Overview
Thematic analysis (TA) is not one method but a family. Braun and Clarke distinguish reflexive TA, where themes are patterns of shared meaning the researcher actively constructs; codebook approaches such as template and framework analysis, which use a structured coding frame; and coding-reliability TA, which treats themes as things to be found consistently by several coders. Many theses mix these without noticing, and examiners notice.
Our support starts by clarifying which version fits your research question, your epistemology and your department's expectations. From there we work alongside you through familiarization, coding and theme development: reviewing samples of your coding, challenging themes that are really topic summaries, helping you build thematic maps and keeping an audit trail of decisions you can show at your viva.
The interpretation is yours. We do not generate themes for you to submit; we ask questions, give feedback on your reasoning and show you model examples on anonymized extracts so you can analyze with confidence.
Who it’s for and the problems it solves
This is for you if you are…
- Undergraduate and master's students analyzing interviews, focus groups or open-ended survey responses for a dissertation
- Doctoral researchers who need their TA to be methodologically coherent with a stated paradigm
- Health, education and social care researchers writing up qualitative findings for journals
- Anyone whose supervisor said their themes are 'just descriptions' or 'too many and too thin'
Common problems we help with
- You have coded everything but cannot see how codes become themes
- Your themes read like interview questions ('Experiences of training') rather than claims about meaning
- A reviewer asked for inter-rater reliability and you are not sure whether reflexive TA needs it
- You are unsure whether to say themes 'emerged' and how to write about your own positionality
- Your write-up is a list of quotations with little analytic commentary between them
What’s included — and what isn’t
Included
- A scoping conversation to place your study within reflexive, codebook or coding-reliability TA and align it with your research question
- Guidance on inductive versus deductive orientation, semantic versus latent coding and experiential versus critical framing
- Feedback on a sample of your coded transcripts: code labels, level of interpretation, consistency and gaps
- Theme development sessions: clustering codes, drafting candidate themes, testing them against the dataset and building thematic maps
- Help writing theme definitions (what each theme is and is not) and choosing illustrative extracts
- Reflexivity support: a reflexive journal template and prompts on positionality
- Review of your methods section and findings chapter for coherence with Braun and Clarke's guidance
- NVivo, ATLAS.ti or MAXQDA project setup for TA if you are coding in software
Not included
- Coding your full dataset and producing themes for you to present as your own analysis
- Writing your findings or discussion chapter; we give feedback on drafts you write
- Reporting agreement statistics for a reflexive TA simply because a template asked for them — we help you explain the methodological reason instead
- Inventing participant quotations or altering extracts to fit a theme
What you receive
- TA positioning note
- A short written summary of which TA variant you are using, why, and the key terms (orientation, coding level, theoretical framing) to state in your methods section.
- Coding feedback report
- Annotated comments on a sample of your coded transcripts, highlighting where codes stay descriptive, where they could go further and where labels overlap.
- Thematic map review
- Feedback on your candidate and final thematic maps, including themes that may be domain summaries, subthemes that deserve promotion and themes that should merge.
- Audit trail and reflexive journal templates
- Simple templates for logging analytic decisions and reflexive notes, so you can evidence quality without borrowing positivist criteria.
- Draft feedback
- Tracked comments on your methods section and findings chapter, focused on analytic depth and coherence.
How it works
Positioning
We discuss your question, data and paradigm, and agree which TA approach you will use and how you will describe it.
Familiarization and early coding
You read, note and code a first set of transcripts. We review a sample and discuss how interpretive your codes are.
Generating and reviewing themes
In working sessions you cluster codes into candidate themes; we test each against the data with you and help you revise, merge or discard.
Defining, naming and writing
You write theme definitions and a findings draft. We give feedback on analytic narrative, extract selection and consistency with your stated approach.
Approaches and concepts we work with
- Reflexive TA (Braun & Clarke)
- Six recursive phases: familiarizing yourself with the dataset; coding; generating initial themes; developing and reviewing themes; refining, defining and naming themes; and writing up. Researcher subjectivity is treated as a resource.
- Codebook TA (template, framework analysis)
- A structured coding frame, often partly developed in advance, that suits applied, team-based or policy research with fixed information needs.
- Coding-reliability TA
- Associated with authors such as Boyatzis and Guest and colleagues; uses a fixed codebook, multiple coders and agreement statistics like Cohen's kappa.
- Inductive and deductive orientation
- Whether coding is driven by the data or by existing theory and concepts; many projects combine both, and we help you describe that honestly.
- Semantic and latent coding
- Coding surface meaning versus underlying assumptions and ideas; the balance shapes how interpretive your themes can be.
- Quality practices
- Reflexive journaling, audit trails, thick description and Braun and Clarke's published quality checklists, rather than saturation claims borrowed from grounded theory.
Why reflexive TA does not report kappa
In coding-reliability TA, agreement between coders is evidence that themes were identified accurately. Reflexive TA starts from a different assumption: meaning is constructed through the researcher's engagement with the data, so two thoughtful researchers may legitimately code differently. Braun and Clarke therefore argue that inter-rater agreement scores do not fit the logic of reflexive TA.
That does not mean anything goes. A second person can still read your coding to prompt reflection, and you still need a transparent audit trail. If a reviewer or ethics template asks for kappa, we help you write a clear, referenced explanation of why your approach uses reflexive quality practices instead, or help you switch to a codebook approach if agreement genuinely matters for your aims.
From topics to themes
The most common examiner criticism of TA is that themes are 'topic summaries': buckets that collect everything participants said about a subject. A theme in reflexive TA is a pattern of shared meaning organized around a central idea, and it usually makes a claim. 'Training' is a topic; 'Training as a performance of competence rather than learning' is a theme.
We work through your candidate themes with this test in mind, check whether each has a clear central organizing concept and help you decide what belongs in a subtheme, what is contextual and what can be let go.
- Does the theme name tell a reader what you found, not just what you asked?
- Can you state the central organizing concept in one sentence?
- Do extracts across several participants support it, including some variation?
- Is there analytic commentary between quotations, not just paraphrase?
Writing up so the method is visible
A strong TA methods section names the variant, the orientation, the level of coding and the theoretical position, and describes what was actually done in each phase rather than repeating the six-phase list. We also help you avoid phrases Braun and Clarke specifically critique, such as themes 'emerging' from data, and write about your positionality without turning it into a confession.
What the work can look like
Illustrative example — fictional data
Illustrative feedback on candidate themes
A master's student interviewed 12 newly qualified nurses about their first year of practice and brought five candidate themes to a session. The study, participants and theme names below are invented to show the style of feedback.
Candidate theme: 'Support from colleagues'
Feedback: this currently collects every mention of colleagues, positive and negative. Your codes suggest something sharper: participants describe informal support as conditional on appearing competent. Consider reframing around that central idea and testing whether it holds across at least most interviews.
Candidate theme: 'Stress'
Feedback: 'stress' appears in nearly all codes and may be a context for the whole analysis rather than a theme. Look at how participants talk about it: several distinguish stress they felt entitled to voice from stress they hid. That tension may be the more interesting pattern.
Next step agreed
The student will redraw the thematic map with three to four themes, write a one-sentence central concept for each and return with two extracts per theme that show variation.
Illustrative example only. The scenario and themes are fictional and show the type of feedback provided, not a real client's analysis.
What you’ll need to provide
- Your research questions and a short description of your methodology or paradigm
- Anonymized transcripts or a representative subset, with consent for them to be shared for support
- Your current codes, memos or software project file, if coding has started
- Any supervisor, examiner or reviewer comments on your analysis
- Your institution's rules on acknowledging support
You don’t need everything ready to ask for a quote — tell us what you have and we’ll say what’s missing.
What affects price and turnaround
Price depends on
- Size of the dataset (number and length of transcripts) you want feedback on
- Whether you need positioning and planning only, or support across all six phases
- Number of feedback rounds on coding, theme maps and chapter drafts
- Software setup needs (NVivo, ATLAS.ti or MAXQDA) versus manual or spreadsheet coding
- Depth of review for methods and findings chapters
Turnaround depends on
- How much coding you have done before feedback starts
- The pace at which you can revise between sessions; theme development is iterative by design
- Your submission or supervision deadline and the size of the sample we review
Every project is quoted individually. How we price our work.
Questions people ask before starting
Can you just code my transcripts for me?
In reflexive TA, coding is analysis, so if we coded your data the analysis would no longer be yours. We can model coding on a short anonymized extract, review your coding and discuss alternatives. For funded or team projects using a codebook approach, where coding by a team member is normal and acknowledged, we can discuss a different scope.
My supervisor wants inter-rater reliability. What do I do?
First, check which TA your supervisor has in mind. If you are using reflexive TA, we help you prepare a short, referenced rationale to discuss with them. If they want a codebook or coding-reliability design, we help you set that up properly, including a codebook and agreement procedure. Your supervisor's decision stands; we help you make an informed case.
How many themes should I have?
There is no fixed number. Two to six well-developed themes is common for a dissertation, but depth matters more than count. We help you judge whether themes are rich enough and distinct from each other.
Is my data kept confidential?
Yes. We ask for anonymized transcripts, use them only for your support and can sign a confidentiality agreement. Please check your ethics approval allows sharing data with a third party for analysis support.
Do I need NVivo or another program for thematic analysis?
No. Many good analyses use Word, printed transcripts or spreadsheets. Software helps with larger datasets and audit trails. We can support either route.
Can you help with open-ended survey responses as well as interviews?
Yes. Short survey responses tend to support more semantic, descriptive themes, and a codebook approach often fits better. We help you choose and describe the approach accurately.