How to Avoid Sampling in Google Analytics
•Download as PPTX, PDF•
1 like•32 views
Find Out: ✔ What is sampling and who should be aware of it. ✔ When, why and in which Google Analytics reports you can find sampled data. ✔ How much money you can lose because of incomplete data. ✔ When and how to fight sampling, and what are the ways to do that in Google Analytics.
Report
Share
How to Avoid Sampling in Google Analytics
- 2. 1. Specializing in implementing Google Analytics 360 Suite and Google BigQuery In our clients’ projects there are more than 2M transactions per week 1. Developing OWOX BI services in Google Cloud Platform Works in Google Cloud Platform and more than 5000 companies worldwide rely on our expertise 1. Organizing professional events
- 3. Wait. What’s in the program? 1. What is sampling and who can face it? 2. In what cases and in what kind of reports sampling occurs 3. Why sampling is an issue 4. Is it worth fighting with sampling and how to handle it (within GA and API)? 5. Methods comparison for sampling avoiding
- 4. What is sampling and who can deal with it?
- 5. the method of selecting a subset of observables from a common set, in order to highlight certain properties of the original set Sampling is….
- 6. When and who can face sampling? Standard Google Analytics Google Analytics 360 500k sessions at the Property level for the used date range 100M sessions at the View level for the used date range
- 7. In what cases and what kinds of reports sampling occurs
- 8. How to define that the data is sampled
- 9. Reports without sampling Default reports from these sections: - Audience - Acquisition - Behavior - Conversions
- 10. Reports with the sampling possibility
- 11. Multi-Channel funnels and Attribution reports Reports with the sampling possibility
- 12. Reports in which sampling is most likely to occur ● Users Flow; ● Behavior Flow; ● Events Flow; ● Goals Flow.
- 13. Why sampling is an issue
- 15. How to avoid sampling (“within” and “outside” the GA interface)?
- 16. How to fight sampling in the GA interface ● Shortening of the date range ● Avoiding usage of “Ad-Hoc” reports, in case default reports fit ● Applying View-level filters to divide the entire amount of data ● Using separate Properties for each platform
- 18. Google Analytics Spreadsheet add-on
- 19. Unsampled reports (GA 360)
- 20. BigQuery Export
- 21. OWOX BI Pipeline (streaming) More about OWOX BI Pipeline
- 22. Comparison of methods to avoid sampling
- 23. Within the GA interface Solution GA 360 Default reports Setting shorter date ranges View-level filters Pros ● Sampling threshold: 100M sessions ● Unsampled reports ● Custom tables Always unsampled thanks to pre-calculated data The shorter the time span, the less data Less data, including only the traffic you want to see Cons Expensive annual license ● Max. 2 dimensions ● Limited set of reports ● More effort to retrieve data for longer time span ● Max. 5 dimensions ● Page-level dimensions inflate user count ● Max. 5 dimensions
- 24. Outside the GA interface Solution Google BigQuery Export for GA 360 OWOX BI Pipeline + Google BigQuery Google Analytics Core Reporting API Google Analytics Spreadsheet Add- on Pros ● Near real time hit data and unsampled session data export ● Max. 200 dimensions ● Raw real-time hit data ● Unsampled session data ● Unlimited number of dimensions ● Free for 14 days ● Programmatic way to pull out unsampled data ● API allows to send up to 50k query per day and returns up to 10k rows per query ● Up to 9 dimensions ● No coding required Cons Available for GA 360 only AdWords data retrieved through BigQuery Data Transfer Service ● Coding required ● Not all dimensions and metrics compatible ● Max. 7 dimensions in a query Unfeasible to use with large amounts of data