Draft the next campaign from customer segments
Act as a fictional JaffleMart analyst, connect orders to customers, read the actual segment distribution, and plan re-engagement only for groups that exist.
0/7 chapters complete
A weekly meeting about the next campaign
The fictional food retailer JaffleMart is preparing its next re-engagement campaign. Sending the same coupon to everyone ignores the difference between a frequent customer and a first-time buyer. Assuming that a churn-risk group exists can also spend budget on an empty audience.
Today, you are the JaffleMart analyst answering this question:
Which customers should we re-engage, and what next action fits each group?

Decision criteria for this mission
- Perspectives in play — Data analysis identifies the customer groups that actually exist and the gaps in their evidence. Campaign planning chooses a next action for each group and which sends to withhold.
- Decision constraints — The 100 raw orders include 12 rows to remove and two missing products to
retain as
PRD-UNKNOWN. Do not invent a campaign for a customer group that is absent from the analysis. - Completion signals — Confirm 88 clean orders and 7
champion, 11loyal, 4new, and 0at_riskcustomers, then explain the three action drafts and why unknown products stay out of personalization.
Workshop goals
- Detect
nulland blank strings as missing values, then apply field-specific handling. - Connect orders to customers and derive segments from recency, frequency, and spend.
- Read the actual 7, 11, 4, and 0 customer distribution.
- Turn segment, region, and tier metrics into a campaign draft.
- Inspect the purchase graph and communicate the limits of
PRD-UNKNOWN.
Seven decisions
Loading the diagram. Mermaid source:
flowchart LR
accTitle: Seven decisions for drafting a customer campaign
accDescr: Define the re-engagement question, clean and connect orders, read the actual segment distribution, draft the campaign scope, check the evidence in a quiz, and explain the decision.
question["1. Define the question"] --> clean["2. Clean and connect"]
clean --> rfm["3. Classify with three questions"]
rfm --> results["4. Read actual groups"]
results --> campaign["5. Draft the campaign"]
campaign --> check["6. Check the evidence"]
check --> explain["7. Explain the decision"]Each chapter follows question → reason → action → observation → interpretation → next decision. Begin with what you want to say to a customer, not with an acronym.
Before you begin
- An analyst account with Editor access or higher
- Enough download space for a scenario ZIP of about 32 KB
- About 90 minutes
You do not need a terminal, Python, or a clone of dhub2-examples. If scenario import is new to you,
start with Import and tour a complete hands-on scenario.
Learning journey
- Start with the re-engagement question15 minImport the E-commerce scenario and separate the order, customer, and result assets needed for a campaign decision.
- Clean orders before connecting customers15 minRun the order-cleaning pipeline and inspect the removal and replacement policies for missing identifiers.
- Turn three customer questions into segments15 minAsk when, how often, and how much a customer bought before learning the RFM acronym and label rules.
- Read the actual segment distribution and customer examples20 minConfirm the four segment counts and select a synthetic customer example to make each campaign action concrete.
- Draft the campaign with honest limits15 minConfirm campaign evidence on the dashboard and purchase graph while separating the limits of unknown products.
- Check the campaign reasoning5 minBefore the final explanation, review order cleaning, the segment distribution, purchase relationships, and campaign scope in four questions.
- Explain the campaign decision with evidence5 minReturn to the re-engagement question and explain both the groups to activate and the work to hold.