Which reflective model should you use? Gibbs, Rolfe, Driscoll and Johns compared
Reflective practice runs through CPD and revalidation for regulated clinicians in the UK, but the model you use to structure it is up to you. If you have ever stared at a half-written reflective account wondering whether you are "doing Gibbs properly", this guide is for you. It sets out the main reflective models, how they differ, and how to choose one that fits the event you are reflecting on and the registration you are meeting.
What the HCPC and NMC actually require
Neither regulator makes you use a particular model. The HCPC is clear that there is no set way to reflect, and no set way to record your reflections, so you can adapt the activity to your role and the way you learn. You can read this in the HCPC reflective practice FAQs.
The NMC asks for five written reflective accounts across your three-year revalidation cycle, recorded on its own reflective accounts form and each linked to the Code. The form comes with its own prompts, so you do not have to layer an academic model on top unless you find it helpful.
In other words, a reflective model is a scaffold, not a rule. Its only job is to stop you describing what happened and never getting to what you learned.
Gibbs reflective cycle: structured and thorough
Graham Gibbs set out his six-stage cycle in 1988. It moves through description, feelings, evaluation, analysis, conclusion and action plan. Because it gives feelings their own stage, it suits emotionally significant events and anyone who wants a clear, step-by-step structure. The trade-off is length. A full Gibbs reflection can run long, which is more than a small learning point needs.
What? So what? Now what? A fast, practical model
This three-question structure is the quickest way to turn an event into a learning point. You describe the event (What?), make sense of it (So what?), then decide what you will change (Now what?).
The three questions originated with Terry Borton in 1970. John Driscoll adapted them for clinical practice, developing his version through 1994, 2000 and 2007, and Gary Rolfe, Dawn Freshwater and Melanie Jasper published their own version in 2001. The two are very similar in practice, and the differences lie mainly in the prompt questions each author suggests under the three headings. For a short account, or when you reflect often and want a repeatable habit, this is hard to beat.
Johns' model of structured reflection: depth and influences
Christopher Johns developed his Model of Structured Reflection for nursing from the mid-1990s, refining it in 2000 and 2004. It uses a set of cue questions and draws on Barbara Carper's patterns of knowing, prompting you to look at what you knew, what you felt, the ethics of your decision and the factors that shaped it. It is more searching than the other models, which makes it well suited to complex cases, ethical dilemmas, or any situation where you want to understand why you acted as you did. It takes longer, so it is overkill for a routine learning point.
Schön: reflecting in the moment and after
Donald Schön (1983) is less a template and more a way of naming two kinds of reflection. Reflection-in-action is the thinking and adapting you do while an event is still unfolding. Reflection-on-action is the deliberate review you do afterwards. Most written reflective accounts are reflection-on-action, but the in-the-moment version is worth naming, because a great deal of clinical skill lives there. You can pair Schön's idea with any of the structured models above.
Kolb: reflection as part of a learning cycle
David Kolb's experiential learning cycle (1984) frames reflection as one stage in a loop: concrete experience, reflective observation, abstract conceptualisation and active experimentation. In plain terms, you do something, think about it, work out the general principle, then try the new approach. It is a useful reminder that a reflection is not really finished until you have applied the learning and can look again at how the change went.
How to choose the right model for your event
You do not need a single favourite. Match the model to the event.
- For a quick, everyday learning point, use What? So what? Now what?
- For an emotionally significant or difficult event, use Gibbs, because it gives feelings a dedicated stage.
- For a complex or ethically tricky case, use Johns, because it makes you examine influences and knowledge.
- For a change you have already made and want to track, use Kolb, because it builds in trying the new approach.
- Whichever you pick, keep Schön in mind as a way to value the thinking you did in the moment.
One practical tip: pick the model before you start writing, not halfway through. Switching models mid-account is how reflections end up repetitive and unfocused.
A worked example
Here is a short event structured with What? So what? Now what?, using a role rather than a name and with no identifying detail.
What? On a busy shift I handed a deteriorating patient over to the receiving team and realised, part way through, that I had not clearly stated the escalation I had already requested. So what? The receiving clinician had to ask me twice, which cost a couple of minutes and left me flustered. Looking back, I had led with the history rather than the current concern, so the most urgent information came last. Now what? I will use a structured handover tool every time, leading with the current problem and the action needed. I have re-read my service's handover guidance and will ask a colleague for feedback on my next few handovers.
That is a complete, useful reflection in three short paragraphs. A larger event might call for Gibbs or Johns, but the principle is the same. Get past describing, and land on a specific change.
Looking after yourself when the event was hard
Some events are worth reflecting on precisely because they were distressing. Reflection should help you learn and move forward, not keep you turning the same moment over. If you notice you are ruminating rather than reflecting, treat that as a signal to pause and reach for support: your occupational health service, your union, or the Samaritans on 116 123, which is free to call at any time. Reflect on what you can change, and be fair to yourself about what you could not.
Let Reflectory do the structuring
If choosing and applying a model is the part that stalls you, Reflectory can do it for you. It interviews you about your event, in your own words, and produces a reflective account you can add to your HCPC or NMC portfolio, with identifiable details screened out and the use of AI assistance disclosed. You choose the event and you stay the author. You can try your first reflection free at reflectory.co.uk.
Frequently asked questions
Does the HCPC or NMC require a specific reflective model?
No. The HCPC states there is no set way to reflect or to record your reflections, and the NMC asks you to use its reflective accounts form for your five written accounts but does not require a particular academic model. Choose whatever helps you reflect honestly and reach a clear learning point.
What is the difference between Rolfe's and Driscoll's reflective models?
Both use the same three questions, What? So what? Now what?, which originated with Terry Borton in 1970. John Driscoll adapted them for clinical practice through 1994, 2000 and 2007, and Rolfe, Freshwater and Jasper published their version in 2001. In practice they are very similar, and the main differences are in the prompt questions each one suggests.
Which reflective model is best for a short reflection?
For a quick learning point, the three-question What? So what? Now what? structure is usually fastest. For an emotionally difficult or complex event, a fuller model such as Gibbs or Johns gives you more structure to work through feelings and influences.