One argument for web and app analytics to continue to be used is to take a lesson from surveys. Yes, those pen-and-paper or phone surveys that are still conducted regularly these days.
I remember filling out a survey at a museum when a staffer asked me if I could help out. Two things stood out in that experience:
- I could have said "no", but I didn't, because my answers would be used to improve the museum's activities.
- Most of my responses were in line with what the staffer expected to see, though some of my responses surprised him.
Add in some statistical calculations, and all of the responses to that survey, including mine, could be extrapolated to be the responses of the general public.
So we already have a tested-and-proven, reliable method of measuring users' behaviours and attitudes that has been around for years, much much longer than when the first website was created.
Do you see where I'm going? I think web/app analytics will move beyond counting exact numbers to becoming more survey-like, with statistics (where machine learning is a glorified form of statistics) playing an important role to help us get a sense of how the general population behaves/thinks.
That's my zen-like $0.02 of the road ahead for web/app analytics. 😊
