Quantitative Data Analysis
We often need quantitative statistics to reduce risk during Business Design sprints. Here is an overview of the data sources we typically use, and how we analyse and present them for the team.
Bernhard Doll
Business Design Maverick
Content
1. Overview
Some activities in the Discover Phase — especially answering open questions about your current business model and the behaviour of your customers and users — and some experiments in the Validate Phase require you to collect and analyse large amounts of data, using descriptive or inferential statistics. Data sources include:
Surveys
Behavioural data from Automated Tracking
Internal applications (e.g. CRM, ERP, issue tracking systems, SCM)
Social media (e.g. Facebook, Twitter, YouTube and LinkedIn)
External applications (e.g. App Store, Google Ads)
Quantitative data analysis is usually based on a handful of key statistics. Choose the ones that fit your data and answer your open questions or test your hypotheses. Before you start the analysis, prepare and clean your raw data in a spreadsheet — check it's complete and remove errors. Present your insights clearly, and map your data to your questions and hypotheses in a dashboard.
Keep in mind
Funnily enough, most people have little real interest in validating their business models. A tangible prototype and a few happy customers feel better than the quantitative statistics that would actually reduce their risk — hard numbers make things real, and real can be scary.
2. Key Elements
Statistics | Description |
Frequencies | Number of times a particular value occurs |
Percentages | Each value's share of the total, expressed as a percentage |
Mean | Average of the values |
Median | Middle value in the ranked distribution |
Mode | Most frequent value |
Range | Distance between the highest and lowest values |
Variance | Variability of the distribution |
Standard deviation | Amount of variability — a higher standard deviation means the data are more spread out |
Correlation | Strength and significance of the relationship between two variables |
ANOVA | Compares the variation between different groups to the variation within those same groups |
Regression | Whether one variable predicts another |
3. Usage Scenarios
Explore your current business model, processes and the behaviour of your customers and users, in the Discover Phase
Explore Antilogs from your Hypotheses & Experiments template
Validate hypotheses from your Hypotheses & Experiments template
4. Example Tools
5. Q & A
How do I build a meaningful dashboard that combines different data sources — Google Ads, Facebook Ads, Twitter Ads, Google Analytics, custom data? Use Google Data Studio. It needs no coding, and it integrates many data sources out of the box.
How do I track custom user behaviour on our site — downloads, configurator actions, scroll position and so on — within Google Analytics? Use Google Analytics Custom Events.
6. Example Dashboard
