Do you think how often UX data should be checked?

Daily, weekly, once in a month, quarterly?

I think you will need to explain more about what you mean by UX (user experience?) data. Are you referring to assessing something like Google analytics information to determine usage patterns or something along those lines? In addition, the answer to your question is likely specific to your situation, which no one here knows anything about.

I was referring to users’ behavioral action on a website, more like data from session replay videos and heatmaps. Since I’m one of the co-founders of a behavioral web analytics tool called VisitorLAB I was trying to measure how often I should send a report to my users.

There is no single answer. In the days and weeks directly after an experience launches or is substantially revised, frequent reports are beneficial for assessing impact. But as a destination and its content mature, patterns are more identifiable with longer intervals.

Once a website is set up and the audience has settled in, I’ve got to assume that in the absence of significant changes, usage patterns would stay mostly the same and gradually change over time.

Of course, if there’s time sensitive content, marketing that’s driving new people to the site, new products, seasonal fluctuations, a change in the site’s structure and other variables that change the dynamics of people’s behavior on the site, it would be very useful to know how those changes have affected usage patterns.

All this, however, seems specific to the website in question. As HotButton said, there’s no single answer. As for sending reports to users, if at all possible, it might be best to let them see whatever data you have at any given moment by setting up a website where they can check whenever they feel a need to do so. If that’s too difficult/expensive to set up, you might allow the users to set the frequency of the reports they get from you, then charge them accordingly.

This is relative to your user frequency. For example, if your website/app is used by people on a daily basis, you could have short periods between reviewing analytic usage data since your data set will reflect unique users en masse.

I would also consider the portions you’re testing. If you’re testing a button’s effectiveness, you’ll want to consider how many of the users would want to click the button among the pool of all users. Is it applicable for all users to click?

I don’t think there is one answer to your question, but rather some loose guidelines based on the types of information you’re collecting and the types of data you’re seeking.