A thesis is made up of several chapters, including an introduction, definitions, related work, proposed solution, and conclusion. You must decide how much time (and pages) to spend on each of them. I call this writing resource allocation.
If this is done poorly, authors will waste a large part of their writing time on chapters that are not central to their thesis; for example, producing bloated definitions or myriad irrelevant technical details. Not only does this distract readers, it also robs authors of the time they need to write their central chapters carefully. Therefore, poor writing resource allocation is a recipe for a bad thesis.
So how do you do this step well? For me, writing resource allocation is a lot like allocating plate space when eating at a buffet. Beware: both problems have a similar solution strategy that is intuitive, widely applied, and reliable to produce poor results.
When I approach a buffet, I feel an urge to fill my empty plate. To get quick results, I put on the first things that I recognize as edible: salad, rice, pasta, and potatoes. Once I have covered some space on my plate, I get pickier and choose the stuff I really feel like eating, like roast pork, some lamb chops, and maybe a chicken breast or two.
When I am just about to leave with a full plate, I see the really tasty food at the end of the buffet. I would be a fool to leave it out! I push the food that is already on the plate together to make space for the scallops and king prawns.
The result is a mess and it is obvious to anyone that not much thought went into its composition.
Following established terminology, I call this the greedy allocation strategy, and I often see it applied to thesis writing as well.
When you approach your thesis, you feel an urge to fill the empty pages. To get quick results, you start writing the first things you know how to. This is often the introduction (you know your topic already), the definitions, and the foundations. It can be motivating, since you have made seemingly good progress by adding three pages summarizing the fundamentals on compiler construction you just heard a course about, and hey, a reader of your thesis needs to know that stuff anyway. So it should be in the fundamentals, shouldn’t it?
Much like at the buffet, however, you pay the price for the greedy strategy when it comes to the important chapters. When it comes time to write the contribution and evaluation, or whichever chapters matter most, you find yourself pressed for time. To make things worse, the really interesting ideas often come only after you have been immersed into a topic for a while; that is, at the end of your writing time. Just like with the tastiest buffet items, there is no space left. They end up either being left out or they compromise the other unfinished chapters (that is, the most important ones). Just as with the messy plate, the thesis composition in unconvincing.
The solution to both problems is simple in principle: Do not use a greedy strategy. Instead, make an up-front choice on how much space each item deserves. You want to have king prawns and scallops? Then take rice and some salad, but leave out the pasta and potatoes. You need time to focus on your contribution and evaluation? Then don’t write three pages on compiler construction fundamentals and instead reference a compiler construction book.
I have outlined a pragmatic thesis writing resource allocation process below.
Determine page budget
The first – and most crucial – step is to determine the number of pages you can realistically write. This is your page budget for planning. For this, determine (a) the number of days available for writing and (b) your average writing speed, and then multiply the two together. Do not count weekends or holidays as you need them to relax and as buffers. For speed, measure the average case rather than the best case. Our goal here is a realistic measure. If it’s too optimistic, you will be planning for pain.
Allocate page budget
The second step is to allocate your budget. For this, create the outline of your thesis. For now, it contains chapters and sections. Depending on the importance of the section, optionally add subsections and a rough collection of the points you want to make, in the form of a bullet list. Do not add too many details, as we want to focus on the big picture. Now assign your page budget to the chapters. You may spend less but you must not spend more. Now break down the budget for each chapter into its sections. Check the sums to make sure that the allocation is consistent.
When you run out of budget, cut content. Do not increase assumed writing speed. This is crucial. The most effective way to free up much-needed writing time is to identify content that is not central to your thesis. Typical candidates for trimming are the introduction (focus on the real problem), fundamentals (cite textbooks), technical implementation details (chances are that the reader looks for less detail here anyway), and future work (it’s mostly guesswork anyway).
To evaluate the allocation, ask yourself if the page budgets of the individual chapters reflect their relative importance to your thesis. If not, take away pages from chapters you feel are relatively unimportant and add them to those that deserve more writing attention. After each change, make sure that your sums are consistent.
Before you continue, get your supervisor’s buy-in on this. Text is much harder to refactor than outlines. Now is the time to identify missing content or different perceptions of relative chapter importance. Iterate until you both agree on the outline and page counts.