I am trying to find the correct way to obtain minute-level data extracts for session/visit starts. It seems as though this is not a standard option within either Report Builder or Data Warehouse where an hourly breakdown is the most granular level available.
I have been able to achieve this granularity in Workspace but that is not a viable option for regular use over large date ranges. In addition to this, I have also noticed that the sum of the minute-level line items exceeds the hourly totals (and, in turn, the sum of the hour-level line items exceeds the daily totals). Please could someone explain this to me? I can understand such a discrepancy when it comes to individual product totals vs. total complete orders but I can't see how this works in the context of user visits.
I have seen some minute-level data from Data Warehouse where the field 'Time stamp - start of session (eVar67) (evar67)' is included but this data includes quite a few rows where there is no entry for this field. Is there a link between these blank entries and the total vs. sum of line item discrepancy within Workspace?
Apologies if there is an obvious answer to this, I'm just trying to get my head around it all before settling on the best solution.
So I've done a bit more research, and the existing answers here are pretty complete. Between a new dimension, minute level data in workspace, and data warehouse, you'll want to weigh your options and pick the best route for your organization.
The only thing that hasn't been mentioned is that workspace will allow the download of up to 50k rows sometime in the next year or so. While this doesn't help immediately, this improvement may be helpful down the road.
Another alternative would be to schedule a workspace project that contains multiple tables, enough for all minutes in the day. Have it scheduled every day, then you can use the spreadsheet data however you'd like. Over time, you'd have the data from each day in it's respective spreadsheet.
You need to set up an integration in Adobe.io. You can do this if you are an admin.
These are two pretty good videos on getting going.
Using Analysis Workspace to Build API 2.0 Requests
jasond73569343, I would advise to implement a variable to store the time value in the format HH:mm:ss. Ideally, the value should be returned by the CMS. Then apply a classification to group all values in the desired buckets. This should help you reach the desired goal.
Use the Entry metric since you are interested in analysing response from TV ads.
Analysis Workspace is limited to 400 rows per freeform table; the api 2.0 is not limited to 400 rows per request. Further more if you are using an api you automate the process to say grab rows 1-50k, then 50001 to 100k etc.
The large advantage of an API is that it will allow you to address this programmatically, while still getting the underlying data you require.
I wonder if you find it the solution for downloading data from Adobe analytics by the minute? I am also looking for this solution but in report builder the minute granularity does not exist.
I hoop someone knows if this is possible,
I think it is possible in Workspace but not a practical solution given that it can only be done one day at a time and it takes a very long time to break the data down by the three categories that I need plus the hour and minute. The prospect of doing this for even a week's worth of data is very daunting, let alone for several months.
Thanks Andrey. I'm looking at this from a reporting perspective - would the existing 'Time stamp - start of session' custom breakdown from Data Warehouse be a valid variable for this purpose?
Having done some more testing today, I can see that the sum of visits per minute (using this 'Time stamp - start of session' field) from Data Warehouse is almost identical to the sum of entries per minute when broken-down to the required granularity from within Workspace. I'm just concerned by the number of rows where there is no timestamp included in the Data Warehouse exports.