Testing team in our organization runs test scripts in production to consistently check certain functionalities/features in our websites. As a result all such hits from such scripts are getting registered inside our Report Suite. Any idea how to exclude all such traffic from Report Suite level.
They have a specific email id that is used to login to the site and run the scripts.
Is there anything in the user agent string to identify these hits? We capture the user agent string into an evar, and then we can use a segment to filter out any of those hits from reporting and analysis.
Additionally, if everything is related to the one email address (and captured in evar/prop), you can create a segment that filters out all hits related to the particular email address.
Are they automated scripts in some sort of headless browser, or are they manual? Also, what are the scripts doing? Are they looking for an Adobe Analytics call to be made with particular key values pairs?
They are automated scripts run daily to check things like broken links, various functionalities working or not, etc. It's not manual. And they are not meant for tracking any analytical call.
Ok, then I guess it depends on how much involvement you want to have. If it is Selenium runner, then you can look at something like browsermob to proxy all of the network calls and just change the sitecat call to sitedog. Or you can use pupeteer and effectively do the same thing with rewriting the request. Another thought could be to write a plugin that injects a js variable onto the page that you read and swap out your report suite name with a test/dev one instead.
I understand the consequence here. Please take those tester IP whoever it belongs and add those in to the BOT rule. BOT rule will remove the data from the know spider (means here your tester) and generate the separate report under site metrics --> bot. It wont add the real production data in to the report suite. I hope this works for you.
Thanks a lot for your reply. Are there any other way of doing that apart from using IP? The reason being, it is not so easy to obtain the IP which involves lots of internal processes. Also not sure whether the IP will be static or dynamic. So wondering any alternative method if exists.
Is there anything in the user agent string to identify these hits? We capture the user agent string into an evar, and then we can use a segment to filter out any of those hits from reporting and analysis.
Additionally, if everything is related to the one email address (and captured in evar/prop), you can create a segment that filters out all hits related to the particular email address.
AGree wit this. Easiest way is to use a campaign code e.g. add cmpid=testers to the end of the URl on first login, then segment out the data. Suggest you use a virtual report suite for all your production data and reporting. Then you can easily apply this segment to the Virtual Report Suite
Currently we are capturing first name and last name of the user. We are not capturing email id. Tried creating a segment through which we can isolate test user name from the report. It is working fine when we are preparing monthly view of traffic data. But when we go inside Downloads report (where also we capture the user name), the segment does not work. Also there are other reports where we capture user level details of their activities where the segment is not working.
Currently we are unaware of the user-agent name through which we can also filter but trying to get hold of it.
Do any of the answers below answer your initial question? If so, can you select one of them as the correct answer? If none of the answers already provided answer your question, can you provide additional information to better help the community solve your question?
