Outline new structure for experiments #9938
Conversation
|
The latest updates on your projects. Learn more about Vercel for Git ↗︎
|
| - Start here | ||
| - New doc with the key concepts you need to understand | ||
| - Outline: |
There was a problem hiding this comment.
I like where this is going. Curious what you think should go here?
| - Start here | |
| - New doc with the key concepts you need to understand | |
| - Outline: | |
| - Key concepts |
| - [Experiments without feature flags](/docs/experiments/running-experiments-without-feature-flags) | ||
| - Should this move to Guides? |
There was a problem hiding this comment.
What's the motivation behind moving it?
| ## Methodology | ||
|
|
||
| - Statistics primer | ||
| - Trends with count | ||
| - Trends with property value | ||
| - Funnels | ||
| - Attribution type: Funnels allow you to have the attribution assigned at any point in the funnel, not just the first step. If the user assigned to a variant later on in the funnel, then they'll be included in the entire funnel. | ||
| - [Traffic allocation](/docs/experiments/traffic-allocation) | ||
| - [Sample size and running time](/docs/experiments/sample-size-running-time) |
There was a problem hiding this comment.
Not entirely sure what we talk about in these sections? But in general I think its a good direction
| ## Methodology | |
| - Statistics primer | |
| - Trends with count | |
| - Trends with property value | |
| - Funnels | |
| - Attribution type: Funnels allow you to have the attribution assigned at any point in the funnel, not just the first step. If the user assigned to a variant later on in the funnel, then they'll be included in the entire funnel. | |
| - [Traffic allocation](/docs/experiments/traffic-allocation) | |
| - [Sample size and running time](/docs/experiments/sample-size-running-time) | |
| ## Under the hood | |
| - Statistics primer | |
| - Trends with count | |
| - Trends with property value | |
| - Funnels | |
| - Attribution type: Funnels allow you to have the attribution assigned at any point in the funnel, not just the first step. If the user assigned to a variant later on in the funnel, then they'll be included in the entire funnel. | |
| - [Traffic allocation](/docs/experiments/traffic-allocation) | |
| - [Sample size and running time](/docs/experiments/sample-size-running-time) |
| - Holdout group | ||
| - Link to holdouts documentation |
There was a problem hiding this comment.
I would still keep holdouts as its own page. Easier to find
| - Evaluating results | ||
| - Use the second part of the existing 'Testing and launching' doc. |
| - Navigate to Experiments and click "New experiment" | ||
| - Explanation of each form field: | ||
| - Name | ||
| - Feature flag key | ||
| - Description | ||
| - Participant type | ||
| - Variants |
There was a problem hiding this comment.
This is mostly the same, right?
|
"Creating a no-code web experiment" could include some mention of targeting devices: https://posthoghelp.zendesk.com/agent/tickets/21678 |
@Lior539 Would it be acceptable to have a "Release conditions" doc under "Features" for both Feature Flags and Experiments?
I feel like it's an important-enough feature that there should be a page dedicated to it. As far as I can tell, this is the canonical reference for it: https://posthog.com/docs/feature-flags/creating-feature-flags#release-conditions |
Changes
WIP