Case 9 · 3 levels
Where the web funnel leaks
Marketing thinks mobile converts badly. Frame the question, check what the session data can and cannot show, build a funnel by device and by year, and write the recommendation.
StartThis case unlocks when you finish any one of these chapters. The levels are readable now.
Jordan Reyes has been looking at the web dashboard and does not like what he sees: mobile visitors buy far less often than desktop visitors, and mobile is now most of the traffic. Before anyone rebuilds the mobile site, he wants to know where the problem actually sits.
A single conversion rate cannot answer that. A visit has to go through several stages before it becomes a sale, and a funnel shows at which stage people drop out. This case builds one:
- Frame the question: which sessions, which stages, which comparison, and what decision it feeds.
- Check the data: duplicates, empty sessions, and the sessions that can never convert in this data.
- Analyse in SQL, pandas or Excel, whichever you prefer: the funnel by device, then the same rates year by year as the traffic mix moves to mobile.
- Recommend: pick the paragraph you would send, with the numbers and the caveats.
The case opens when you finish any one of SQL chapter 6, pandas chapter 6 or Excel chapter 3. It
uses conditional counts (SUM(CASE WHEN ...), groupby with boolean columns, or SUMIFS on a
prepared sheet); the lessons walk through each.
Levels
- Frame the mobile conversion question
from Jordan Reyes, Head of Marketing
smart goal, problem framing, funnel, conversion rate, segmentation
- Check and build the web funnel
from Jordan Reyes, Head of Marketing
data validation, conditional aggregation, funnel, conversion rate, mix shift, simpsons paradox
- Recommend: where should the mobile work go?
from Jordan Reyes, Head of Marketing
recommendation, caveats, funnel, mix shift, data quality, data storytelling