INSIGHT on INSIGHT: Understanding Quantitative Insight
Research, or any set of data, can be classified as quantitative or qualitative depending on how the data was acquired. Like primary, secondary and tertiary research, quantitative and qualitative research each has its place. And each comes with pros and cons.
This article focuses on quantitative insight. Follow this link to the complementary article on qualitative insight.
QUANTITATIVE INSIGHT
Quantitative insight (“quant” for short) refers to large sets of numeric data allowing for accurate and repeatable analysis and measurement. Data is typically acquired through an impersonal mechanism like a survey, sales transactions, or digital breadcrumbs.
These data sets can range from a brief internet survey among a couple hundred respondents to loyalty card data with millions of unique entries and thousands of variables.
The results are often expressed as percentages, with meaning derived from both the absolute values and the relative comparison between or across numbers, both of which are typically viewed as statistically accurate enough to be treated as factual.
Absolute values are often used to emphasize size: 57% of our buyers are age 21-35. 41% of respondents claim to have done online research before making a purchase.
Relative comparisons define the relationship between two groups more than the numbers themselves: Shoppers are twice as likely to prefer red to blue. Concept X attracts 30% more unique buyers compared to Competitive Benchmark Y.
When clients have a limited budget and have done little or no prior research, it is typically wise to invest in quantitative research first because there is so much more bang for the buck. Quantitative research is not fundamentally better than qualitative research, but dollar-for-dollar it can typically answer more questions and tells a more compelling story.
After all, doing any type of research is ultimately about getting better information to make better decisions. And I’m more comfortable making recommendations based on the results of a thoroughly analyzed internet survey compared to the directional perspective from a limited number of in-depth interviews.
In my world, quantitative research most commonly comes from internet-based surveys or POS sales data. However, there are plenty of other datasets, including those derived from online traffic analytics.
Read the list pros and cons to better understand why I like quantitative research so much:
PROS OF QUANTITATIVE RESEARCH
- It is easy to balance the sample of people included in the data set to make sure it is representative of the population being studied
- Data can be filtered to look at specific profiles or sub-segments
- There is greater confidence in the accuracy or reliability of the data
- The data can typically be projected or extrapolated to a larger population
- By having a single set of data, it is possible to filter and cross-tabulate any subset of variables
- It is possible to merge and compare demographic, psychographic and behavioral data
- There are numerous question and analysis techniques available to accomplish a variety of objectives
- The data can be designed to cover a broad number of topics or dive deep into a single topic
- Data acquisition can be repeated over longer time periods to reveal trends
- There are numerous options to visualize and present quantitative data to tell stories with few words
- In my experience, quantitative data is more readily accepted as truth or fact, is generally less likely to be questioned, and generally more likely to drive decisions.
CONS OF QUANTITATIVE RESEARCH
- Upfront planning must be done to make sure the data set contains all desired variables
- Fresh quantitative data often requires spending time to clean and organize before analysis can begin
- Analysis requires use of software, which may be as simple as running formulas or pivot tables in Excel or involve more complicated package like SPSS or R
- Accurate analysis requires a clear understanding of how the data is structured and labeled and what it means.
- There is risk of making mistakes during analysis (such as incorrect calculations) unless quality controls exist
- Quantitative data generally comes with a margin of error, which increases the smaller the sample size of the data is
- There is typically limited ability to dive deep into the ‘why’ behind the data
- It can be difficult to incorporate non-numeric data
- Results do not always reveal the underlying meaning or cause behind the key data points
Overall, quantitative insight can come in many forms and can produce a broad range of information. It should provide numerous options for analysis and presentation of results. And the objective, numeric nature of the data should instill confidence when using it to base important decisions on.