Step-by-step guide
Grounded Theory Steps: How to Do It
Grounded theory is more iterative than linear, but it does have a recognisable sequence of steps. This guide walks through how to do grounded theory step by step — from your first round of coding to a saturated core category — and the documentation each step needs.
Remember the defining feature: data collection and analysis happen together. You do not collect everything first and then analyse; you analyse early, and that shapes what you collect next.
The grounded theory steps
- 1Collect initial data — begin with a broad sample relevant to your area of interest — often a small number of interviews or observations to start.
- 2Open coding — code the data line-by-line or incident-by-incident, generating many initial concepts that stay close to the data.
- 3Constant comparison — compare each new incident against existing codes and categories, refining and merging as patterns appear.
- 4Memo-writing — write analytic memos capturing your developing ideas about categories and their relationships — these become the substance of your theory.
- 5Theoretical sampling — let the emerging analysis decide what data to collect next, to develop and test your categories.
- 6Axial and selective coding — relate categories to one another, then integrate them around a central core category.
- 7Reach theoretical saturation — continue sampling and comparing until new data adds no new properties to your categories.
- 8Write the theory — present the integrated, explanatory account grounded in your categories and supported by data and memos.
How to document rigour at each step
Grounded theory examiners look for evidence that you actually did constant comparison and theoretical sampling — not just claimed them. Keep memos dated and linked to the data that prompted them, keep an audit trail of how categories developed and merged, and be explicit about how you decided saturation was reached.
QualIntel OS supports the iterative coding and keeps a timestamped audit trail of how each category developed, so the comparative process at the heart of grounded theory rigour is documented as you work — and it generates the AI disclosure statement if you used AI assistance.
Frequently asked questions
What are the steps of grounded theory?
Collect initial data; open code it into concepts; use constant comparison to refine categories; write analytic memos; use theoretical sampling to guide further data collection; move to axial and selective coding to integrate categories around a core category; continue until theoretical saturation; then write the integrated theory. The steps are iterative — data collection and analysis run together.
How do you start a grounded theory study?
Start with a broad area of interest rather than a fixed hypothesis, collect an initial round of data (often a few interviews or observations), and begin open coding immediately. Your early analysis then guides theoretical sampling — what data to collect next. Avoid imposing a detailed framework up front, especially in classic grounded theory.
What is constant comparison in grounded theory?
Constant comparison is the technique of continually comparing new data and incidents against your existing codes and categories, refining them as you go. It is what keeps the developing theory grounded in the data and is one of the defining methods of grounded theory. Examiners look for evidence you genuinely did it.
How do you know when to stop in grounded theory?
You stop when you reach theoretical saturation — the point at which collecting and analysing more data adds no new properties or insights to your categories. Demonstrating saturation (rather than just asserting it) is part of grounded theory rigour; an audit trail of how categories stabilised supports the claim.
Related guides
Build grounded theory in QualIntel OS
Iterative, comparative coding with an audit trail of how every category developed — and AI disclosure generated as you go. One project free, no card.
Start for free