Skip to main content

Running Multiple Experiments

Method 1: Layers​

Using layers allows you to run multiple landing page experiments without needing to update the code on the website for each experiment. Layers are a powerful way to run many experiments in sequence, one building on top of the learning from the other.

Running an experiment within a layer is simple. Create a layer with the page_url parameter defined first like here:

image

Then you can go ahead and create a new experiment as detailed above.

When you run experiments as part of a layer, you should update the script to specify the layerid instead of expid.

<script
  src="https://cdn.jsdelivr.net/npm/statsig-landing-page-exp?apikey=[API_KEY]&layerid=[LAYER_NAME]">
</script>

Method 2: Multiple Fallback Experiments​

Another way to run multiple experiments is using fallback experiments with different targeting gates. For example, if you have two experiments, one for organic visitors and the other for paid visitors, you might want to set up two experiments and pass them to the landing page. You can use multiexpids query param to indicate this.

<script 
  src="https://cdn.jsdelivr.net/npm/statsig-landing-page-exp?apikey=[API_KEY]&multiexpids=expid1,expid2,expid3">
</script>

This will first evaluate if the user is in exp1 and if the user is not part of the experiment, fallback to exp2 and so on. When the user is allocated to one of the specified experiments, then the page_url property is consulted and the redirect happens.

In these cases, it's helpful to know on the final redirect which experiment was consulted for the redirect and that can be easily fetched by looking at the _configname query param that's passed on to the redirected page.

info

If you don't want to expose the experiment names in the final redirect to your users, you can suppress the inclusion of the experiment name in the url by setting a parameter named excludeConfigNameInUrl to true in your experiments.

image