One of the most exciting features of Google Universal Analytics is the ability to capture offline data that can be correlated with online activity. Prior to this offering, digital analysts, programmers and/or DBAs would have to develop a methodology for correlating disparate online and offline data sources. Today, there is an automated way to get outside data where digital analysts and marketers need it. The programming process of capturing and sending offline data to Universal Analytics is a technical one. This overview is for marketers and product managers on how Universal Analytics offline data capture works and to offer some ideas about how offline data can be used for online optimization.
As with any process, proper planning is key is setting up offline and online correlations. Begin your task by clearly defining what it is you want to measure. Specify the metrics and dimensions you want in your definition. You may find it helpful to write your objective in sentence form. Make sure your final sentences contain your metrics, dimensions, and any specifics. For example:
- I want to measure how many visitors (metric) to our site use a loyalty card (offline custom dimension) in one of our stores.
- I want to measure how many visitors (metric) to our site make a purchase (segment) and use a loyalty card (offline custom dimension) in the store.
As a practice, the sentences were written and then the metric, segment and dimension notations were added as a check. Clearly defining what you want to measure, or what hypotheses you are trying to test, will reveal the roadmap for ensuring the measures you need will be available.
Simply saying “now you can go and bring your offline data in” is an unfair, oversimplification of your technical requirements. A very high level explanation of the import process is programmers capture and send data in an http request to Google’s data servers. Google has detailed documentation for programming and capturing offline behaviors in its Measurement Protocol Development Guide and in its Measurement Protocol Reference. When the behind the scenes wizardry is complete and the data is available in your Universal Analytics account, you can apply metrics, build segments, and correlate events like and other data element.
There are countless industry-specific uses for correlating offline and online data. Below are a few ideas that use available APIs or company data sources to illustrate the power of this feature:
- Weather Related Data
Many businesses, especially retailers, see a difference in customer behavior depending on the weather. Retailers can adjust content or promotions based on weather forecasts or conditions. Sites planning and content optimization can use trending data to correlate sales and engagement to weather patterns.
- Crime Statistics
Realtors and homebuilders can correlate visitor engagement with neighborhood crime statistics. This data can be used to adjust content, online promotions, or the voice of specific pages.
- Center for Disease Control Data
Healthcare providers and pharmacies can adjust promotions and content based on health and safety trends. Trend data can reveal the best time to run promotions or communicate specific messages with customers.
- Loyalty Cards
Customers that use store loyalty cards provide a wealth of purchasing information. Customers that register their cards online can be matched in Universal Analytics to reveal how purchases are researched and completed across multiple retail channels.
The Bottom Line
Correlating online and offline data in Universal Analytics can be a complex technical task, but it has the potential for huge reward in the insights and optimizations companies can gain. There is a vast number of APIs online that give companies access to environmental data, or companies can use their own data sources to correlate to online activity. Thetechnical wizards at Search Discovery are available whenever you need help with capturing data and making sense of your Digital Analytics.