A DoorDash Sales Analysis

Evaluating Marketing Effectiveness and Insights

  • Role: Data Analyst (Marketing & Growth)

  • Objective: Evaluate campaign effectiveness and customer segmentation to inform future marketing investment

  • Constraints: Limited data availability, Excel-only tooling

  • Outcome: Actionable recommendations with known confidence limits

  • Dataset: 2,025 customers (2014–2016), originally from iFood (DoorDash-like delivery platform). Available on GitHub and Kaggle.

Read the blog post here for a narrative overview of the case study.

Business Problem

The goal is to find business opportunities and insights and improve campaign strategy.

The marketing team wants to improve its campaigns by:

  • Targeting the right audience

  • Promoting the right products

  • Executing effectively across channels

Data Overview & Constraints

Included

  • Customer demographics (income, dependents, education, marital status)

  • Spend per product category

  • Campaign participation 

  • Channel usage (store, web, catalog)

  • Complaint count

  • Order recency

Not Included (Key Constraints)

  • Campaign costs (CAC)

  • Subscription retention, churn, or lifetime value

  • Returns, refunds, and detailed complaint reasons

  • Product SKUs, margins, or supplier data

  • Geospatial or time-series ordering behavior

  • Customer experience and operational signals

Implication:
Recommendations emphasize what can be responsibly acted on given these constraints.

Analytical Approach

Segmented customers by campaign participation and retargeting status

  1. Normalized spend behavior by income bands

  2. Compared campaign reach vs value (avg spend per customer)

  3. Analyzed product category preferences by campaign

  4. Evaluated channel mix differences between members and non-members

  5. Flagged overlap and retargeting effects that complicate attribution

Key Findings

1. Campaign Membership Correlates With Higher Spend — But With Diminishing Returns

  • Campaign members spend more than non-members at comparable income levels

  • Income explains a significant portion of spend variance

    • R² ≈ 70% (campaign members)

    • R² ≈ 62% (non-members)

  • Customers in multiple campaigns eventually spend a lower percentage of income, suggesting saturation rather than loyalty at higher exposure levels

Implication: Campaigns increase spend, but indiscriminate scaling risks diminishing marginal returns.

2. Product Preference Is Strongly Skewed Toward Wine

  • Wine is the dominant category across all campaigns

  • Meat is consistently the secondary category

  • Other categories (fish, sweets, fruit) remain relatively stable and minor

Implication: Product-led campaign strategies should center on wine, with selective bundling rather than broad category expansion.

3. Campaign Effectiveness Varies Sharply by Reach vs Value

  • Campaign 6 has the highest registrations but elevated complaint indicators

  • Campaigns 1 and 5 generate the highest average spend per customer with low discount reliance

  • Campaign 2 shows no new customer acquisition and relies entirely on retargeting

Implication: Reach alone is a misleading success metric without value and friction context.

4. Channel Preferences Differ Between Campaign Members and Non-Members

  • Campaign members purchase more via catalog

  • Non-members purchase more in-store

  • Higher catalog usage correlates with higher complaint rates in certain campaigns

Implication:
Channel mix influences both value and operational risk.

5. Customer retargeting causes overlap between campaigns that complicate attribution spend behavior and causality.

  • There are a total of 993 registrations, the majority of whom were retargeted and joined multiple campaigns

  • Campaign 2 is entirely sourced and retargeted from other campaigns

  • Impossible to delineate product market fit without product distribution per order in addition to per campaign



Implication:
Analysis is limited to categorizing non campaign members from campaign members and retargeted customers.

Campaign Synthesis

Testable Hypotheses

Customers with low discount reliance generate higher average spend

  • Wine-focused campaigns outperform mixed-category campaigns in value

  • Higher catalog usage is associated with increased complaint rates

  • Retargeting overlap inflates perceived campaign effectiveness

  • High-reach campaigns without CX alignment increase friction risk

These hypotheses require validation using CAC, retention, and complaint-level data.

Recommendations by Stakeholder

Sales / Revenue

  • Prioritize C1 and C5 for premium offerings

  • Focus upsell and cross-sell around wine-based bundles

  • Avoid scaling low-value, high-friction campaigns

Marketing

  • Evaluate campaigns using reach + value, not registrations alone

  • Use catalogs for curated, high-value segments

  • Use stores/web for broader acquisition

CX / Operations

  • Investigate complaint drivers in C3 and C6 before further investment

  • Align customer expectations with channel and campaign messaging

  • Add operational KPIs before scaling acquisition spend

Risks & What’s Missing

No marketing cost data limits ROI conclusions 

  • No purchase order detail limits correlations on buying patterns to campaigns

  • No retention or churn data limits lifetime customer value assessment

  • Complaint counts lack root-cause classification

  • No operational data limits profitability conclusions 


Next Data Needed:
CAC, retention rates, purchase order detail, complaint taxonomy, and margin by product SKU.

Decision Confidence Summary

This case demonstrates my ability to frame ambiguous business questions, analyze under real-world constraints, and deliver decision-safe recommendations.

Next

Department of Education - Data Visualization and Dashboards