Quantitative data deals with hard numbers, while qualitative data reveals insights from users’ subjective experiences. Quantitative data is generally used to find out what is happening in your product or site, while qualitative data is good for telling you why.
In SaaS, good data is the holy grail when it comes to sound decision-making. But how do you know which data to use, and when?
In this guide, we’ll introduce you to the two main types of data—qualitative data and quantitative data—and help you understand the difference between them. You’ll also learn how to make the two work together.
In brief, quantitative data is “hard” data—data that can be captured by a number, and measured exactly. Qualitative data is “soft” data—data that’s conveyed through language, story, and subjective experiences. Below is a table that briefly outlines the key differences:
Want to upgrade your quantitative analytics with qualitative insights? Check out this on-demand webinar!
What Is Quantitative Data?
As mentioned above, quantitative data is helpful for teams who are trying to put a hard number on something—that is, when they’re trying to figure out “how often,” “how much,” or “how many.”
Quantitative numerical data sets can often be displayed in easy-to-digest graphs. They can help you find correlations you weren’t previously aware of, give you statistics to base larger generalizations on, and help you determine where to invest your team’s resources. Quantitative data is usually gathered through analytics tools, experiments, polls/surveys, controlled observations, and other methods.
When it comes to SaaS, examples of quantitative data metrics you might want to measure are customer churn rate, natural rate of growth, and burn multiple, among others. Metrics like this can be calculated by looking at user data, which can easily be tracked, counted, and put into context with tools like Looker, Tableau, and Microsoft Power BI. Other web and mobile analytics tools built for product teams, including Heap, offer behavioral data that quantify clicks, drop-offs, and other actions users take on web or mobile.
With quantitative research, sample sizes are also important to consider. The higher your sample size is, the higher your chances are of finding statistically significant information.
What Is Qualitative Data?
Qualitative data generally means descriptions expressed in words, not numbers. It can be gathered through methodologies like structured or unstructured interviews, user observation, or written-response surveys. In general, qualitative data is focused on answering “why” and “how” questions—something quantitative data can’t do.
Depending on the stage of your SaaS company or feature-development process, qualitative data might actually be the most appropriate type of data to use. This is because it can give you valuable insights that quantitative data might not.
For example, quantitative data can tell you how many users are leaving a specific user flow, or aren’t renewing their contract. Qualitative data like interviews or Session Replays can show you why—what specific problems users have, what workflows they use when using your software, where they get stuck, what users find confusing or don't see, how they use your software with other tools, and a host of other pieces of useful info.
This kind of information tends to be invaluable for making design decisions, or figuring out what kinds of messaging works best with different users.
Qualitative data also lays the groundwork for grounded theory, a qualitative research method approach sometimes used by data analysts to build theories based on ideas or concepts that that “emerge” from qualitative data.
What Are the Different Types of Quantitative and Qualitative Data?
While numerical data and non-numerical data each have an established place in data analysis at SaaS companies, both types are essential to give product teams a robust picture. Below, we’ll break down some different types of quantitative and qualitative data that are relevant to product analytics.
Types of Quantitative Data
Quantitative data can generally be broken down into two types: discrete and continuous.
Discrete data is information that is expressed through concrete numbers with fixed, specific values. While it doesn’t always require whole numbers, it always consists of distinct or unconnected elements.
Some examples of discrete data in SaaS might be:
The number of users who downloaded your app
The number of employees in a company
The number of people who respond “yes” instead of “no” on a survey
There are limited ways discrete data can be expressed—bar graphs, pie charts, or tally charts often work best.
In SaaS, discrete data is essential, whether you’re measuring conversions, how many users pay for subscriptions, or how many users used a new feature you recently introduced. The variables are endless.
Unlike discrete data, continuous data can take any value. It can also vary over time, while being measured (instead of counted) on a scale or continuum. Continuous data can also be meaningly divided into smaller parts, depending on how precise you want to be.
A few examples of continuous data include:
How many sales a company has in a year
Duration of customer service calls
Metrics like daily active users, load time/latency, user growth rate, customer acquisition cost (CAC), and others
Since continuous data changes over time, it is often displayed on a line graph, histogram, or scatter plot. If points of measurement can be broken in half and still make sense, your data is continuous.
For SaaS companies, continuous data can also be used to measure efficiency in customer service, marketing, and other business departments.
Types of Qualitative Data
There are two key types of qualitative data that you should keep in mind when doing qualitative data analysis: ordinal and nominal. Both types are important for building a roadmap and improving your product.
Like discrete and continuous quantitative data, ordinal data has to do with numbers. But a key difference is that these numbers don’t add up to anything. Instead, they refer to a certain position on a scale. You can’t do math with ordinal data, because the numbers only make sense in context of their relative position.
A few examples of ordinal data include:
Letter grades (like A, B, C, D, F)
Satisfaction survey ratings
Search engine rankings
A key example of ordinal data in SaaS companies is NPS or CSAT ratings, which both measure customer experience within different parameters on a scale of 1-10. It can also be used to gauge user interest in potential new product features or gather user rankings of what key features they consider to be most important.
Nominal data is used to put labels on items that have no order or quantitative value. While it can’t do numerical tasks, nominal data is helpful for categorizing information that is part of the story or situation you’re investigating.
A few examples of nominal data include:
User demographics (nationality, gender, marital status, etc.)
Customer satisfaction survey responses
SaaS companies can gather and use nominal data for market segmentation, cohort building, identifying target audiences, and many other qualitative data analysis processes. When you can start building user categories and better understand how customers feel about your product, you can better understand how you can improve on the value you offer them.
Tools for Collecting Quantitative and Qualitative Data
So, how do you gather qualitative and quantitative data? There are quite a few ways. Whether you’re asking individuals open-ended questions or measuring how many users clicked on your CTA, the way you gather data can dictate what type of analysis methods you use afterward.
Tools for Collecting Quantitative Data
Quantitative data collection methods can vary, but generally, you’ll gather it using one of the following methods:
Internal and external analytics tools. These can be platforms like Google Analytics, Heap, or Looker. They’ll show you raw numbers and, if they’re robust, specific patterns and insights you can glean from those numbers.
Experiments and observations. Whether you’re running an A/B test or taking note of how many users are behaving a certain way, experiments and observations are reliable quantitative research methods that can give you key insights into how your product is performing.
Database reports. These can give your cold, hard numbers a bit of structure, making it easier for leaders in your business to easily digest the information so they can make sound decisions based on it.
Industry reports. While relying on data from external sources will require fact checking (like with market research, for example), industry reports can give you a fast, broad look at relevant information for your SaaS business.
Surveys. In quantitative user research, answers to survey questions will have a numerical value. Surveys are cost-effective, flexible with sample size, and great if you want a real-time pulse on information that matters to your team.
Tools for Collecting Qualitative Data
Qualitative data collection methods also vary, as do the ways to analyze your qualitative research data. They typically include one or more of the following:
Session replays. These help SaaS companies get an instant replay of any given user’s journey on their website or app. Understanding customers’ actions can give you important insights into what parts of your product work well and what parts don’t.
Heatmaps. If you’re in the SaaS industry, heatmaps will show you visually what your collective user base is doing most often in your product. While heatmap tools can validate design decisions that are working, they can also reveal behaviors you weren’t anticipating. This can pave the way for new product improvements, features, and strategies.
User or prospect Interviews. Again, in-depth interviews can be structured or unstructured. You can talk with users of your product and gain insights based on the experiences and thoughts they share.
Case studies. These are detailed studies of a particular subject in a real-world context. They allow teams to explore the depths of a product, process, event program, activity, person, or group of individuals.
Focus groups. This method of qualitative data collection involves bringing together a small group of people in a moderated setting to answer a set of research questions. It can give you insights into customer decision-making, perception of product quality, and other key aspects of people’s subjective experiences.
Questionnaires. Also known as qualitative surveys, this method involves asking open-ended questions and collecting written answers from respondents. Questionnaires can be a great precursor for interviews, focus groups, or case studies, allowing you to identify issues or themes for further exploration.
How Do You Know When to Use Quantitative vs Qualitative Data?
It can be challenging for data newbies to know when to use qualitative vs. quantitative research to gain insights into what users really need. When you consider the differences between the two types of data, however, the value of each—or the value of using a mixed methods approach—becomes clearer.
Use quantitative data to confirm a test or hypothesis
If quantitative data is a key part of your project strategy, you’re likely looking for many—perhaps even millions—of objective data points surrounding a particular angle of inquiry, because you’re trying to gain statistically significant facts from the final numbers.
Often, you’ll want to run tests (such as A/B tests) to prove or disprove certain hypotheses about your product or how people are using it. This structured approach to quantitative research offers a clear path toward micro- or macro-level goals, which can vary depending on the scale of your SaaS company or the stage it’s in.
Use qualitative data to understand concepts and experiences
Qualitative data, by contrast, is for when you truly care about what individual users have to say. It will enlighten you about the real stories behind the quantitative data—for SaaS companies, this means customers’ subjective experiences, habits, and opinions.
There’s no way to hit statistical significance with qualitative research data, because statistics are not the goal. Meaningful patterns and nuances are, and they’ll sometimes help you build new product narratives that you won’t be able to glean from quantitative data.
Use a mix of both
Because both quantitative data and qualitative data can be extremely valuable, a mixed method analytics approach can help you build the most coherent narrative about your service or product. Platforms like Heap that offer a mixed method solution are great for SaaS companies, because they can help leaders make sound decisions with product design, operations, and marketing.
How to Analyze Quantitative and Qualitative Data
Now comes the actual quantitative and qualitative data analysis! To do this, you can take a few different approaches, which we’ll outline at a high level below.
Analyzing Quantitative Data
When you’re doing quantitative data analysis, it’s important to get a clear understanding of what the numbers are telling you. To do this, you’ll likely use both descriptive statistics and inferential statistics.
Descriptive statistics describes or summarizes the characteristics of a data set. It’s the most common and fundamental form of data analysis, and it can be used at every level of a SaaS company, including measuring product metrics. Below are key terms you’ll need to be aware of:
Mean. This is the mathematical average of a set of two or more numbers, calculated by adding all the numbers together and dividing the total by the number of figures within the set.
Median. This refers to the middle value in a set of numbers. It can be calculated by ordering all numbers in a set from lowest to highest (or vice versa) and finding the middle number. It is easiest when you have an odd number of figures!
Mode. This refers to the value that occurs most frequently in a data set.
Percentage. This is calculated by 1) counting a subset of items in a group; 2) dividing that number by the total number of items in a group; and 3) multiplying the result by 100.
Frequency. In statistics, this refers to the number of times an event occurs.
Range. This refers to the difference between the smallest and largest values of a numeric variable, calculated by subtracting the minimum from the maximum.
Standard deviation. This is a measure of how dispersed data is in relation to the mean. Low standard deviations are dispersed closely to the mean, while high standard deviations are more spread out.
Skewness. This is a measure that describes the shape or symmetry of the distribution of a data set. If the distribution is normal and perfectly symmetrical, the skewness is zero. If it’s high, the distribution is very asymmetrical.
Inferential statistics builds upon descriptive statistics by taking the data gathered and making inferences and predictions about what a more extensive representation of that data might be. With inferential statistics, you’ll use:
Cross tabulations. These can help you quantitatively analyze the relationship between multiple variables in a table—including relationships that may not immediately be apparent—and are often used on categorical data (data that can be divided into mutually exclusive groups). A pivot chart is an example of cross tabulation.
Regression analysis. Usually shown in a graph, this refers to a set of statistical processes that help you estimate the relationship between variables in a data set—particularly dependent variables as they relate to independent variables. It can help you figure out the strength of the relationship between variables and model future relationships between them.
Cohort analysis. This is a type of behavior analysis that groups data sets into related categories before doing the analysis. For SaaS companies, it’s a great way to identify and analyze behaviors of users with common characteristics.
Factor analysis. Also known as data reduction, this technique reduces a set of variables by grouping common characteristics under one umbrella. It’s another great way to identify relationships in data that might not immediately be apparent.
Analysis of variance (a.k.a. ANOVA). This is a formula that lets you calculate variances across averages of different groups. It helps you identify statistically significant differences when comparing groups.
MaxDiff analysis. This survey-based data-gathering technique is used to quantify preferences, often in market research. For SaaS companies, users might be surveyed about what features are most and least important to them, rated by a number.
Time series analysis. This helps you analyze a sequence of data points collected at consistent intervals over time. It’s useful, because it shows how variables can change over time, helping you better predict future events.
Analyzing Qualitative Data
If you’re a SaaS company taking a qualitative approach, there are a few key methods of digging into the information once you’ve gathered it:
Pattern analysis. In SaaS, this method helps teams uncover valuable insights that can inform product development, marketing, and customer engagement strategies. It allows businesses to recognize and respond to users’ behavior patterns, needs, and preferences. Three main types of pattern analyses are:
Demographic. This involves examining data related to certain characteristics of users, including age group, gender, income levels, and more. This can help SaaS companies tailor marketing strategies or product features to specific types of users.
Geographic. This is helpful for SaaS companies that want to understand regional preferences, adoption rates, and other behaviors. It can help inform decisions about expansion, localization, sales strategies, and more.
Psychographic. This relates to the study of user behavior, attitudes, and preferences related to SaaS products. It can help you understand motivations and preferences of users and how they influence choices users make.
Content analysis. This technique helps you determine the presence of certain words, themes, or concepts in qualitative data like textual, video, or audio content. This can help SaaS companies monitor user feedback and identify areas of improvement.
Thematic analysis. This is usually applied to a set of texts and refers to the process of reading it and identifying concepts, themes, and topics that come up repeatedly.
Narrative analysis. This is the process of interpreting core personal stories gathered from people in a particular study group. It also examines how a story is conveyed in an effort to deduce how a user experienced something.
Limitations of Quantitative and Qualitative Data
While quantitative data and qualitative data are both great for SaaS companies, they each have their limitations.
Quantitative data, for example, doesn’t have the power to give you a solid understanding of the depth or context of the situation(s) that generated it. All you have are the numbers. While they can tell you a lot, they don’t tell you everything!
Qualitative data, conversely, doesn’t have numbers to back it up. It can be biased, and a team member’s skill level and experience can potentially influence their interpretation of it. Unlike quantitative data, qualitative data also isn’t easily scalable. It often has to be waded through by individual team members, which can be costly and time-consuming.
Explore Your Data with Heap
When you’re looking for third-party analytics tools to help you auto capture user behavioral data (and make sense of it!), choosing one that incorporates a qualitative layer on top of a quantitative layer can be very helpful. Heap is just such a tool, as it records every single action a user takes on your web or mobile app before helping you interpret the data, build case studies, and bolster qualitative data with contextual quantitative information.
Interested in a demo of Heap’s Product Analytics platform? We’d love to chat with you!