> First, is it possible to develop software without inconveniencing business users with temporary ... regressions and inconveniences? The trick is to avoid int

> First, is it possible to develop software without inconveniencing business users with temporary ... regressions and inconveniences? The trick is to avoid introducing change without simultaneously introducing perceived value to the user. Predictability and reliability are incredibly important for business users, because they're going to build processes and documentation and training around however they interface with your system. This includes both directly interfacing with your system and creating supplementary processes outside of your system to work around deficiencies your system has in supporting whatever need they have. If you introduce change, you inherently reduce the temporary predictability and reliability of the system, and the friction created for end users needs to be perceived as worth whatever value your change provides. If they perceive a benefit to themselves, it'll incentivize them to embrace and adapt to the change. If they perceive it as a regression to their processes/needs, they'll fight it and it'll eventually lead to resentment and friction between the business users and the developers. Slack's WYSIWYG input box is a prime example - it unlocks capabilities which is convenient from Slack's perspective, and is potentially generating value for users not familiar with Markdown and are used to traditional GUI-based rich-text editors. Helping them better support the type of corporate-wide adoptions that are lucrative for them. But due to Slack's origins, a large chunk of their user base prefers the older, Markdown-based editor and perceive a decrease in value from the change. If you're going to introduce changes such as this which will be perceived as a regression by a large enough subset of users, then you need to account for that in your change management process such that it isn't abrasively disruptive.