Optimizely – A/B testing¶

Optimizely is an easy way to implement A/B testing. Try different decisions, images, layouts, and copy without touching your website code and see exactly how your experiments are affecting pageviews, retention and sales.

Installation¶

To start using the Optimizely integration, you must have installed the django-analytical package and have added the analytical application to INSTALLED_APPS in your project settings.py file. See Installation and configuration for details.

Next you need to add the Optimizely template tag to your templates. This step is only needed if you are not using the generic analytical.* tags. If you are, skip to Configuration.

The Optimizely Javascript code is inserted into templates using a template tag. Load the optimizely template tag library and insert the optimizely tag. Because every page that you want to track must have the tag, it is useful to add it to your base template. Insert the tag at the top of the HTML head:

{% load optimizely %}
<html>
<head>
{% optimizely %}
...

Configuration¶

Before you can use the Optimizely integration, you must first set your account number.

Setting the account number¶

Optimizely gives you a unique account number, and the optimizely tag will include it in the rendered Javascript code. You can find your account number by clicking the Implementation link in the top right-hand corner of the Optimizely website. A pop-up window will appear containing HTML code looking like this:

<script src="//cdn.optimizely.com/js/XXXXXXX.js"></script>

The number XXXXXXX is your account number. Set OPTIMIZELY_ACCOUNT_NUMBER in the project settings.py file:

OPTIMIZELY_ACCOUNT_NUMBER = 'XXXXXXX'

If you do not set an account number, the Javascript code will not be rendered.

Internal IP addresses¶

Usually you do not want to track clicks from your development or internal IP addresses. By default, if the tags detect that the client comes from any address in the OPTIMIZELY_INTERNAL_IPS setting, the tracking code is commented out. It takes the value of ANALYTICAL_INTERNAL_IPS by default (which in turn is INTERNAL_IPS by default). See Identifying authenticated users for important information about detecting the visitor IP address.