Qual or quant or both?

Which type of research, in which order; a quick guide for designers

Today’s topic - whether to run qualitative research (e.g. interviews) first or quantitative (e.g. surveys). Or the other way round. Or both at the same time in one big bang approach to getting the research done.

Some teams I’ve worked with are cautious with research and want to do one careful step at a time. Others want to plough ahead and chuck everything live at once.

Which is the correct approach? Well, yes.. It depends.

Note: This is for UX designers, Product Designers and Service Designers - all of you will come up against this question in your research planning.

It depends…

The decision to run qualitative then quantitative research, quantitative then qualitative research, or both simultaneously depends not on your team culture, your organisation’s attitude to research or how you feel on any given day but rather on:

• The problem you are investigating

• The type of insights you need or specific goals

• How much you understand about the current problem

• The resources available or constraints you are operating within.

Option 1: Quantitative first, then qualitative

Do this when:

If you already understand the problem space but want to know the size or scale of specific issues or trends first.

Quantitative research is excellent at answering "what" is happening, "where" it’s happening, and "when" it occurs. Also the size of the issues or problem. But less useful for any kind of '“why”.

Because:

Quantitative research is great at identifying large-scale patterns. For example:

• Use surveys to measure customer satisfaction (e.g., "How satisfied are you with the current billing process?" on a scale).

• Use analytics or transactional data to understand trends, such as the percentage of customers choosing different payment methods.

Once you know the size of the ‘what’, you can dig into the ‘why’.

---

---

Why it will work:

• Provides a clear understanding of the size or scale of any given issue

• Helps you choose where to focus your questions when designing qual research such as interview discussion guides

• Useful if stakeholders respond to numbers - as it gives you robust data to secure buy-in for further qualitative research.

Option 2: Qualitative first, then quantitative

Do this when:

If the problem space is less defined or it’s a new area of discovery, then start with qualitative research first. This approach is great for uncovering customer or user attitudes, unmet needs, and emotional drivers.

Because:

If you’re in a new problem space, then you want to be generating hypotheses and identifying themes. For example

• Interviews with customers which uncover how they feel or why they respond in certain ways to product and service experiences

• Observational research, contextual inquiry or shadowing customers in their environments to understand the context of their behaviours

Once you’ve found key themes, follow up quant research can show how widespread the issues are - their scale and importance

Why it will work:

• Provides deep, contextual insights into customer/user behaviours and attitudes that won’t show up in quant.

• Helps provide clarity, helps you refine the problem, identifies unexplored issues or unmet needs

• Generates hypotheses for validation (via quant).

Option 3: Simultaneous qualitative and quantitative research (mixed methods)

Do this when:

This approach is generally used when you have limited time available, and the problem is partially understood, but you need both large-scale validation and deeper insights simultaneously.

Because:

Running both methods at the same time means that quantitative data can reveal broad patterns, while qualitative sessions can provide the context and reasoning behind those patterns.

It allows you to ask ‘what’ and ‘why’ at the same time.

Why it will work:

• Saves time by running both streams of research concurrently

• Provides a balance of measurable findings with in-depth context

• Useful for teams that need quick, actionable insights

• Depending on your timelines, you may be able to iterate between the two in realtime based on findings in flight however…

BUT - big caveat.

By running both at the same time, you remove the ability for one type of research to inform the other.

If the problem space is really unknown, I would always recommend running consecutive research methods to give you the opportunity to either understand scale (quant second) or deep dive into interesting signals appearing at scale (qual second).

Other things to think about when making your decision

The main variables for deciding which of these three approaches to take is as follows: