UX attrition rates are a serious problem for most digital service and product providers and it’s been one for a long time. The past was a different world, one in which computers and all things digital were just inherently difficult to use, and little of competitive nature was being developed. This isn’t the world we live in in 2013, where even our phones are little smart computers with complex operating systems and schools of design for multi-purpose software solutions. UX attrition rates are a serious concern and for a multitude of reasons. The biggest nail in the coffin for any UX design is simple obsolescence. This can come from two directions, the primary being that better interface standards and control options have been made available by refinement of platforms and hardware, removing limitations that were in place when your software was designed. This isn’t something you can actually do anything about, it’s the one aspect of attrition I really can’t give you any advice on preventing, because frankly, there is nothing that can be said for it. Sadly, in this circumstance, we just have to roll with the punches.
Thankfully, SaaS makes responding to and handling this attrition, to update the UI in response to obsolescence of this sort. It also makes combatting the other aspect of obsolescence easier to deal with, though it’s not as final as the passage of time itself. The other cause of this is due to changing trends in design, which often vary on the base standards of practice of the time, but nuances, aesthetics, navigation patterns and the like will drift in varying directions at a given time. It’s best to just pick a neutral plan, and stick with it, and possibly, if you’re up to the challenge, allow some customization by users, which lets them tweak these nuances as they please. This prevents your base UX from becoming “uncool” or unappealing, by letting the users take over this. This is the essence of self service, used in a new way.
Still, other factors contribute to attrition and these are things that contingencies can combat. The biggest problem is that once a system is out there and the user base grows and spends time with the system, their requirements of it will grow in scale and demand and they will attempt increasingly more complex (but possible) tasks. These tasks can get very involved and inconvenient to go about after a time, because the UX, both in UI, navigation and presentation is not designed to simplify such things. This is where testing, contingencies and a period of NDA-sealed user “playtime” is a good way to prevent this kind of breakdown by demand. This is actually the biggest preventable cause of UX attrition rates being so high, on account of designers not accounting for the more adventurous and evolving demands users will have of their software in time. Naturally, if this occurs anyhow, it can be addressed by updates, which again are much less of a costly chore to conduct in SaaS than the old traditional models, and we can be thankful for that.