Braun and Clarke Thematic Analysis - How to do the six-step process in your qualitative research project
Braun and Clarke’s six-stage thematic analysis is one of the most widely used approaches in qualitative research. If you’re working with interview transcripts, focus groups, or other forms of qualitative data, there’s a good chance you’ve come across it already.
Of all the PhD researchers I’ve supported over the past two decades, there’s a moment that comes up again and again. I ask how they’re planning to analyse their data. They say, “I’ll probably use Braun and Clarke.”
On the surface, the process appears quite clear. There are six stages, each with a defined purpose, and a logical progression from raw data through to writing up your analysis. But as many of my own PhD supervisees have quickly realised, understanding the stages is one thing. Working out how to apply them in a way that feels coherent and defensible is something else entirely.
In this post, I’m going to walk you through the six-stage process step by step, using a simple example to make it concrete. As you read, it’s worth keeping in mind that thematic analysis is not just a sequence of tasks. It’s a way of working with data that involves interpretation, judgement, and ongoing decision-making.
Braun and Clarke’s approach to thematic analysis provides a structured way of identifying and developing patterns, or themes, within qualitative data. It involves working through detailed textual material and building an account of that data through themes that you actively generate as a researcher.
What I want to do here is show you how that process unfolds in practice, while also highlighting where it tends to feel more straightforward - and where it often becomes less clear.
The example we’ll use is from a project looking at people's experiences of remote working during the COVID-19 pandemic. I've deliberately kept this extract very short and compact for illustrative purposes, so you can see how thematic analysis would work. Below, I have also included the research title, aims and objectives, and research questions.
Research Questions
Let’s begin by outlining the six-stage process:
Familiarising yourself with the data
Generating initial codes
Generating initial themes
Reviewing themes
Defining and naming themes
Writing up
(1) Familiarising yourself with the data
This first stage is about getting properly acquainted with your data.
At this point, you’re not trying to decide what anything means yet. You’re just beginning to notice what’s there.
A simple way to approach this is in two passes: reading, then brief summarising.
Reading. Start by reading through your data without taking notes. This can feel slightly uncomfortable - especially if you’re used to highlighting or coding straight away. But it gives you something important: a sense of the data as a whole.
You begin to pick up tone, direction, and the kinds of issues that seem to be sitting underneath the surface - not just isolated fragments. Most people rush this part - and later on, that tends to show.
Summarising. After that first read, you can begin to write short summaries - a few sentences per page is enough. You’re still not analysing in detail. You’re just starting to externalise what you’re seeing.
This small step makes a noticeable difference later. It gives you something to work with when coding begins, rather than starting from scratch each time.
(2) Coding
This is where the analysis starts to feel more tangible.
You begin working through the data in detail, identifying segments that relate to your research focus and assigning labels - your codes.
Each code captures something of interest. As you move through the data, those coded segments start to build a picture of what’s present.
At this point, many researchers feel a sense of progress because you can see what you’re doing. You can point to it. It feels like the analysis is taking shape.
It is, but it’s still early.
The purpose of coding isn’t just to label content. It’s to prepare the ground for something more interpretive later on.
An example of this is shown below, where specific excerpts of data have been assigned codes that capture their key features.
At this point, most qualitative researchers feel reasonably comfortable (well - as comfortable as you can be with qualitative data!). Coding feels tangible. You can see what you’re doing, and there’s a sense of progress.
It’s afterwards that things start to shift.
It’s no longer just about what’s in the data, but what you’re doing with it. How those codes begin to take on meaning, how decisions start to matter.
This is where thematic analysis moves beyond process and into interpretation - and where many PhD researchers begin to feel less certain about whether they’re “doing it right.”
(3) Generating initial themes
Up to this point, you’ve been working closely in the data - breaking it down, organising it, labelling it.
Now you’re being asked to step back and work across it. To look at those codes and ask: How do these connect? What do they begin to say, collectively?
This is often where the wobble starts, because there isn’t a single obvious way to do this.
This is why Braun and Clarke use the term generate rather than discover.
Themes are not sitting in your data waiting to be found - this is not a treasure hunt. Themes take shape through your engagement with the data - through your perspective, your questions, and the decisions you make as a researcher.
This means this stage is unavoidably interpretive. You are not a neutral observer here. You’re actively shaping what the analysis becomes.
As you begin to develop themes, the task isn’t simply to group codes together. It’s to make sense of those groupings.
Why do these pieces of data belong together?
What is the central idea holding them in place?
What are you actually saying about your topic?
Those questions don’t always have immediate answers. For many of the PhD researchers I’ve supported through this stage, this is the point when things feel uncertain and the imposter syndrome kicks in. They ask themselves, “Who am I to say that this is what’s going on in the data?!”.
What themes are (and what they’re not)
A theme is a way of organising your data around a central idea - something that starts to say what it means, not just what is there.
This is where the shift happens.
Coding breaks the data down into manageable pieces.
Themes begin to build something from it.
For example, codes like “communication challenges” and “productivity impact” might start to come together around a broader idea about changing workplace relationships.
Or codes like “boundary blurring” and “social isolation” might sit within something larger about how work and home life are shifting.
(4) Reviewing themes
Once you’ve generated some initial themes, the next step is to revisit them to see whether they actually hold up.
This is where you start asking slightly different questions:
Do these pieces of data genuinely belong together?
Is this theme doing one clear piece of work - or trying to do too much?
Does the overall set of themes make sense as a whole?
It’s very common for things to move around here.
Some themes become clearer, some need to be split apart, others turn out not to hold in the way you first thought.
For example, you might start with a theme like “Boundaries between work and home life.”
As you look more closely, you might realise it’s carrying multiple ideas - boundary blurring, isolation, and perceived benefits - all at once. Separating it into more focused themes can make the analysis clearer and more precise.
What you’re really doing in this stage is strengthening your own reasoning. That can feel less certain than the earlier stages, because you’re no longer just organising the data - you’re deciding what counts as a coherent account of it.
(5) Defining and naming themes
Now, each theme needs to be defined, named, and described in a way that captures its central idea.
In practice, this often means writing a short description - a sentence or two that explains what the theme is capturing and how it relates to your research focus.
Simple, on the surface. However, this is also where something interesting tends to happen.
When you try to explain your themes clearly to someone else, you’re forced to see how well they actually hold together.
Does this make sense?
Is this really one idea, or several?
Does this reflect the data - or what I think the data is saying?
It’s very common at this point to realise that some themes need refining. You might rename them, tighten their focus, or go back and adjust how they were developed in the first place.
That’s not a setback it’s part of how the analysis becomes clearer.
By this stage, you’re no longer just describing your analysis, you’re making your reasoning visible.
This is where things either start to click or become harder to articulate.
What makes a difference here is how well the earlier stages - coding, theme development, interpretation - have actually been thought through and connected.
(6) Writing up
The final stage is writing up your analysis.
This can sound straightforward: present your themes, include data extracts, and explain what they mean in relation to your research question.
You’re expected to show how your themes fit together, how your interpretations are grounded in the data, and why your analysis is credible.
Which means making your reasoning visible - in a way that feels coherent, deliberate, and defensible.
If your coding and theme development from earlier on feel clear, writing tends to follow more easily. If they don’t, this is usually where that becomes harder to articulate.
There’s also the added layer of expectations.
Different disciplines and institutions have their own conventions around how thematic analysis should be presented - which can shape how you structure and write your chapter.
The example below gives you a sense of what this can look like in practice.
Next steps
If you’ve read through this and thought: “I understand the stages… but I’m not completely confident in what I’m actually doing within them”, that’s a very typical place to reach.
For many PhD researchers, the way it’s explained in textbooks doesn’t always translate easily into practice. It can feel abstract, overly rigid, or simply difficult to apply to your own data.
The way I teach thematic analysis is a little different.
It’s based on how I’ve supported PhD researchers over many years to actually work through coding, theme development, and writing up - not just understand the stages, but make sense of what they’re doing within them.
If you’ve found yourself reading the textbooks and still feeling unsure, this approach may feel more workable.
My Braun & Clarke Reflexive Thematic Analysis Guide is designed to give you that kind of structured support.
It walks you through how to work within each stage - particularly where things tend to feel less clear - so you can make your reasoning visible and develop an analysis that feels coherent and defensible. It’s here if and when you need it.
Feel like you want to throw your analysis away and start again “properly”?
This is exactly where most people get stuck with Braun & Clarke’s thematic analysis.
Not sure if you’ve coded things “right”?
Worried your themes don’t quite make sense (or feel forced)?
Reading papers over and over and still thinking, “I don’t get it…”
That doesn’t mean you’ve done it wrong - it usually means you’ve hit the messy middle
This guide shows you how to make sense of what you’ve already done and move forward with confidence, without starting from scratch.
Inside, you’ll find a clear, step-by-step breakdown of Braun & Clarke’s six-stage process, with practical examples and worksheets to help you actually do your analysis, not just read about it.
Swipe through the images to see exactly what’s included.
Got questions? Contact me using this form, I’ll be happy to help.
By purchasing this product, you agree to our Terms & Conditions.