Skip to content

mustafasencer/campaign-performance-challenge

Repository files navigation

Campaign Performance Analysis

codecov Test

📦 Dependencies

make deps
make dev-deps
make run-infra-background

docker and docker-compose should be installed on the host machine.

💻 Migration

make migrate-scenario-(1st|2nd)

💨 Run

make run-scenario-(1st|2nd)

💄 Format and Lint

make format
make lint

🚨 Tests

make test
# or
make test-with-coverage

📝 Docs

1st Scenario

Assumptions:
  • Data deduplication decision has been taken based on the data analysis done in here.
  • The decision of keeping the last duplicate event has been taken.
  • click_through_rate has been calculated based on a 30 min time window check where the presence of a ReferralPageLoad event is checked before firing of the ReferralRecommendClick event.
    • Belonging to the same customer and user.

2nd Scenario

Assumptions:
  • Data deduplication has been discarded and consequently each event has been processed.
  • The same click_through_rate assumptions has been applied.

📝 The presence of various duplicate events and the way they are treated by the 2 scenarios has resulted in different numbers for page_loads numbers.

About

Simple ETL Pipeline with Pandas and Postgres 🐿️

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors