Image edited

Market research: the ultimate how-to guide

Curious about market research? There’s lots of noise out there on the topic. But no fear, this guide is all you need to get started—plus we’ll give you a behind-the-scenes look at how we do market research at Typeform.

So, you’ve got the next billion dollar idea that will blow the top off your profit margins. You just know you’re onto a winner! Time to throw a huge budget (or your life savings) at this idea, right? Woah, not so fast...

You're not likely to get very far in the marketplace if you only rely on your gut instincts. In fact, recent research from Harvard Business School professor Clayton Christensen states that over 30,000 new products are introduced every year—and a whopping 95 percent fail.

How can you know if your idea even has a chance at surviving in the cutthroat marketplace? Crystal balls sadly went out of fashion a few centuries ago.

The answer: market research. A realistic prediction, based on data, of your chances of success. Basically, it’s a way to find out the market viability of your idea.

If you’re new to market research, don’t be intimidated. This guide will take you from basic concepts through to advanced techniques. Plus our in-house experts will walk you through real life examples of how we do it here at Typeform.

US- fullrender

What is market research and why does it matter?

In a nutshell, market research is the process of collecting information about your target market and customers so you can:

  • Learn who your customers are

  • Find out what they want and/or need

  • Gauge potential market size

  • Discover trends in your industry

  • Get wise on what your competitors are up to

  • Determine how you can stand out

This way you’ll better understand how to serve your customers, prioritize, and get higher returns on your own marketing and product development efforts. Market research is an essential part of any business’s strategy, whatever the size of your company.

Plus, it’s big business, with the global market valued at $75.8 billion in 2019. There are many different ways to approach market research, and at Typeform we’ve developed our own spin on it, thanks to continuous testing, and the insights we get from being a market research tool ourselves (forms and surveys).

In uncertain times, market research helps to reduce the uncertainty

Strategy and business decisions are leaps of faith, but you can at least leap into a situation where you have much higher chances of success. Research is key.

avatar-logoOscar CarbonellDirector of Strategy @ Typeform

Uncertainty is an inevitable part of business, and if 2020 taught us anything, it’s that the world is now more uncertain than ever. However, it’s still possible to reduce some of the uncertainty.

This is where market research is your best ally. Nothing is guaranteed, but making an informed decision based on comprehensive research beats a stab in the dark. Oscar Carbonell, Typeform’s Director of Strategy, has spent many years deep in the heart of market research.

He swears by understanding what your customers want, what the competition is doing, and how to navigate that foggy area in between. Market research helps to reduce the thickness of that fog to see what your options are, and in which direction you might want to take.

You may be competing with people you didn’t even realize were your competition. Market research helps you to maximize your visibility in a rapidly changing, murky marketplace. And most importantly, maximize value to your end customer.

avatar-logoOscar CarbonellDirector of Strategy @ Typeform

Convinced you shouldn’t be sleeping on market research? Great—let’s dive deeper.

Section one: Types of market research

Market research is like water: it is fundamental and remains essentially unchanged, but must adapt to fit the industry, and flow with the context.

Oscar CarbonellOscar CarbonellDirector of Strategy @ Typeform

Finding what works best for you is a must for useful and actionable market research. We don’t believe in a cut-and-paste approach for all businesses and markets, nor in one definitive “right” way to do things. However, there are some basic principles that apply across the board.

Secondary and primary research

Secondary market research delves into information that you don’t create yourself. It’s data that’s already out there, which you can buy or access for free, and is great for benchmarking. Examples: industry reports, census data, research papers, articles in journals or newspapers. Primary market research involves collecting information yourself. This may be more expensive and time-consuming than secondary research, but it’s a better investment in the long run. You focus on your own target audience and gather information directly relevant to your goals. Example: surveys, interviews (face-to-face or over the phone), focus groups, user testing. The ways of doing primary market research can be broken down into two broad categories:

Quantitative and Qualitative

Ahh, the classic “quantitative vs qualitative” dichotomy.

Quantitative market research gathers data that is numerical, descriptive, and structured. You can draw statistics from quantitative research. It involves more of the “what” questions, and can be done at scale.

This type of research is usually carried out through surveys and questionnaires and can be internal or external. Internal quantitative research examines your current customers, while external can help you identify new customers and see the actual distribution of the whole market. External is more likely to be objective, as your own customers already know you and will have formed opinions.


Examples of quantitative questions:

  • “Where do you live?”

  • “How much do you spend on electricity per month, on average?”

  • “Do you use this product?”

  • “How often do you go to the gym?”

  • “On a scale of 1–10, how satisfied are you with our service?”

“On a scale of 1-10, how satisfied are you with our service?"

For more examples, check out this article on survey question types.

Qualitative market research involves more of the “how” and “why” questions. It’s done at a much smaller scale, and is less structured and more exploratory, aiming for insight rather than certainty. It helps you find out how customers feel about your product, their opinions and preferences—in other words, things that cannot be quantified.


Examples of qualitative questions:

  • “Why did you choose product A over product B?”

  • “How does this image make you feel?”

  • “What do you feel is missing from this service?

  • “Describe the last time you purchased something online”

  • “What are your favorite brands for dog grooming products?”

Usually this type of research is done through surveys with open-ended questions, or interviews. A small number of interviews are conducted, which are then projected to apply to a larger population. Dive deeper with this article on qualitative and quantitative research. But quantitative and qualitative research don’t need to be seen as opposite or distinct techniques. It can be an and, instead of an either-or. For example, in the same survey you can ask both quantitative questions (where are you located?) and qualitative (how do you feel about our product?

This hybrid is known as the mixed methods approach, which we’re into at Typeform (read more about the mixed methods approach in our example case study later in this article).

Market research for product (making stuff) and for marketing (selling stuff)

Market research tends to inform two main areas in a business: product development and marketing efforts. Whether it’s creating a new product or a new set of features, at Typeform we always start from the end. Who’s going to use this? Who will buy it? How do I justify engineers spending time on this? Market research is one of the most important tools to answer these questions. Nobody wants to invest time, money, and effort into making something that no one wants or needs. Market research allows you to assess the market size, its opportunities, and your competitors. This is also where user research and market research inform one another.

Likewise, market research and your marketing strategy go together like Netflix and chill. Segmenting the market is one of the main activities in market research, as it gives you your target audience(s) (more on segmentation down below). How else will you know who is buying from you already, who to market to, and which marketing messages work best?

Competitor analysis, another cornerstone of market research, helps you to craft your positioning. In simple terms: how you are different from your competitors, and why buyers should pick you.

Now you’re well-versed in the “what”, let’s move on to the “how”...

Section two: How should I conduct market research?

So you can probably see now how varied market research is. The way we do our own market research here at Typeform has evolved over the years through testing and experimentation. After much trial and error, we finally landed on the approach that works best for us. Curious about what it is? Read on…

First things first: have a goal

Before we even think about launching market research of any scale, we make sure to have a clear objective in mind. Are you trying to enhance a particular metric (such as customer numbers, customer satisfaction level), gauge potential market size, or something else? If you can’t specify a goal that’s clearer than “I wanna find out some stuff”, then you’re probably not ready to start. Define your objective(s) first, then move on to the next step.

The only person that matters

Whatever your approach, the next thing you should always have at the front of your mind is your customer. Oscar explains, "Research done without the end user in mind doesn’t make sense."

There are many different ways of cutting markets and too many companies are still doing it with products and features at the center of everything. You need to segment a market based on customer profiles and needs.

Oscar CarbonellOscar CarbonellDirector of Strategy @ Typeform

Still, focusing on the customer can mean different things to different people. Which brings us to the next point...

Focus on jobs, not personas

Brace yourself, because we’re about to say something controversial.

Don’t focus on buyer personas.

This flies in the face of what most other market research guides will tell you: research your audience to create buyer personas, and frame your offering around them.

Not that buyer personas are not important—they are. And at Typeform, we definitely use them, and other companies use Typeform to create awesome buyer personas. But we also follow the Jobs To Be Done (JTBD) model. This is the backbone for how we conceptualize everything, from our marketing messaging to our product development. It informs how we see our customers, and how we segment them.

The Jobs To Be Done model is neatly summed up by the now famous quote from Harvard Business School professor Theodore Levitt: “People don’t want to buy a quarter-inch drill. They want a quarter-inch hole.” Buyer personas may help you to know where to advertise, but jobs take you to the core of your customers’ motivations.

How many people in your business speak directly to customers? The bigger your organization, the smaller this number is likely to be, and the further removed the customer becomes from the decision-making. The job creates a consistent framework for everyone to work with, and remain close to the customer’s needs.

As you identify needs that intersect, you can begin to find unique differentiators for your product. For example, an obvious job to be done for any Typeform user is to collect information. What differentiates us from all the other form builders out there?

We have identified another job that is important to our users: building relationships. We have a product that not only collects data, but helps our customers build relationships with their customers, through a conversational experience, and a customizable interface. This is a major differentiator for our product. Knowing that this matters to our customers allows us to market our product in a way that matches up to their job.

At the end of the day, your customers don’t care about you or your product or its features. They care about the job or jobs they are trying to get done, and if you provide the best solution they will pay you for it. If you don’t, they will move on to your competition faster than you can say “job to be done.”

The job is the centrepiece of all of our strategy and all of our actions: nothing else makes sense.

Oscar CarbonellOscar CarbonellDirector of Strategy @ Typeform

So how does this all relate to market research?

Rather than framing your market research efforts on creating buyer personas and targeting them, frame them around jobs your customers are trying to get done. There will be some natural overlap with personas, but you need not be wed to them. For more on how we implement the model, check out this article on jobs to be done.

Market segmentation (i.e. different groups for different jobs)

Market segmentation is the act of dividing a target market into groups (or segments). This lets you tailor your efforts to each segment, whether that be your marketing strategy or deciding on features for your product.

How to segment

As mentioned before, it’s all too common that companies segment their market based on products and features, rather than customers. However, there are any number of ways to segment your customers.

Finger market

The four most common methods of segmentation are:

  • Demographic - age, gender, ethnicity, income, industry, job

  • Psychographic - lifestyle, values, personality traits, interests

  • Geographic - country, region, city, town

  • Behavioral - spending habits, internet browsing habits

Depending on your situation, any of these might be useful focus points, and all of them no doubt provide valuable insight. However, as previously mentioned, our main method at Typeform for segmenting customers is the Jobs To Be Done model. Once you get clear on the different jobs your customers want to accomplish, you can segment them on this basis, and target accordingly.

The benefits of segmentation

  1. A better experience for customers: better understanding of your customers can only really be a win-win. You’ll be able to tailor each part of your customer’s experience, from marketing message to product experience, based on their segment.

  2. More targeted marketing: in other words: better use of your marketing resources. Rather than casting the net wide and crossing your fingers that you haven’t just thrown a lot of time and money away, your segments let you focus your efforts where they’re likely to have the most return.

  3. Improved product development: knowing the real demands of your target audience will allow for product development that they will actually appreciate (read: pay for).

Section three: Planning market research

Now that you’re convinced of the importance of market research and how it can help your business, you’re probably pumped to get started. Having even a basic plan can be the difference between a piece of research that has a real and lasting impact on your business, and gathering some interesting insights that are forgotten in two weeks. Always start with the question: why? What’s the purpose of the research? Your objective shouldn’t be “to do some research”, nor should you select a method first, whether that be a JTBD-based questionnaire, customer interviews, etc. Make sure you’re always starting with a question you want to answer, and adapt the method to the question.


For example

  • “How can we increase conversions?”

  • “Why are people churning after two months?”

  • “What is the appetite for this product?”

  • “Which product features are most useful to our customers?”

  • “In which region(s) should we focus our next marketing campaign?"

Let this always be front and center as you go about planning and executing your research.

Quick-start tips

Sometimes we just need some professional help.

Cate Antinarelli is a Senior Business Analyst and seasoned market researcher at Typeform. She’s got both the academic chops and extensive hands-on experience in market research. Here are her expert top tips for getting started with your planning:

Do preliminary research: have a basic understanding of the industry and the landscape you’ll be investigating. It doesn't have to be extremely in-depth, but it’s important to have a foundation. This ensures you ask the right questions, know what to assess, and can get a more accurate vision of the market.

Align with potential stakeholders: there may be others in your organization who could benefit from the data you are about to gather. It may be worthwhile checking around to see how you could maximise your research efforts. Even just one extra question on your survey might provide essential data for someone else.

Use the right tools for your research purposes: make sure that whichever tools you use are fit for purpose. Using the right tools will not only save you lots of time and energy, it is essential for correct and high-quality data.

For example, Typeforms have a Logic Jump feature which allows you to qualify respondents. If you only wanted responses from people located in the USA, you can ask this at the beginning of the survey. With a Logic Jump, anybody who matches this criteria will be moved onto the rest of the survey, and all others will be sent straight to a thank you screen.

We’ve put together a handy list of market research tools at the end of this page.

How you ask is everything

The questions you ask depend on your objectives. Here’s a handy guide on how to write qualitative market research questions that are meaningful. And if you’re hungry for more, here are some tips on market research survey questions that will help strengthen your relationship with your customers. Best of all, you can hit the ground running with this template for market research readymade and good to go—just customize to your heart’s content. It’s also a great example of what a market research survey might look like.

Cate also recommends you consider running a test first, depending on the scale of your research. Sending your survey to a smaller population and analyzing the first few responses will let you check that you’re getting useful responses that are answering your research questions.

Sometimes, until we start getting results, we’re unaware that a question is ineffective. This may be because the question uses terminology not understood by the target audience. For example, you may ask “what SaaS tools do you currently use?” If you get responses like “iPhone 11” and “desktop computer”, then you know you need to adapt your questions better to your audience! Here at Typeform we sometimes send out test emails to smaller populations (around 10% of the target audience) for this purpose, and adjust our surveys if necessary.

Getting the right number of respondents

400 is the magic number.

Well no, in actual fact there is no magic number, sorry.

Generally speaking, 400 is the standard recommended sample size—this just means the number of people who responded to your market research survey. But this number can vary greatly depending on your total population (i.e. all the people that this research will apply to) and the way you segment them. But there’s a mathematical explanation for the popularity of 400: with 400 responses your margin of error is 5%. For example, say you got 400 customer responses to your market research survey. 80% of your respondents answered “yes” to the question “would you buy from us again?” That means there’s a 95% chance that in your total population of customers, around 80% would buy from you again. Here at Typeform we tend to aim for 1,000 responses with our market research surveys. This is because we then segment responses into Jobs to Be Done, aiming for at least 100 respondents per job. Confused? To simplify things, here’s a sample size calculator that will do the hard work for you. Don’t forget that in order to reach your target sample size, you will need to reach out to many more people! If sending out surveys by email, open rates tend to hover around 15-25%. The percentage of people who then go on to complete a survey will be even lower. To increase your chances of opens and completions, offering an incentive is never a bad idea. Prize draws or discounts on your product work well for us. And of course the experience of answering a market research survey is paramount for completions: ensure your form is user-friendly with a smooth and beautiful interface. And nobody does a beautiful form like Typeform! One last thing! Before you hit SEND, Cate would like to gently remind you to aim for a sample that will be a good approximation of your overall population. There’s a risk of bias, depending on the channel through which your research survey is shared. For example, if you share it on social media, you might get a younger average age of respondent, which may not be accurately representative of your total population of customers.


Challenge your approach

So you’ve got an objective and a plan, and you’re raring to go! Just one last thing before you get started: have you checked what data you already have? Nobody likes duplicate work. There’s a chance that you may be sitting on valuable data that you could use to answer your questions, even if partially. And always: are we clear on what decision we’ll take based on the results of this research? Yes to all of the above? Then let’s get ready to rumble.

Ready-to-use plan outline

Below is a simple outline, based on the one we use at Typeform, for planning a piece of primary market research.



A brief summary of why this research was started:

  • What led to this research being done/requested? What needs to be validated or explored?

  • What has been done prior to this research? E.g. competitive analysis, brainstorming, previous research

  • What insights will this research generate? How will these insights be used?

Group 1731

Business/Product Objectives

We cannot emphasize enough the importance of having a clear goal in mind. What metric(s) are you trying to enhance? E.g. more conversions, less churn. This helps people understand the bigger picture of this research.

Group 1731


State what decisions are going to be made or impacted based on the research. As a general rule, if you’re not prepared to make changes, don’t run the research.

Group 1733

Research Objectives

State the high-level objectives for this research. Try to keep it specific, actionable, and max. 2–3 points.

Group 1736

Research Questions

Provide a list of questions that you plan to answer during this research (these questions are not the interview questions).

Group 1739

Participant Criteria

List the primary characteristics of the people you will recruit for the research, e.g:

  • Geography

  • Job(s) to be done

  • Industry

  • Job title

Also decide on the minimum and maximum number of participants you will need for your study.

Section four: Execution (how to use market research for success)

You’ve got the theories, the models, the key terms and even a helpful template to get you started. How about doing the darned thing?

By now you may have a good idea of the type of market research you want to do, whether it’s a big quantitative survey sent out to thousands of people, or a handful of interviews with your most loyal customers.

However, if you are still unsure, fear not. We’ll walk you through a couple of case studies from our own market research efforts, to show how it can be done. There are a million and one ways to conduct market research, but let these serve as illustrative examples.

Market research example: doubling our SaaS customers in three months

Here’s a case study of how market research helped us reach a concrete goal in just a few short months, using a mixed methods approach. Jahed Momand, Head of Research at Typeform, headed it up, with pretty awesome results.


Optimize customer journeys to double the number of new SaaS customers in one quarter

How did we do it?

  1. Diving into the data

We started by mining our existing customer data to identify those customers who were the most “valuable”—i.e. those who spent the most, stayed the longest, and were most engaged with our product. We found that these customers tended to be Software as a Service (SaaS) companies, of a certain size, in the English-speaking world. With a bit of research (i.e. customer interviews), we knew we could find out how they’d heard of us, their motivations, and their jobs to be done. Then we could figure out how to improve customer journeys, and increase the number of new customers with a similar profile.

2. Deciding who to interview

In other words: better use of your marketing resources. Rather than casting the net wide and crossing your fingers that you haven’t just thrown a lot of time and money away, your segments let you focus your efforts where they’re likely to have the most return. The data let us refine the parameters for deciding who we wanted to interview, e.g. the date of sign up. This was all data that we already had. With this quantitative analysis done, we could get down to the qualitative interviews. We reached out to a subset of customers for interviews and conducted around 15 (online because, well, 2020). We asked questions such as:

  • How did they hear about us?

  • What did they do on the site?

  • What did they fail at doing in the early parts of their journey?

  • What was the first form they launched?

  • How long did it take them to receive responses?

  • When did they think that Typeform was for them?

In other words, things that cannot be answered with quantitative data.

3. Detecting patterns

Once we finished the customer interviews, it was time to do some affinity mapping (read more on affinity mapping here). Jahed and two designers in the marketing team spent two days going through all of the qualitative data, tagging it and aligning it with set parameters: jobs to be done was the main one, naturally. So, being led by the customers’ input (rather than our own assumptions) we began to look for patterns. Were they clustering around certain jobs? We found lots of similarities and contrasts, but each of them presented opportunities. A key finding was that people found us most through:

  • Word of mouth

  • Social media

4. Testing hypotheses

We don’t take the word of a small number of customers and accept that as truth, even if they all agree. Instead, we generate hypotheses from our findings, and test them. For instance, social media seemed to be a good channel for our customers, so we investigated further, e.g. by looking at the last ten social media ads that our interviewees clicked (with their permission of course!). This gave us more clues. Qualitative research is good for generating ideas, which we can then test against millions of people at scale. So in this case we were able to generate, test, and optimize much more effective social media ads. The ads were created with the help of the affinity mapping from the research, and tested on people who matched the profiles of our target customers.

The end result?

We doubled our number of new customers in a quarter.

Section five: Taking action on market research insights

Remember: data is not reality, says Professor Jill Walker Rettberg from the University of Bergen in Norway. However, market research can give you a pretty decent proxy of reality.

Another gentle reminder from Cate: data can be unpredictable. Missing a small detail can skew the results significantly, so try to be as methodical and meticulous as you can.

Here are Cate’s top tips on what to do once you’ve gathered all that juicy data:

  • Clean it up: Your data needs a spring clean before you start drawing conclusions. Nix the incomplete surveys to avoid different sample sizes for different questions. It’s also good practice to check for extreme outliers which might create bias and skew your averages. It’s worth putting the data into a table (an application like Google Sheets would do just fine) to check for any outliers.

  • Check it out: As you’ve already put your data in a spreadsheet from the previous step, you might as well make the most of it. Having data laid out like this will enable you to look at distribution and more easily spot any patterns.

  • Cut it down: This is where our friend segmentation comes back into play. At this stage you might have a lot of raw data on your hands. Now you can start slicing it up, and segmenting in order to address your question.

  • Carry it out: Not to sound like a broken record, but we’re going to reiterate: data isn’t there to just look pretty—it should drive action. Once you’ve drawn conclusions from your research, it’s time to think of actions you will take based on your findings.

Remember: market research is not just for big companies. Anyone can do it. You are doing research when going on a date, meeting friends, buying on Amazon—market research is part of our day-to-day. Your customers are doing market research all the time.

And finally...

Here are some words of wisdom from Oscar to gas you up before you go:

Don't be scared of never having done it before or not being able to afford an expert. It's just about having a hypothesis and testing it. It could be as simple as asking: do you like this? You're the expert. Nobody knows your customers better than you.

Oscar CarbonellOscar CarbonellDirector of Strategy @ Typeform