+97% of Time Reduction to Onboarding New Metrics

Geek Friday / HelloFresh

I am presenting a successful showcase that has caused more than 97% timing reduction due to process automation. This automation was led by me when I worked on the Experimentation team. The slides will give you a context for the talk, going through the problem, and explaining the solution, and outcomes.

Transcript

1. +97% of Time Reduction to Onboarding New Metrics How?
2. who.is www.thulioph.com • Thulio • Software Engineer • Experimentation Team • 󰎙
3. Agenda Sorry, I have only 5 min 🏎💨 1. Context 2. Problem 3. Solution 4. Outcomes
4. Context
5. a.k.a Experimental Kitchen 📊 EK "Platform that allows the analysis of experiments and calculates relevant metrics to enable decision making."
6. Experimental Kitchen
7. Experimental Kitchen
8. Metrics Data
9. All phases when adding a new metric 🚥 Data phases 1. Insertion 2. Processing 3. Visualization
10. Insertion
11. Processing
12. Visualization
13. Problem
14. Problem
15. Problem
16. Problem
17. Problem
18. Problem
19. Problem
20. Problem
21. Problem
22. Problem
23. Solution
25. Data Insertion Automation 1/3 🤖
26. Data Insertion
27. Data Processing Automation 2/3 🤖
28. Data Processing
29. Data Visualization Automation 3/3 🤖
30. Data Visualization
31. Data Visualization
32. Outcomes
33. From 40 days To onboarding a new metric. To 1
34. of time reduction +97%
35. From people point of view: • The team is more confident with the ecosystem of tools • Building horizontal knowledge of the domain From product point of view: • Application becomes reliable • Adding new metrics, does not require code changes • Effectively scale the metrics management • Solid data contract between API and clients • Removing any friction to analyze the experiment results Highlights
36. Links • Miro board • Initiative Page • Jira Project • PRD: Automate metric updating process • API contract using JSON Schema • Architecture Decision Record • Metrics onboarding: end to end guideline
37. Danke!