layout | title | lead |
---|---|---|
page |
Help |
All-new documentation for an all-new Myna. |
This is a work-in-progress help site for Myna v2, which is currently in public beta.
Myna v2 is pre-release software - it is subject to change and may contain bugs and missing features. Please report any issues via our contact page.
In the meantime, we recommend using the regular dashboard for mission critical experiments until the software comes out of beta. Please refer to our existing help site for support with the regular dashboard.
We just launched a brand new version of Myna that makes it much more powerful and easier to use. Here are the new features:
-
Dashboard improvements - we've completely overhauled our dashboard, providing it with additional power and flexibility:
- Search for experiments by name or ID. Mark experiments as draft, published, or archived to focus on a particular stage in the lifecycle.
- Preview variants directly from the dashboard using the new Myna inspector.
-
Deployment improvements - it's far easier to set up experiments on your web site:
- New variant settings allow you specify colours, layouts, and text on your dashboard and use them in your HTML on your site.
- New deployment snippets allow you to paste one tag into your page template to get access to all your experiments.
- Configure experiment settings directly from the dashboard without having to access your source code.
- Our deployment scripts are hosted on a CDN, meaning blazingly fast page load times all around the world.
-
API improvements - we've added new features to our API to allow developers to test anything they need to:
- API keys allow you to run experiments without storing your log-in details in your code.
- Assign settings to experiments and variants - remote control your experiments via our API.
- Override our suggested variants when you need to - create complex experimental setups that are part Myna, part your own design.
-
Algorithmic improvements - we've overhauled our machine algorithms making Myna even faster at learning your users' preferences, whilst simultaneously being able to cope with a much wider set of experimental constraints.
We've also made it possible for you to remotely tweak and configure experiments direclty from the dashboard without
The source for this help is all on Github. Sending a pull request with improvements will make us extremely happy. Alternatively just drop us an email.