I am currently using clickstream data to build my own reports / visualization (as it is a piece of a bigger project) and was wondering how to go about to analyze my data using segment.
Is there a specific approach to follow?
Anyone who can help with the above?
Apologies there, I got my tools mixed up Data Warehouse is where you can apply segments, and not data feeds.
Pertaining getting data based on visits/visitors, there are a few good writeups on the topic. The following would be one I consider the best: Using Adobe Analytics Data Feeds and SQL for Basic Reporting – The Data Feed Toolbox
There's also a data feed lab at Summit this year that I anticipate would be valuable for your organization.
Thank you very much for sharing the link.
I will have a look at it when i have a moment.
Thank you for the info.
I do have the PDF but nothing really highlight best practice to look at data based on a segment as mentioned in my previous comment.
I was hoping to have some best practice which will help get the data based on visitor or visits etc...
Additionally, as far as i have seen while setting up data feeds, there are no option to apply a segment unless i am missing this option completely.
If it's going into an SQL-type database, you can use commands like WHERE or HAVING to effectively segment data. If you need to know what each column means, you can refer to this: Clickstream Data Column Reference
If you only care about the segmented data (and nothing else), you can apply a segment when creating the data feed, so the resulting file is only the data you care about. You'd be able to use the built-in segment manager to create the criteria you're looking for.
Thank you very much for you response.
For the moment, it will be very manual. FTP > extract and push into mysql (as I will start testing with a day or 2 of data).
I am thinking of using python to do the rest. As a start I just want to understand what I need to do for building segments as the Adobe PDF simply highlights the calculations for certain metrics.
I would really appreciate if you have any info / additional resources or advise.
Ideally, I would think that I would go towards Hadoop and apache spark. What do you think?
What tool are you importing your clickstream data into?