To write a thesis, treat it as an argument you have to defend, not a record of everything you did. Check the expected length in your field, write methods first, then results, discussion, literature review, introduction and conclusion. Cut anything that doesn’t support your argument, and write a little every day.
Most PhD students sit down to write their thesis by opening a blank document called “Chapter 1: Introduction”. That’s the hardest place to start. Here’s the approach I used to write mine in a couple of months, and what I’d tell any student staring at that blank page.
Your thesis should be persuasive, not impressive
A thesis is an argument that you defend with evidence. It isn’t a lab notebook and it isn’t a list of everything you tried over three or four years. (If you’re not sure what the difference is, here’s my explainer on what a thesis actually is.)
My argument was that you could make relatively high-efficiency solar cells using a nanoparticle dispersion in the active layer. Every chapter had to support that.
So be ruthless. If a set of results doesn’t support your main argument, cut it. A peer of mine cut an entire chapter and finished on time. Those results didn’t vanish. They just didn’t belong in that thesis.
Check the length and structure in your field
Thesis length varies wildly between fields. My physical chemistry thesis was about 200-odd pages. I’ve seen maths theses of around 20 pages and physics theses of about 60, and some fields run much longer. Look at recent theses from your department and ask your supervisor what’s expected. I’ve covered this in more detail in how long a PhD dissertation should be.
The typical structure is introduction, literature review, methods, results, discussion and conclusion. Your field or university may differ, so check. My thesis outline guide walks through it.
The order to write your thesis
You read a thesis from front to back. You don’t have to write it that way. This is the order I recommend:
- Methods. You already know what you did. Make it super detailed: suppliers, software versions, settings. Someone should be able to repeat your work from it.
- Results. Only the results that support your argument.
- Discussion. What the results mean and why they matter.
- Literature review. Now you know exactly what your work contributes, so you know which literature to frame it with.
- Introduction. Set up the argument you’ve now actually made.
- Conclusion. Last, once everything else is settled.
Writing the concrete chapters first is common advice. The ANU’s guide to chapter writing also covers how to approach each chapter.

Your literature review has to critique, not summarise
This is where most students fall down. A literature review that goes “Smith found this. Jones found that.” is a summary. You need to synthesise: group studies, compare them, point out where they disagree and where the gaps are, and show how that leads to your work. I’ve written a full guide on literature synthesis.
Write before you’re ready, and write every day
Don’t wait until you feel ready to write. Writing is how you work out what you think. Ideas come from putting words down, not from staring at your data.
Work in small daily chunks. Half an hour a day adds up far faster than waiting for a free week that never comes.
Get your supervisor to turn chapters around quickly
I finished writing in a couple of months, and a big reason was that my supervisor returned chapters within two days. I’ve heard of chapters sitting on a supervisor’s desk for six months. Agree on turnaround times early, and chase politely if a chapter goes quiet.
Supervision style matters. In the video I mention a paper titled “Superhero or hands-off supervisor”, a study of Russian universities, which linked hands-off supervisors with the lowest student satisfaction and the longest expected time to degree. If your supervisor is hands-off, find support elsewhere: a co-supervisor, a postdoc or a PhD student near the end of their thesis.
Once the thesis is submitted, the next hurdle is the oral exam. Here is my guide on how to prepare for your thesis defense.
If you want structured help with the writing side of your PhD, take a look at the Academia Insider membership and course.
FAQs
Which chapter of a thesis should I write first?
The methods. You already know what you did, so it’s the easiest chapter to get down, and it gets you writing. Make it detailed enough for someone to repeat your work.
Should I write my thesis in order?
No. I recommend methods, results, discussion, literature review, introduction, then conclusion. The intro and literature review are much easier once you know what argument your results actually make.
Does my thesis need to include everything I’ve done?
No. Include what supports your main argument and cut the rest, even a whole chapter if needed. A thesis should be persuasive, not impressive.
When should I write the introduction and abstract?
Near the end. Write the introduction after your results, discussion and literature review, and leave the conclusion until last. The abstract summarises the finished thesis, so it makes sense to write it once everything else is done.
How do I structure each thesis chapter?
Each chapter should push your main argument forward. The usual layout is introduction, literature review, methods, results, discussion and conclusion, but check the conventions in your field and university.
How long should a PhD thesis be?
It depends on your field. Mine in physical chemistry was about 200-odd pages; I’ve seen maths theses of around 20 pages and physics theses of about 60. Check recent theses in your department.
Written by Dr Andy Stapleton


