I am trying to create an active list using the regex option described in #3 for filtering by page views in the HubSpot knowledge base article for active lists.
I've used a debugger to test out the expression and it looks like it is correct, however, my list returns no contacts that match the filter, despite contacts who do match it. Is the issue here that it's taking HubSpot's internal processes more than one second to test the contact against the condition and therefore HubSpot is automatically returning a value of false as that KB article states as the timeout limitation or is there an issue with the regex formatting I'm trying to implement?
The team has confirmed URL-based list filters ignore utm_ query parameters, # symbols, and everything that follow them.
Thus, visited URL and regular expression filters containing a /#/ or //# unfortunately will not match with contacts who have viewed URLs containing those symbols
We still plan to explore options to support list segmentation on URLs with fragments and query parameters in the future, but the team does not have a definite time frame on implementing such a feature. Apologies for the inconvenience.
At this time, the regular expression https:\/\/app.zizera.com\/.*\/publisher\/login\/\.\.* is the closest we can get to finding contacts who view URLs like https://app.zizera.com/whatever_value/publisher/login//#/ZIZERA/login. I've updated this test list as an example.
I was able to confirm that the analytics system does not strip all query strings — only utm parameters, native HubSpot tracking parameters, and URL fragments. Per this article:
HubSpot will not strip non-UTM query strings from visited URLs on contact records (e.g. 1www.domain.com/page?string`).
Thus, you should be able to use and segment on query parameters as long as you don't use # or utm variations.
OK, thanks for the clarification. Order to work around this we will have to change our implementation and manipulate the logged URL's in such a way that we will be able to segment contacts on exactly what they have done on our website and in the system we provide. In order to do this we need to get information on what characters in a URL that shall not be used. if there are more than "?" or "#".
Thanks for posting the feature request, I have already voted on it and will also comment it later on.
I sympathize with your frustration. Frankly, I was surprised to find this to be the case as well.
While the team is aware this is a paint point for customers like you, they don't have an immediate plan to address it. They have chosen to continue canonicalizing URLs in lookups for the time being because they believe changing the functionality would greatly confuse users who are unfamiliar with query parameters. The issue is on their backlog, however.
I want to see page view filters match your needs and expectations, so I have created this feature request on the HubSpot Community Ideas Forum to capture feedback. Please upvote and comment on it.
The product teams regularly monitor posts and generally prioritize ideas which have the most engagement from and greatest potential impact on users. As such, engagement on my idea will help keep this issue top-of-mind for the product team. I recommend including as much detail as possible per these best practices. The more the product teams understand a feature's use case, the more accurate they will be when scoping it.
I will eventually close this topic in the to maximize feedback on this idea.
Hi @Isaac_Takushi and many thanks for looking into this further.
Unfortunately this is very bad news as all our URL's contain a "#" and the important part to filter on happens after the "#" so the regex you are suggesting does not help us at all. Browsing around the web it is very common that a "#" is included in the URL so this issue is potentially already affecting other customers of yours. Are there any plans to have this addressed in the near future?
The team has confirmed URL-based list filters ignore utm_ query parameters, # symbols, and everything that follow them.
Thus, visited URL and regular expression filters containing a /#/ or //# unfortunately will not match with contacts who have viewed URLs containing those symbols
We still plan to explore options to support list segmentation on URLs with fragments and query parameters in the future, but the team does not have a definite time frame on implementing such a feature. Apologies for the inconvenience.
At this time, the regular expression https:\/\/app.zizera.com\/.*\/publisher\/login\/\.\.* is the closest we can get to finding contacts who view URLs like https://app.zizera.com/whatever_value/publisher/login//#/ZIZERA/login. I've updated this test list as an example.
"We still plan to explore options to support list segmentation on URLs with fragments and query parameters in the future", any updates on this @IsaacTakushi or a workaround to solve this challenge? Thank you!
When I checked the first contact's Last page seen values on November 15, they did not contain views of https://app.zizera.com/sodimac/publisher/login//#/ZIZERA/login. Now they do:
Yes, I removed \/\/#\/ZIZERA\/login from the test list's filter because I wanted to show a more inclusive list which you could use as a starting point for analysis. I recognize that it didn't include all the parameters you were looking for, but I thought 300+ matching contacts was better than 0.
I'm still looking into why we're running into issues with the #.
Hi again @Isaac_Takushi,
I now see that you have removed the "#" from the regular expression which is a problem. The URL´s containing the "#" are for users who have successfully signed in but the modified regex you have provided also returns users who have visited the login page but have not logged in and those are not supposed to be included in this list.
Almost all our URL´s contains the "#" character, why is this causing a problem?
This is indeed very interesting and I agree it is a good start! However your first statement is incorrect, the first user in the list has indeed visited a page with the expected URL pattern, how have you determined that it hasn't?
I will look further into the double slahes and see if I can get the list to work.
The first contact does not have any Last page seen values which contain login, so I would not expect the regular expression to match for him, however I did find something for the other two:
The second contact has viewed the page https://app.zizera.com/candicetomkins/publisher/login//#/ZIZERA/login:
I've been testing the corrected expression with this list, but have yet to populate the list.
Since we've corrected the expression, I believe we must be encountering the timeout issue and can't get more specific than https:\/\/app.zizera.com\/.*\/publisher\/login.*, which this test list reflects. All three of the contacts you provided are in this list, so it might be a good starting point!
Unfortunately I cannot provide you a list as URL wildcards are not working when filtering on page views hence I need the regex option to work in order to provide you a list
I have yet to find any issues with your expression, either. All the RegEx testers I've tried match that URL.
That said, if there's one part of the expression I'm suspicious of, it's that #. The RegEx testers accept when I escape the character with \# but it doesn't seem to make a difference to HubSpot.
While I'm leaning toward a timeout, could you share a few contacts (or better yet, a list of contacts) who should match the expression?