Why Panel-Based Attribution Doesn’t Work for Localized Advertising

19
Oct
2017
10/19/2017

WHAT MARKETERS NEED TO KNOW

Advertisers want to know what amount of foot traffic actually visited their physical location as a result of seeing their mobile ads. The ability to measure and report on “last mile” conversions is an incredibly important part of measuring the success of mobile advertising.

But the marketplace is crowded and many advertisers find it overwhelming to differentiate between the various mobile attribution solutions in-market today.

In order to provide advertisers with clarity, we recently analyzed the main players in the mobile attribution space. We uncovered two fundamental reasons as to why panel-based attribution does not work for localized advertising:

1. Reporting is based on “estimated visits” versus “actual visits.”
2. In localized advertising, nationwide panels are not effective.

HOW PANEL-BASED ATTRIBUTION WORKS

Estimated vs Actual Visits

Most third-party attribution providers leverage a “panel-based” methodology which estimates projected location visits based on a small sample of opt-in mobile users. Panel-based attribution tracks the users they have data on and scales up the measurements – which means visits to a specific location are estimated based on broad assumptions.

A commonly sourced statistic from a panel-based attribution provider states that they can reach 1 in 100 US adults. Sounds impressive at first brush but unfortunately that is only 1% of all users – a very limited pool of mobile users that requires estimations for scale.

Under this model, an estimated 99% of reported visits are projections, not actual visits. This means you would see reporting for 100 visits; while in reality only 1 visit was verified. On the flip side, the reverse is also true. Actual visits based on valid lat/long data may not be included due to the small sample size projection.

In Localized Advertising, Nationwide Panels are Not Effective
For national campaigns focused on reaching large, broad-based audiences, panel-based attribution can provide a good indication of behavior. Panels can work well in situations where there is enough volume to get accurate signals. However, nationwide panels are not effective in localized advertising, where audiences are often measured in thousands, not millions.

For example, if a third-party attribution provider has a panel with 3 million opt-in users, then that panel represents approximately 1% of the current US population. Here are two examples of how panel numbers shrink when applied to localized campaigns.

Digging deeper, we know that 3% of a population is typically in the market for a new product or service at any time. Here we see how this impacts the reach of panels even further.

For this example, we’ll look at people planning to switch insurance providers in Fort Worth. Based on panel-based attribution, the effectiveness of a Fort Worth insurance campaign would be measured in total by 255 people.

And now, this image represents people shopping for a car in Knoxville. Using panel-based attribution, this Knoxville auto campaign would be measured by 55 people.

THE SIMPLI.FI WAY

Measure and Report on Actual Location Visits

We believe that measuring effectiveness and generating insights with panel-based attribution that makes assumptions based on a small percentage of measurable traffic is not an acceptable solution for marketers, especially for localized campaigns.

Simpli.fi’s Conversion Zone technology measures actual location visits (not projections). Conversion Zones are virtual boundaries drawn around an advertiser’s business location via a GPS map. For consumers that were previously served an ad after visiting another geo-fenced location, our technology recognizes that unique user and upon visiting the predetermined conversion zone, attributes their visit as an offline conversion for the mobile campaign. This happens for every user visit and without estimated projections.

Additionally, through our Geo-Conversion Lift metric, we use a convenient and intuitive baseline for comparison that includes natural converters–users who visit the target zone and then the Conversion Zone without being served an ad. By tracking this way, Conversion Zones allow advertisers to more accurately gauge and optimize the effectiveness of their mobile campaigns.

Effectively Localize Campaigns

Compared with competitor’s published numbers, Simpli.fi’s Data Network is many times larger, covering 90% of all smartphone users, over 1.3 billion registered devices and serving ads on more than 600,000 apps. These numbers reflect a significant increase over the 1% reached by panel-based attribution.

We see over 3 billion hourly records and 500 million unique location data points a day. We have a team of dedicated data scientists who are constantly improving our models in order to ensure we use only the highest quality data. After our data is cleaned, we see approximately 600 times the amount of location data signals when compared to others in our space.

This immense set of unstructured data allows Simpli.fi to deliver a deep level of granularity and flexibility to our clients. Advertisers can target, bid, optimize and report on the individual data element level. This combination of data granularity and transparency delivers scale and performance for hyper-targeted local campaigns.

THE KEY TAKEAWAY

In today’s world of performance-based digital media, advertisers should accept no less than transparent reporting on the actual impact of their online-to-offline efforts. Simpli.fi’s clients have access to a proven solution that effectively localizes campaigns and tracks actual visits – not estimated projections based on a small sample.

Would you name the winner of the Daytona 500 after only 2 laps?


Would you call the winner of a football game 2 minutes into the game?

Could you accurately predict the ending of a feature length movie after less than a minute?

If your answer is no to these questions, then panel-based attribution is not the best way to measure the success of your localized advertising campaigns.

You can download the PDF of this content here for easy sharing within your organization. Whether you are just getting started in programmatic or want to drive more relevance and increase performance, reach out to us at hi@simpli.fi.