Hampus Poppius

Hampus Poppius's articles

See All Blog Articles

Bad controls: why splitting your experiment on a post-treatment variable backfires

Data Science

Experiments

Bad controls: why splitting your experiment on a post-treatment variable backfires

Hampus Poppius

August 31, 2026

Segment analysis in experimentation: how to avoid fooling yourself

Data Science

Experiments

Segment analysis in experimentation: how to avoid fooling yourself

Hampus Poppius

August 28, 2026

Why summing your experiment wins overstates impact

Experiments

Guides

Data Science

Featured

Why summing your experiment wins overstates impact

Hampus Poppius

July 9, 2026

What does a holdout test actually measure?

Experiments

Guides

Data Science

Featured

What does a holdout test actually measure?

Hampus Poppius

July 9, 2026

A practitioner's guide to treatment effects in experimentation: ATE, CATE, ITT, LATE & ATT explained

Experiments

Guides

Data Science

Featured

A practitioner's guide to treatment effects in experimentation: ATE, CATE, ITT, LATE & ATT explained

Hampus Poppius

May 1, 2026

Your experiment lift is an average — which users actually benefited?

Experiments

Analytics

Data Science

Your experiment lift is an average — which users actually benefited?

Hampus Poppius

April 1, 2026

Ready to ship faster?

No credit card required. Start with feature flags, experimentation, and product analytics — free.

Get Started

Book a Demo

Simplified white illustration of a right angle ruler or carpenter's square tool.

White checkmark symbol with a scattered pixelated effect around its edges on a transparent background.

Realtor.com on guardrail metrics, the 30/30/30 rule, and AI as your junior data scientist

How Zalando connects every experiment to its North Star

Learneo on testing the opposite of every hypothesis

How to Use Feature Flags for Trunk-Based Development

Build vs Buy: Should You Build Your Own Feature Flag Tool?

How to Build a Feature Flagging Governance Framework

How to Scale Your Experimentation Program

The four questions Early Warning asks before any A/B test

A/B testing 300 million players without breaking their trust: the Supercell approach