids for as much of their traffic that matches to those different geographic locations as possible.
More granular location types are also tied to geographic attributes more specific to the population of that area (ex. average income of a ZIP code much more useful than the average income of a city), so advertisers want to embrace the most granular targeting possibilities.
However, they need to be confident that targeting a campaign to more granular location types won’t result in large shares of the population not getting targeted due to Google’s inability to track them to that level of location granularity.
To both of these ends, any improvement in search engines’ ability to track user location is a good thing, and makes not only standard geographic location targets more effective, but also makes tools like Google’s HHI bidding levers (which rely on ZIP code level location) and proximity bidding more valuable.
Better bids based on location result in a more efficient allocation of spend for advertisers, and should lead to an increase in spend overall going to Google – a win-win.
For obvious reasons, both Google and advertisers want to serve the most relevant ad copy possible to searchers with local intent. Particularly with campaigns targeted to very small radiuses around businesses or other physical locations, granular location tracking is pivotal to providing users with the best experience possible.
Particularly with campaigns targeted to very small radiuses around businesses or other physical locations, granular location tracking is pivotal to providing users with the best experience possible.
As many searches with local intent take place on mobile devices, it’s been especially promising to see Google’s advancements in tracking traffic from these devices to granular locations.
Google rolled store visit tracking based on user proximity to advertiser locations out of Beta in December as part of Estimated Total Conversions. Tracking those users which have Location History activated on their smartphones to instances when they arrive at a storefront, Google then extrapolates these tracked visits to provide estimates for the total number of brick and mortar visitors with a paid search click.
Tracking those users which have Location History activated on their smartphones to instances when they arrive at a storefront, Google then extrapolates these tracked visits to provide estimates for the total number of brick and mortar visitors with a paid search click.
These estimates are obviously heavily reliant on granular location tracking, and will only become more reliable as Google is able to track more users to brick and mortar locations. This is especially important for mobile, which typically sees a bigger lift from accounting for offline impact than desktop and tablet devices. For example, one RKG advertiser tracking these in-store visits sees one in-store visit for every three mobile conversions, compared to one visit for every ten desktop conversions.
This is especially important for mobile, which typically sees a bigger lift from accounting for offline impact than desktop and tablet devices. For example, one RKG advertiser tracking these in-store visits sees one in-store visit for every three mobile conversions, compared to one visit for every ten desktop conversions.
Hyper-segmentation and targeting is becoming increasingly important for advertisers as we strive to take account optimizations to the next level. However, these optimizations rely on the technology behind the scenes being able to correctly identify and bucket user context at a granular level.
As such, advertisers need to be aware of just how good search engines are at attributing not only geographic location, but also more user-specific information like gender or age.
When a campaign is set to target all of the 18- to 25-year-old males in the zip code of a college town, those ads aren’t actually reaching all of the 18- to 25-year-old males in that ZIP code. They’re only being served to the users for which the platform can assign an age, gender, and ZIP code level location to, and which meet the specified criteria.
Thus, advertisers may want to consider adding in “safety nets” to catch relevant searches that may not be tracked as granularly as the level they are targeting. For example, adding a city level target to a campaign targeting a radius around a store.
As Google will respect the most granular location target to which it can properly attribute the search, the most specific bid modifiers will be used for those searches which are tracked to the radius around the store, while the city target will allow the campaign to also garner those clicks which take place nearby but which can’t be assigned a location by Google as granular as the radius specified.
The good (GREAT) news is that Google has been making steady progress in terms of how well it can track user location, and is making particularly strong gains in the area of mobile devices, which have been harder to track than tablet and desktop devices in terms of granular location attribution. This progress will only continue to make for a better experience for users, more effective optimizations for advertisers, and increased spend heading Google’s way.
The post Google’s Steady Gains In Paid Search Location Tracking A Win-Win appeared first on Search Engine Land.