Template Analysis in Qualitative Research: A 2026 Guide
Qualitati Research Team · 2026-10-01 · 10 min read
Short answer: Template analysis is a codebook style of thematic analysis, set out by Nigel King and colleagues, in which the researcher starts with a small set of tentative a priori themes, builds an initial coding template on a subset of transcripts, then revises that template as it is applied to the rest of the data. It suits studies with clear research questions, many participants and a need for a transparent, hierarchical coding structure.
Template analysis is one of the most practical ways to analyze qualitative interview data when you already know, at least roughly, what you are looking for. It is common in organizational, health and psychology research, and it sits between fully inductive reflexive thematic analysis and the matrix-driven framework method. This guide explains what template analysis is, walks through its six steps with a worked example, compares it with neighboring methods and shows where AI-assisted coding helps and where it does not.
Key takeaways
- Template analysis lets you define some themes in advance, but treats them as tentative: they can be redefined or dropped (Brooks et al., 2015).
- The initial template is built on a subset of the data, not on every transcript, and then revised over several versions.
- Templates are usually hierarchical, often four or more levels deep, which is deeper than most reflexive thematic analyses.
- Integrative themes capture ideas that run across several clusters of the template.
- It is epistemologically flexible: it works from realist, subtle realist and contextual constructivist positions.
- AI can speed up applying a template to large datasets, but defining, revising and interpreting the template remains the researcher's job.
What is template analysis?
Template analysis is a form of thematic analysis that organizes coding around a "template": a hierarchical, written list of themes that the researcher develops early and keeps revising. In their methodological paper in Qualitative Research in Psychology (open access via PubMed Central), Joanna Brooks, Serena McCluskey, Emma Turley and Nigel King describe it as a flexible style of thematic analysis with a defined procedure, and position it alongside framework analysis as a "codebook" approach.
The method has two signature features. First, it permits a priori themes: you can start with themes drawn from your research questions, interview guide or theory. Second, it is iterative by design: the template is a working draft that changes every time real data challenges it. For a book-length treatment, see King and Brooks' Template Analysis for Business and Management Students (Sage, 2017).
The six steps of template analysis
Brooks et al. (2015) describe the procedure in six steps. The order is fixed, but steps four to six usually loop several times.
1. Familiarize yourself with the data
Read the full dataset, or a representative subset if the dataset is large. Note first impressions and anything that surprises you relative to your research questions.
2. Carry out preliminary coding
Highlight passages relevant to the research questions. If you defined a priori themes, use them here, but code anything relevant that does not fit them.
3. Organize emerging themes into clusters
Group codes into meaningful clusters and decide how they relate. This is where hierarchy appears: broad higher-order themes with narrower sub-themes beneath them, plus any integrative themes that cut across clusters.
4. Define an initial template
Write the first version of the template based on a subset of transcripts. Brooks et al. distinguish this from approaches that code every account before defining structure. Choose the subset to cover the range of your sample, for example different roles, sites or participant groups.
5. Apply the template and revise it
Code further transcripts with the template. Where data do not fit, modify it: insert a new theme, delete one that is not doing work, redefine its scope, or move it to a different level. Keep each version and a short note on why it changed.
6. Finalize the template and code the full dataset
Stop revising when no substantial section of data relevant to your research questions remains uncoded. Then apply the final version to the whole dataset and use it as the backbone of interpretation and write-up.
A worked example: a template for remote-work interviews
Suppose a research team interviews 30 employees about remote work. The research questions concern autonomy, coordination and career visibility, and the interview guide has a section on each. An initial template built on eight transcripts might look like this:
- Autonomy over work
- 1.1 Control over schedule
- 1.2 Control over workspace
- 1.3 Monitoring and its effects
- 1.3.1 Formal tracking tools
- 1.3.2 Informal "always online" pressure
- Coordination with colleagues
- 2.1 Meeting load
- 2.2 Asynchronous tools
- Career visibility
- 3.1 Access to managers
- 3.2 Promotion concerns
- Integrative theme: trust (runs through autonomy, monitoring and visibility)
When the next ten transcripts are coded, two changes are typical. A theme the guide never anticipated appears, for example caring responsibilities shaping schedule choices, so it is inserted under 1.1. And "Asynchronous tools" turns out to say little on its own, so it is merged into "Meeting load" and renamed "Coordination practices". Each change is logged with the transcript that prompted it, which becomes part of your audit trail.
Template analysis vs other qualitative analysis methods
The table below compares template analysis with three neighbors, based on Brooks et al. (2015), Gale et al. (2013) on the framework method and Braun and Clarke (2021) on reflexive thematic analysis.
| Criterion | Template analysis | Reflexive thematic analysis | Framework analysis | IPA |
|---|
| A priori themes | Allowed, tentative | Generally avoided | Common (analytical framework) | Not used |
| Coding structure | Hierarchical, often 4+ levels | Usually 1 to 2 levels | Framework categories in a matrix | Within-case themes, then cross-case |
| Main output | Final template plus interpretation | Interpretive themes | Case-by-theme matrix | Idiographic account |
| Focus | Across cases | Across cases | Across and within cases | Within case first |
| Typical sample | Medium to large | Small to large | Medium to large | Small, homogeneous |
| Teamwork | Well suited (shared template) | Possible, reflexive emphasis | Well suited | Usually single analyst |
For deeper comparisons, see our guides to framework analysis, thematic analysis, interpretative phenomenological analysis and deductive vs inductive coding.
Template Revision Log: a checklist you can reuse
The weakest point in many template analyses is not the template but the undocumented path to it. This Qualitati-owned checklist turns each revision into an auditable decision.
| Field | What to record | Why it matters |
|---|
| Template version | v1, v2, v3 with date | Shows the analysis evolved with the data |
| Change type | Insert, delete, merge, redefine, move level | Makes the kind of revision comparable over time |
| Trigger | Transcript ID and passage | Grounds each change in data |
| A priori status | Was the theme defined in advance? | Reveals whether prior assumptions survived contact with data |
| Rationale | One or two sentences | Supports reflexivity and peer review |
| Coverage check | Any relevant passages still uncoded? | The stopping rule for finalizing the template |
A useful signal: if no a priori theme was ever redefined or deleted, ask whether the template was really tested against the data or just filled in. Pair the log with an audit trail and a codebook template.
Who template analysis is for, and when not to use it
Use it when: you have defined research questions or a theory to extend; a semi-structured guide shapes the data; the sample is too large for case-by-case idiographic work; or a team needs one shared, explicit coding structure.
Avoid it when: the study is genuinely exploratory and a priori themes would steer you too early; you need an idiographic account of a few lived experiences (IPA fits better); or you aim to build theory from data through constant comparison (grounded theory fits better).
Limitations and methodological concerns
- Anchoring on a priori themes. Starting themes can become a lens that hides disconfirming data. The revision log above is the main safeguard.
- Over-elaborate hierarchies. Deep templates are tempting but can fragment meaning. Depth should follow richness in the data, not a target number of levels.
- Template as output. A finished template is a structure, not a finding. The write-up still needs interpretation, quotes and links back to the research questions.
- Quality claims. Template analysis often uses independent scrutiny of the template by co-researchers, but inter-coder agreement is not automatically required; justify your quality checks against your epistemological position.
Using AI for template analysis
Template analysis maps naturally onto AI-assisted coding because the template is an explicit, written instruction set. Research on deductive coding with LLMs, such as Xiao et al. (2023), combines expert-written codebooks with model prompts and reports useful agreement with human coders on some tasks, while also showing that performance depends on how clearly codes are defined.
A sensible division of labor:
- Researcher: defines a priori themes, builds the initial template on the subset, decides every revision and interprets the final structure.
- AI: applies the current template to remaining transcripts, flags passages that fit no theme (candidates for revision) and retrieves quotes per theme.
- Researcher again: reviews AI-applied codes on a sample before trusting them, and treats unexplained AI codes as prompts for reflection, not findings.
Human review is essential for sensitive methodology claims: report how AI was used, which template version it applied, and how its output was checked.
Where Qualitati fits
Qualitati is a European, budget-friendly qualitative research platform for universities and research companies, with data privacy and GDPR as central priorities. For template analysis, the relevant pieces are:
- QDA Workspace: generate initial codes from your research questions or interview outline (your a priori themes) into a code tree, code manually or with AI-assisted suggestions, apply multiple codes to the same passage, and produce a three-level codebook report (themes, categories, codes).
- ThemeLens: AI thematic analysis across up to 100 transcripts at once, mapping codes to research questions with participant-anchored quotes, useful as a second reading against your template.
- Data collection: AI-moderated interviews in text and voice in 10 languages, so the semi-structured guide that shapes your template also structures the data.
Qualitati does not replace the analytic decisions that make template analysis rigorous; it shortens the mechanical parts so more time goes to revising and interpreting the template.
FAQ
Is template analysis the same as thematic analysis?
It is a style of thematic analysis. Its distinguishing features are permitted a priori themes, an initial template built on a subset of data, iterative revision and a typically deeper hierarchical coding structure.
How many levels should a template have?
There is no fixed number. Brooks et al. (2015) note that templates often reach four or more levels, but depth should reflect where the data are richest, not a quota.
Is template analysis inductive or deductive?
Both. It starts with tentative deductive themes from research questions or theory and then revises them inductively as transcripts are coded.
How is template analysis different from framework analysis?
Both are codebook approaches. Template analysis gives more guidance on developing and revising a hierarchical coding structure, while framework analysis centers on charting data into a case-by-theme matrix.
Can AI do template analysis?
AI can apply an existing template to transcripts and surface data that do not fit, but defining a priori themes, deciding revisions and interpreting the result should stay with the researcher.
Bottom line
Template analysis gives qualitative researchers a disciplined middle path: start with what your research questions tell you, build a template on part of the data, and let the rest of the data rewrite it. Document every revision, keep interpretation human, and use AI for the repetitive application of codes. Ready to try it? Start free with 30 credits, explore the QDA Workspace, or view transparent pricing.
Last updated: October 1, 2026
This guide is an independent editorial summary by the Qualitati Research Team. Methodological claims are attributed to the cited sources; please review them against your discipline's standards before use in a thesis or publication.