How Do I Use Saunders’ Research Onion?

by | Mar 1, 2026

YouTube video

🎯 The Short Answer: The research onion is a planning framework that helps you design your methodology step by step, from big-picture philosophy down to specific data collection techniques. Use it early on to clarify your research design, then adapt those decisions to fit your university’s required structure for the methods chapter.

If you’ve come across Saunders’ research onion and felt a bit overwhelmed, you’re not alone. Many first-time researchers aren’t sure how to actually use it, or how it connects to writing the methodology chapter.

In this post, we’ll break it down in plain English. We’ll look at what the research onion is, how to work through it step by step, and how to turn those decisions into a clear, well-structured methods chapter.

🧅 What Is The Research Onion?

The research onion is a framework created by Saunders and colleagues to help you think through the design of your research methodology. It’s usually shown as a set of concentric circles, like layers of an onion, that you work through from the outside in.

Each layer represents a key decision you need to make. You start with broad, big-picture thinking and gradually move toward very specific choices about data collection and analysis. It’s not a chapter template. It’s a thinking tool to help you design a coherent study.

🌍 Start With Your Research Philosophy

The outer layer of the research onion focuses on your research philosophy. In simple terms, this is about how you believe knowledge is created and understood.

For example, if you believe there is one objective reality that can be measured, you might lean toward a positivist stance. If you believe reality is shaped by people’s experiences and perspectives, you might lean toward interpretivism. At this stage, you’re answering a big question: What assumptions am I making about knowledge and reality?

Don’t panic if this feels abstract. The key is alignment. Your philosophy needs to match your research question. For example, if you’re exploring how employees experience remote work, an interpretivist approach often makes sense because you’re interested in personal meaning.

🔎 Define Your Research Approach

The next layer moves you closer to practical decisions. Here, you choose your research approach. Are you taking a qualitative, quantitative, or mixed methods route?

If you’re testing hypotheses with numerical data, you’re likely in the quantitative camp. If you’re exploring opinions, experiences, or perceptions through interviews or open-ended questions, you’re likely going qualitative. Mixed methods combines both.

This is where we often see students get stuck in our private coaching sessions. They choose a method that sounds impressive, but it doesn’t actually fit their research question. A simple rule of thumb is this: Let your research question drive your approach, not the other way around.

🛠 Choose Your Strategy And Design

As you move further into the research onion, you’ll decide on your research strategy. This is the overall design of your study. For example, will you run an experiment, conduct a survey, carry out a case study, or use in-depth interviews?

This is where your study starts to take shape. If you’re examining customer satisfaction across a large population, a survey might make sense. If you’re analysing one organisation in detail, a case study could be more appropriate.

At this stage, you’ll also clarify things like sampling. Who exactly will you collect data from? How many participants do you need? The goal is to ensure that every design choice supports your research objectives in a logical way.

⏳ Consider Your Time Horizon

The research onion also asks you to think about your time horizon. In plain terms, this means deciding whether your study looks at one point in time or over a longer period.

A cross-sectional study collects data once, like a single survey distributed in March. A longitudinal study collects data over multiple points in time, such as tracking employee engagement over a year.

Be realistic here. Longitudinal studies can be powerful, but they require more time and coordination. If you’re working within tight deadlines, a cross-sectional design is often more practical.

🎯 Get Specific With Techniques

At the centre of the research onion are your techniques and procedures. This is the nitty gritty detail of what you’ll actually do.

For example, if you’re conducting interviews, how long will they be? How will you record them? How will you analyse the data? If you’re running a survey, how will you distribute it and which software will you use to analyse the results?

This is where your methodology becomes concrete. You should be able to describe your process clearly enough that someone else could replicate your study. Specificity builds credibility.

✍️ Turning It Into Your Methods Chapter

Here’s the important part. The research onion is not a template for your methodology chapter. You should not structure your chapter as Philosophy, Approach, Strategy, and so on, unless your university specifically tells you to.

Instead, think of the research onion as a planning exercise you complete early in your dissertation journey. It helps you clarify your decisions and discuss them with your supervisor before you start heavy writing.

When it’s time to write your methods chapter, follow your university’s handbook and look at past dissertations from your department if possible. Use the structure they expect, then plug in the decisions you’ve already made through the research onion process. This way, your chapter will be both well thought out and aligned with institutional expectations.

📌 Key Takeaways

  • The research onion is a step by step framework for designing your methodology, not a chapter template.
  • Work from the outside in, starting with philosophy and ending with specific data collection techniques.
  • Always align your philosophy, approach, and strategy with your research question.
  • Use the research onion early on as a planning and discussion tool with your supervisor.
  • When writing your methods chapter, follow your university’s required structure and insert your design decisions accordingly.

P.S. Have a question? Join our next Live Q&A Session – it’s free!

Don’t stop now…

Which Qualitative Analysis Software Is Best?

Which Qualitative Analysis Software Is Best?

🎯 The Short Answer: If you're working with a large volume of data (50+ interviews or 200+ documents), dedicated software like NVivo or Dedoose is (potentially) worth the investment. For smaller projects, a simple spreadsheet approach works just fine. One of the most...

What Do Dissertation Markers Really Want?

What Do Dissertation Markers Really Want?

🎯 The Short Answer: Dissertation examiners want to see clear, coherent research that demonstrates your methodological competence, strong engagement with existing literature, and honest acknowledgment of your study's limitations. They're looking for evidence that you...

Qualitative Analysis 101: The Big Picture Process

Qualitative Analysis 101: The Big Picture Process

🎯 The Short Answer: Qualitative analysis follows a structured process: collect and record your data, clean and verify your transcripts, code your data to identify patterns, organize codes into themes, and then write up your findings with supporting quotes. If you're...

Too Many Qualitative Codes? Here’s What To Do.

Too Many Qualitative Codes? Here’s What To Do.

🎯 The Short Answer: Start by cleaning up similar codes, then create category layers that nest related codes together. This lets you write about the bigger picture while preserving the detailed nuance underneath. If you're sitting on a hundred or more qualitative codes...

How Do I Choose the Right Statistical Test?

How Do I Choose the Right Statistical Test?

🎯 The Short Answer: Start by reviewing the statistical tests you've already learned, then match your research question to what you're actually trying to accomplish (describe, compare, find relationships, or predict). Different tests do different things, so...

What Does P-Value Actually Mean?

What Does P-Value Actually Mean?

🎯 The Short Answer: A p-value tells you the probability of getting your statistical results if there's actually no real effect or relationship (in other words, by chance). It's not the probability that your hypothesis is true, and it's definitely not a measure of how...