I don't think it's usually a problem of "remove either A or B", it's usually "should we remove A or invest work in making it compatible with changes ahead?".
I see how usage telemetry can be useful in deciding whether or not it's worth it to keep supporting a feature, but I offer two counterbalancing points:
1. What people may be worried about - what I myself am worried about - is the methodology creep; it's too easy to end up having telemetry drive feature removal decisions, as in, "monthly report says feature X is used by less than 1% of users, therefore let's schedule its removal for the next sprint". The problem here is, telemetry alone will likely lead you astray. It's useful as a data source, not as the optimization function of product development.
2. If a feature you're worrying about has significant use, you most likely already know it without telemetry - all it takes is following on-line discussions mentioning your product (yes, someone might need to do it full-time). If removing the feature will have major impact on your maintenance budget, and non-telemetry sources don't flag this feature as being actively used, you can just axe it - revenue hit from lost userbase you've missed is unlikely to be big.
From this follows that the telemetry is most useful for deciding the fate of features that aren't used much, and don't cost much to maintain. At which point, I wonder, do you really have such low margins that you can't afford to carry the feature a while longer? I'm strongly biased here, because I'm going only by my personal experience - but I'm yet to see a software company[0] that doesn't have ridiculous amounts of slack. Between the complexity, management mess-ups, piles of technical debt and the nature of knowledge work being high-variance, having a feature slow your current development down by half[1] won't have much long-term impact.
--
The question thus is, are the gains from usage telemetry really worth the risk and potential ethical compromise? Would those gains be significantly lessened, if the telemetry was opt-in, and the company put more work into getting to know the users better? I suspect the answer is, respecting your users this way won't hurt you much, and may even benefit you in the long run.
--
[0] - Other than one outsourcing code farm I briefly worked in (my boss loaned me for a couple weeks to a friend, to help him meet a tight deadline), but these kind of companies don't make product decisions, they just close tickets as fast as possible.
[1] - And hopefully leading some devs notice the need for a refactoring, in order for that feature to not be a prolonged maintenance burden.
I'm a hundred percent certain that 1 is what Spotify does. It seems every release they remove or hide features that I use occasionally (e.g. clicking on the currently playing album picture to reveal the active playlist). It's extremely frustrating that since power users now are in a minority they are completely ignored.
At the first release like this they basically said "This new version has a lot less features, but you can vote on which ones we will add back!". They added none of the ones that got votes, and eventually removed the feature voting system altogether.
I see how usage telemetry can be useful in deciding whether or not it's worth it to keep supporting a feature, but I offer two counterbalancing points:
1. What people may be worried about - what I myself am worried about - is the methodology creep; it's too easy to end up having telemetry drive feature removal decisions, as in, "monthly report says feature X is used by less than 1% of users, therefore let's schedule its removal for the next sprint". The problem here is, telemetry alone will likely lead you astray. It's useful as a data source, not as the optimization function of product development.
2. If a feature you're worrying about has significant use, you most likely already know it without telemetry - all it takes is following on-line discussions mentioning your product (yes, someone might need to do it full-time). If removing the feature will have major impact on your maintenance budget, and non-telemetry sources don't flag this feature as being actively used, you can just axe it - revenue hit from lost userbase you've missed is unlikely to be big.
From this follows that the telemetry is most useful for deciding the fate of features that aren't used much, and don't cost much to maintain. At which point, I wonder, do you really have such low margins that you can't afford to carry the feature a while longer? I'm strongly biased here, because I'm going only by my personal experience - but I'm yet to see a software company[0] that doesn't have ridiculous amounts of slack. Between the complexity, management mess-ups, piles of technical debt and the nature of knowledge work being high-variance, having a feature slow your current development down by half[1] won't have much long-term impact.
--
The question thus is, are the gains from usage telemetry really worth the risk and potential ethical compromise? Would those gains be significantly lessened, if the telemetry was opt-in, and the company put more work into getting to know the users better? I suspect the answer is, respecting your users this way won't hurt you much, and may even benefit you in the long run.
--
[0] - Other than one outsourcing code farm I briefly worked in (my boss loaned me for a couple weeks to a friend, to help him meet a tight deadline), but these kind of companies don't make product decisions, they just close tickets as fast as possible.
[1] - And hopefully leading some devs notice the need for a refactoring, in order for that feature to not be a prolonged maintenance burden.