The software company I work for has a short release cycle, with only have eight weeks of development to put out a new version. Since my product is fairly in demand - or perhaps it would be more honest to say that enhancements are demanded - this means I'm pretty much always creating designs in addition to the other product management work I do.
Like project work, design work is subject to phases, which are magnified by the compressed release cycle. People who work on projects or design work may identify with some of the following phases:
Phase I: 'Out with the old, in with the new.' I just finished the designs for the current update and it's time to start thinking about the next update but I'm not quite ready yet. I hover between two updates and poke at a few things but it's hard to let go of the topics that have claimed my attention for the last eight weeks and get started. I tend to feel unconnected and out of sorts during this phase, maybe even a little bit burned out. Fortunately, this phase tends to be short because there's a deadline to meet.
Phase II: 'Ramping up.' I've started the designs for the next update but there are some questions from development and other groups around the current update so my attention is divided. I force myself to make steady progress but haven't really hit my stride yet. I don't yet have a stake in the current design beyond the obvious (and important) one that someone pays me to work on it. This is a restless, unfocused phase but reasonably productive.
Phase III: 'Can't. Talk. Must. Finish. Design.' Something clicks and suddenly the design owns me. Maybe it's something someone said, or maybe the design has just reached a critical mass, but the different design threads pull at me all the time, insisting that I resolve them into a cohesive pattern. There may be a number of false starts before I find the right balance of prioritizatation of requirements, easy configuration and use, and architectural fit. And I don't know how many people realize this about software design but expertise will only take you so far - a good designer also also needs empathy in order to anticipate mistakes people are likely to make and help them not make them. Empathy makes me grumpy and hungry so if you see me glaring at my computer with a cookie in my hand, you might want to save that question for later.
Phase IV: 'Bring it on home.' I'm 95-98% done. I've crossed some sort of design threshhold where the pattern has integrity and holds together but that last 2-5% still needs to be done. It's usually pretty boring but someone needs to do it. I look around for my minion but I don't have a minion so that leaves me. Around this time the current release is ready for testing and that's less exciting than the wide open field of design work, but not less important. Now it's time to start horse trading features with my colleagues in development, who invariably feel that I've given them too much work. And they're right about that, because I have surrounded the most important requirements with non-essential pawns that I'd like to see but may be sacrificed if necessary to protect the core design.
The cycle repeats. I start thinking on a high level about the next update and gradually easing back into Phase I for the next round... better hurry up and ask me that question.