Frequently Asked Questions

A curated summary of the top questions asked on our Slack community, often relating to implementation, functionality, and building better products generally.
Statsig FAQs
OpenAI affirm Univision Notion Microsoft nyt ea beyond Ancestry Rippling Thumbtack Collective
GENERAL

How can we efficiently test a large number of copy variants without manually creating each one?

Date of slack thread: 7/1/24

Anonymous: If I wanted to run a large number of copy variants, what’s the recommended way to do that? For example, on one page I have Headline, Description, and Button. If I have 5 options for each of those that can be combined together that’d be 125 variants (555). What would y’all recommend so that we don’t have to go in and manually create all of those?

Anonymous: Hey <@U073HQ81F8V> - it depends a bit on what you’re looking to test out (i.e. how much you want all combos): I’d generally advise against trying to run 125 variants - there’s a lot of room for false positives there (i.e. it’ll be 124 variants compared against the single control variant - with a 95% Confidence interval, you’d expect 6 or 7 to show as positive even if all were the same experience). I’d probably generally lean towards actually running them as separate tests, to get a bit more insight into if a single one of the Headline / Description / Button - are super impactful. Depending on the test - and what you’re trying to measure, this could actually be a pretty interesting use case for https://docs.statsig.com/autotune (our multi-armed bandit). If you’ve got 8 headlines you want to try out - you can really just set it and forget it if there’s a certain event you’re able to optimize around (i.e. clicks on a given button).

Anonymous: What we’re looking to understand is which combination of the three leads to increase in conversion. Would we be able to run multiple variables inside the same autotune? Or would we need to reduce the number to 8?

Vijaye (Statsig): I don’t think you’d want to deal with a combinatorial problem. My recommendation would be to treat these three properties as separate experiments while keeping the others constant. Find the one that’s best in each and then run a separate experiment picking the best of three to find the lift over baseline.

In most cases that should work well. There are other cases where it’s usually clear for the product team combinations that work well. In those cases pick fixed combinations and test them.

Join the #1 experimentation community

Connect with like-minded product leaders, data scientists, and engineers to share the latest in product experimentation.

Try Statsig Today

Get started for free. Add your whole team!

What builders love about us

OpenAI OpenAI
Brex Brex
Notion Notion
SoundCloud SoundCloud
Ancestry Ancestry
At OpenAI, we want to iterate as fast as possible. Statsig enables us to grow, scale, and learn efficiently. Integrating experimentation with product analytics and feature flagging has been crucial for quickly understanding and addressing our users' top priorities.
OpenAI
Dave Cummings
Engineering Manager, ChatGPT
Brex's mission is to help businesses move fast. Statsig is now helping our engineers move fast. It has been a game changer to automate the manual lift typical to running experiments and has helped product teams ship the right features to their users quickly.
Brex
Karandeep Anand
President
At Notion, we're continuously learning what our users value and want every team to run experiments to learn more. It’s also critical to maintain speed as a habit. Statsig's experimentation platform enables both this speed and learning for us.
Notion
Mengying Li
Data Science Manager
We evaluated Optimizely, LaunchDarkly, Split, and Eppo, but ultimately selected Statsig due to its comprehensive end-to-end integration. We wanted a complete solution rather than a partial one, including everything from the stats engine to data ingestion.
SoundCloud
Don Browning
SVP, Data & Platform Engineering
We only had so many analysts. Statsig provided the necessary tools to remove the bottleneck. I know that we are able to impact our key business metrics in a positive way with Statsig. We are definitely heading in the right direction with Statsig.
Ancestry
Partha Sarathi
Director of Engineering
We use cookies to ensure you get the best experience on our website.
Privacy Policy