“users hate change”

reddit comment with no particularly significant context:

Would be really interesting to put a number on “users hate change”.

Based on my own experience at a company where we actually researched this stuff, the number I would forward is 30%. Given an existing user base, on average 30% will hate any given change to their user experience, independent of whether the that experience is actually worse or better.

“Some random person on reddit” isn’t scientific evidence or anything, but it definitely seems pretty plausible to me that some very significant portion of any user base will generally dislike any change at all — I think I’ve been one of those users for software I don’t develop, I’m thinking of recent changes to Google Maps, many changes to Facebook, etc.

I’m not quite sure what to do with that though, or how it should guide us.  Because, if our users really do want stability over (in the best of cases) improvement, we should give it to them, right? But if it’s say 1/3rd of our users who want this, and not necessarily the other 2/3rds, what should that mean?  And might we hear more from that 1/3rd than the other 2/3rds and over-estimate them yet further?

But, still, say, 1/3rd, that’s a lot. What’s the right balance between stability and improvement? Does it depend on the nature of the improvement, or how badly some other portion of your userbase are desiring change or improvement?

Or, perhaps, work on grouping changes into more occasional releases instead of constant releases, to at least minimize the occurrences of disruption?  How do you square that with software improvement through iteration, so you can see how one change worked before making another?

Eventually users will get used to change, or even love the change and realize it helped them succeed at whatever they do with the software (and then the change-resistant won’t want the new normal changed either!) — does it matter how long this period of adjustment is? Might it be drastically different for different user bases or contexts?

Does it matter how much turnover you should expect or get in your user base?  If you’re selling software, you probably want to keep all the users you’ve got and keep getting more, but the faster you’re growing, the quicker the old users (the only ones to whom a change is actually a change) get diluted by newcomers.   If you’re developing software for an ‘enterprise’ (such as most kinds of libraries), then the turnover of your userbase is a factor of the organization not of your market or marketing.  Either way, if you have less turnover, does that mean you can even less afford to irritate the change-resistant portion of the userbase, or is it irrelevant?

In commercial software development, the answer (for better or worse) is often “whatever choice makes us more money”, and the software development industry has increasingly sophisticated tools for measuring the effect of proposed changes on revenue. If the main goal(s) of your software development effort is something other than revenue, then perhaps it’s important to be clear about exactly what those goals are,  to have any hope of answering these questions.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s