I've thought a lot about digital unionization [0] in the past year or two, primarily because I think it's an important approach to a number of problems we are witnessing. The thoughts below led some friends and I to start building
https://izens.net/ as one general approach to this problem.
The first thing to notice is that the calculus of unionization or collective action is very different than what we historically know. As a user, whether you see yourself as a customer or a laborer vis-a-vis Google, in both cases your position is different than traditional customers or laborers. The importance of convenience for all your transactions is much bigger. We probably aren't willing to spend an hour a week going to union meetings, because Google isn't as significant to us (as users) as a traditional employer, and because we have a similar relationship to a variety of other companies (Facebook, Amazon, Apple, etc.) as to Google, which would by analogy thereby entail a lot of time.
Moreover, it is much harder to build a social relationship with people in similar positions. At a job, or at a checkout line, you meet people in the same place as you. With a Google service, it's just doesn't happen naturally that you encounter people (and can easily and spontaneously communicate with people) in the same position as you (even though there are likely so many more). The companies themselves control how you inhabit their "space" - whether you can "meet" other people in the relevant way currently depends, for instance, on whether there is a chat or forum connecting you to those people. And of course, relying on their graciousness for that space is not going to go down well most of the time [1] as soon as this collective organization starts becoming significant. (This sort of control of how social interactions take place has been compared to the British enclosures.)
There are probably a number of valuable ways to collectively organize in this space. But the above considerations, and the fact that our own skills (and hence ability to contribute) are in programming and design, led a few friends and I to an idea: that we could create those spaces for collective organization via browser extensions and phone apps. The idea is that you install a browser extension (we haven't started on the app), and join one or more unions (or campaigns). The browser extension, based on the policy files of the unions you've joined, can a) block the website for some period of time for all its members, with some message; or b) display a banner when a member visits a site. The union members discuss and decide what the policy will look like. From there you can organize boycotts and make demands.
(https://izens.net/ (and the browser extension) is still in it's early phases, so feedback is much appreciated!)
[0] I agree that it may be misleading to think of it as a consumer union in most cases, for reasons other people have mentioned. On the other hand, the Posner & Weyl idea that we are data laborers suggests it might be a labor union. Ultimately, none of this matters much to the broader points.
[1] Compare this to moving out of WhatsApp, which I'm in the process of doing. I can use WhatsApp itself to get in touch with the people I want to coordinate with, thus making that coordinating relevant and timely.