I was asked a question recently about the differences in the UX process when starting with a new website from scratch, compared to the UX process of a redesign for an already existing website. I said that the processes are largely the same but with a existing client you have the benefit of having current statistical analysis to help inform your decisions. But it had me thinking if this is the right approach and is there more to this? What are peoples thoughts on this?
At a high level I would consider the main difference to be the maturity of the solution.
In one case you have presumably a relatively mature existing solution. Work has been done, decisions made and possibly expectations set about the understanding of the problem, including users, technical aspects, content, organisational aspects, information architecture etc.
On the other hand, you may have none of this.
How this impacts on a process could be that you don't need to perform some tasks because past decisions are perceived correct. E.g. The branding maybe? the content? This might manifest in people expecting you to re-use things, and this may impact on your process because you aren't starting with a clean slate. There may be organisation resistance in performing tasks they perceive as completed. There may be less budget because some has been spent already. You may have previous research you can use, but that research may or may not be valid.
There are extremes around this, but the middle ground might be that the expectation for a new client is that your spend more effort trying to understand the problem, but for an existing solution you might spend more time trying to refine an existing system. Not always the case, but can be.
The UK government runs most of its projects based on the GDS standards which involve distinct discovery, alpha, beta, live phases. You might be coming onto a project where discovery has been done, or where alpha is running and you need to consider the previous work in your solution.
Alternatively, it could be that the old system is to be thrown away and the first thing you need to do is work out exactly where people are failing with it and avoid those mistakes.
As with everything, the answer is "it depends"!