I've found two concepts, projects and next actions, which are a key part of the GTD system, to be particularly helpful for this kind of problem. In brief, the idea is to maintain a list of projects, which are usually high-level goals, that will take significant time and effort to achieve. For example, "write paper about structural advantages of geodesic domes" might be a project. But you can't just go and do that and then tick it off a list - it's going to take a lot of work. If it's research, you probably have only the vaguest idea what you actually want to write!
That's where actions come in. Think hard for 30 seconds about the very next step you could productively take to move your project on. Then write that action down on an action list. You will have one next action for each project you care about.
For example, here the next step might be any of these:
- meet with supervisor and confirm research direction
- sketch out broad research aims
- brainstorm key ideas and identify a list of key papers to read
- search literature for recent work in this area
- email my friend who is an expert for some insight
- write a 1000 word literature review using what I've read
- plan the software to prove my idea about why geodesic domes are awesome
The point about actions is that they are concrete and achievable. You can sit down at the start of the day and pick one, and then do it, and then tick it off your actions list. When you tick it off, you think carefully, and then come up with the next action for this project, and add it to the list. This process allows you to seamlessly jump between the big picture view (what you are trying to achieve overall) and the practical next step you need to make. Without the big picture, it's hard to know if what you are doing right now is useful for anything. Without the next step, it's easy to be overwhelmed by the enormity and uncertainty of your project, which can lead to loss of morale, followed by procrastination and loss of productivity.
Looking back over your list of achieved actions is particularly helpful in the case of research work, because it allows you to see that you have made progress and to see that your meandering path towards a solution is actually productive. In academia in particular, it is easy to feel that much of what you do is a dead end or waste of time. Being able to trace your thought process and objectively state all the things you did is a great way to stay motivated, justify your time and the decisions you made, and to make explicit the fact that finding a cool answer often means doing a lot of stuff wrong first.