12 tips for PhD students from Philip Guo:

Tip 1. Start simple

Tip 2. Look at raw data

Tip 3. Be concrete

Tip 4. Incremental results

Tip 5. Find something interesting

Tip 6. Push hard on clean signals

Tips 7. Actionable

Tips 8. Comparative advantage is key

Tip 9. Find solid levers to pull

Tip 10. Document success and failures

Tip 11. Inspect outliers

Tip 12. Eliminate sources of bias