Preface: This is more of a solicitation for people to post feedback on their own projects and experiences, basically in an effort to offer free advice to others.
My undergraduate training, several internships, and first job out of college included a significant amount of systems design and analysis and general project planning/management. I noticed, however, that when I entered my first digital humanities position as a web developer/programmer that I had entered uncharted territories full of people without this training. (Does anyone else think this training could be useful for dissertation writing....) I felt like many of the projects I had been assigned to work on obviously involved a lot of initial effort in getting the grant money, but much of this time was spent on the humanities narrative and very little time had been spent on the work plan, realistic budget, project scope and possible obstacles, etc. I felt, and still feel to a large extent, that this is detrimental to projects. I realize that a typical corporate project plan may not work for various reasons (e.g. lack of funding, staff, tight schedules, bureaucratic requirements to granting institutions,etc) but I also believe that project planning could and should be more organized from the beginning. One way I have tried to alleviate problems related to poor project planning in my projects is by following several steps during the grant writing process, even while the narrative or other pieces are still being finalized:
1) map/sketch out (literally) whatever the final project is going to look like on paper or a whiteboard and discuss it with other project members --- then take your idea and discuss it with people not involved in your project. Specifically, make sure you can explain not only the why the project is significant and needs to be done this way (which is what we humanities are so good at doing) but also how something is going to get done. Can someone repeat back to you what the point of your project is and how you're going to do it? Also, if you don't yet know how, then you need to solicit the advice of someone who might know how ASAP. [There are a number of project planning tools out at this point, but I'd like to hear what people are currently using for whiteboard drawing/planning.]
2) outline in some detail the steps/process for how this project is literally going to happen. What steps have to take place and in what order? This is a big problem for those not informed with how the technology behind something actually works. A search engine on a site cannot really search anything unless a) the site is at least somewhat complete in terms of design b) any necessary databases or underlying infrastructure is setup and c) something is there to be searched. Dummy data can be used for a while, but it only goes so far, and the programming process is not a one-step process. Lots of prototyping and testing will need to start early and continue through several phases of the project.
3) a timeframe needs to be attached to each of these steps and to all tasks, and anything else in the project that might simultaenously take place (such as content development or data collection outside of the final project). This timeframe should leave some extra room for scope creep: if you think two weeks then consider that it might take you four and write that instead. (After all, can you really have the design of the site finalize in two weeks if you can't meet with the designer in those two weeks because you're interviewing people or collecting data for the project elsewhere?) When you write down a step/task it is vital to note who is going to actually complete that task. How much time does that person have? How many additional people will need to be hired to help complete that task and how much will you need to pay them? If you can't answer that, you need to find someone who can help you determine and assess how much time and money this costs in the real world. Humanities people (myself included) tend to take on too much at once and lack the ability to say 'no' to additonal tasks. (Talking to someone who works in the field professionally can give a better sense of how long something like scanning boatloads of documents for OCR, or programming a new widget into a wordpress installation, or writing PHP code, actually takes.)
4) During the grant writing process, as well as post-grant, regularly use a project management or task completion tool for tracking project progress and to keep yourself on task, particularly if more than one or two people are concurrently working on the proposal. I have used basecamp in the past and liked it. Basecamp has writeboards, calendars with deadlines, task lists, the ability to upload documents, and you can manage or be a part of multiiple projects at once. It's essentially a centralized location for a project and its associated documents, task lists, etc, without having to mass e-mail and spam people constantly and without having to search through folders of old e-mails for what you need. [I would love to hear people's experiences with other tools like this.]
In the end, I have found that people and projects who stick to these basic steps are able to avoid major scope creep and tend to be the most successful ones, where measure of success are according to who finishes roughly on time having completed the majority of the tasks set out in the initial grant, or even expanded into something new and exciting, ready to apply for a bigger and better grant.