Your cart is empty.
Keep shoppingData-Driven Management: KPIs, Analytics and Decisions [EN]
2,688 students
Data-Driven Management: KPIs, Analytics and Decisions [EN]
data driven decision making | kpi | business analytics | people analytics | critical thinking | data driven management
Last updated 5/2025
English
English [Auto],
What you'll learn
- Tell a symptom from a problem, and state the problem so it can actually be measured
- Recognise the biases that corrupt a conclusion before the data is even collected
- Design metrics that measure the outcome instead of the activity
- Segment data and find the insight hiding inside the average
- Run correlation and regression analysis, and interpret the output honestly
- Design an experiment with a control group and read the result
- Build a forecast and know how much confidence it deserves
- Present analysis to leadership in a form that produces a decision
- Learn alongside Mike's 1.6 million students from 185 countries
- Get the author's experience from Preply, Wargaming, iDeals and Alfa-Bank
Explore related topics
Course content
14 sections • 38 lectures • 18h 23m total length
Requirements
- Decisions you are responsible for, and data you suspect you are not using well
- No statistics background required. Regression is built up from zero
- A spreadsheet you can experiment in, since several lessons work through calculations
- English at intermediate level or above, since all lessons are in English
- Willingness to find out that a favourite metric measures nothing
Description
This course contains the use of artificial intelligence.
Most companies that call themselves data-driven have a dashboard nobody opens and a set of metrics that were gamed within a quarter.
Why the dashboard did not help
The failure almost never happens at the analysis stage. It happens two steps earlier, when somebody measured a symptom instead of the problem, or built a metric that rewarded the behaviour rather than the outcome. I have introduced metrics that were being worked around inside six weeks, and the analysis behind them was fine. The question was wrong. No amount of regression rescues a question that was badly framed.
What this course actually covers
Five blocks, in the order the work happens. First, thinking: what makes an argument strong, which cognitive biases corrupt a conclusion before you reach the data, how to judge a source, and the decision models for acting under uncertainty. Second, framing: telling a symptom from a problem, eliciting what people actually need, and stating a question as problem, goal, constraint and metric. Third, measurement: designing an indicator, deriving it from critical success factors, displaying it properly, and the six myths that break most systems — starting with tying metrics to pay. Fourth, analysis: segmentation, lifetime value, funnels, correlation, multiple regression, forecasting, and designing an experiment with a control group. Fifth, eight lessons on a company that made all of this its operating system, including how it decides who to hire and how it predicts who will leave.
A note on the data
Eighteen of the thirty-eight lessons work with people data — turnover, engagement, hiring funnels, employee value. The method does not depend on that: segmentation, key driver analysis, multiple regression and experiment design behave the same whatever the rows represent. I am telling you what the examples look like so you can decide before buying rather than in lesson twenty-one.
Who is teaching this
I am Mike Pritula. I built the people system at Preply as it became a unicorn, and I have worked at Wargaming, iDeals and Alfa-Bank. More than 1.6 million students have enrolled in my courses across 185 countries, and over 150,000 specialists have gone through my programmes. I hold PHRi and SHRM-CP certifications and represent HRCI in more than ten countries.
What is included
Lifetime access to all 38 lessons
Active instructor support in the Q&A section
A Udemy Certificate of Completion
Working frameworks: Problem → Goal → Constraint → Metric, impact mapping, the 10-80-10 rule, key driver analysis, regression, experiment design
Real cases, including how Google runs its decisions on data
Where to start
Pick the metric your team reports most often and ask what decision it has changed in the last six months. If the answer is none, that metric is decoration, and this course is about the difference. Enrol now and start the first lesson today.
Who this course is for:
- Managers who have dashboards and still decide on instinct
- Leaders whose reports are full of numbers that never change a decision
- Analysts who produce correct analysis nobody acts on
- Operations and business specialists building a measurement system from scratch
- Founders deciding what their company should track at all
- Consultants who need to prove an effect rather than assert one
- Anyone who has watched a metric get gamed within a quarter of being introduced