Qualitative and quantitative data analysisJul 07, 2021
By Nick Bottai - Co-Founder/Director @ The Marketing Leaders Ltd.
1. Quantitative Data
2. Qualitative Data
3. Case Study
5. Sharing is Growing
Data are important to make informed decisions and raise the odds to pick the right one. However, we can’t measure everything.
We know how many people visit our website because it's a number, a quantity: quantitative data.
We do not know WHY these people visit and stay on our website. What do they like? This is qualitative data.
- Quantitative data tell us how many (size)
- Qualitative data tells us why (behavioural)
To carry on with the website visit example, you need to evaluate both sets of data if you want to improve the performance. When you know why people like your website, you can tailor landing pages for this segment and measure how many visitors stay on it. You can repeat the process with different segments and build high converting landing pages for each of them.
Or you can guess and hope to find the right one…
When we want to know "what" or "how many" specific quantities, we run Quantitative data analysis. It’s as simple as that.
Quantitative data give us the size, nothing more. They are useful for statistics, but we miss the opportunity to give meaning to these numbers if we consider only them.
We can compare, for example, the current University students in the first year with the previous years. For sure, it tells us something, and we immediately know if they are increasing or decreasing and therefore take the appropriate course of action accordingly.
Quantitative analysis and research methods often include:
- Closed-ended questionnaires and surveys
- Large-scale data sets
- Analytics gathered by machines
- Random sampling
- Structured data
- Tracking software such as CRMs, marketing automation, advertising
If we want to understand why, what motivates a customer, and gives meaning to quantitative data, we need to perform Qualitative data analysis. These data help us to understand why we have those numbers.
Once we know that the first-year students at the university increased, we can do qualitative research to understand who they are: male, female, ethnicity and so on. We might find out the majority of them comes from a specific country abroad or joined the University for a specific course only.
Qualitative analysis and research methods often include:
- Focus groups
- Open-ended questionnaires and surveys
- Unstructured interviews
- Unstructured observations (like reading social media posts)
- Case studies
Years ago, I had to run the European campaign for a B2B international company.
The data from previous campaigns showed mainly quantitative data, segmented by nation.
It was a good starting point but not enough for me. The previous campaign strategy was based on numbers only: everything focused on statistics and conversion rates and they set the investments accordingly.
I decided to change approach and to dig deeper into the data: I wanted qualitative data. Why did customers in a specific country like the product?
We found three major reasons:
1. the colour palette
2. the packaging
3. the words used on the product (and therefore the promise or the perception of it)
Then we contacted agencies in each relevant country to work with them to improve the product following the qualitative data analysis whilst maintaining the brand identity and communication (I bet many of you understand the challenge here).
Once we had everything approved, we prepared all the touchpoints along the funnel to adapt to the new communication and then we launched the campaign.
I must say, I used all the budget. I used it differently because part of it went for analysis and rebranding.
In the end, the campaign returned a higher ROI than the previous ones, much higher. And the following year, with more data available, and the new practice, the ROI grew up again.
Is it possible to use Quantitative data only?
The answer is yes.
Considering the amount of data we have today, if we pick the right ones, compare different clusters and use statistical modelling, they can be used to predict behaviours. And by understanding the cause and effect of certain data, you can get to the why without the need of running a qualitative analysis.
It’s a long process and it requires resources.
However, AI and predictive data analysis is growing and becoming more accurate and can simplify this process.
Understanding the customer is the key to win. Quantitative and qualitative data analysis give us a better understanding of the audience we are dealing with. I recommend using both of them until you think you have enough data to make your decision. Sometimes you just need a couple of analyses, sometimes more.
Sharing is Growing
What approach do you use? Following the topic of this article, collecting your favourite approach and why you use it (qualitative and quantitative data) we can open a debate that can enrich everyone.