Once you’ve created your account and installed Kameleoon script, you are ready to create and launch your experiments! This article will guide you step by step towards the launch of your first experiment.
Kameleoon’s editing mode enables you to create variations of your webpages and to launch and manage your experiments.
Via your website
To launch Kameleoon, go to one of the webpages on which you have installed the Kameleoon script and press Maj + F2 or, il you are a Mac user Fn + Maj + F2.
If nothing happens, you can try another method to launch Kameleoon: simply add this code at the end of your URL:
For example, your URL will become:
Once Kameleoon is loaded, a connection pop-in opens. Fill in your user name and password to access the editor.
Then choose the type of experiment you want to create.
When the editor is open you will see a header above your webpage.
Via your personal space
You can also launch the editor from the back-office. First of all, you need to log into your personal space.
Click on the “New experiment” button on the homepage or on the Experiments dashboard page.
Create an experiment
New classic A/B experiment
When Kameleoon has loaded, a pop-in opens offering you to choose between the creation of a new classic A/B experiment, a multivariate test and the edition of an existing one.
We choose to create a new classic A/B experiment.
Name your experiment in the new popin. You can also add a small description and tags to your experiment. Then click on “Validate”.
The Kameleoon header appears above your website: you are now able to create your variation!
When you are creating a new experiment, two variations are automatically created:
- Original: this is your original page without any changes. You cannot change or delete this variation as it is the one used as a reference for your experiment.
For further information about variations, please read our article Managing experiments.
Edit an element
To change an element on your page, use your cursor to select the element of your choice. A selection frame will help you to select it easily.
Once the element selected, an edition panel appear on the left of your screen, composed of 3 tabs: “Style”, “Position” and “Tracking”.
For each tab, the most common and most likely changes will be presented. To see all available tools, click on the “More Options” button.
To learn more about the graphic editor, see our articles on the subject.
Validate your choice to apply it to the selected element.
We now have our new variation ready to be used in an experiment.
Launch the experiment
Once our new variation is created, we are ready to launch your experiment. This step will only be possible if you have properly installed Kameleoon script on your website.
To launch your experiment, click on the “Finalize” button on the right of the header.
The following menu appears, allowing you to configure and launch your experiment:
To launch the experiment, simply follow the 3 steps on your screen:
- 1st step: Select the variation to include in the experiment, by checking the corresponding box. If you have created several variations, you can select as many as you want to. To launch a experiment, you must select at least one variation.
- 2nd step: Choose the visitors targeted by your experiment. By default, the experiment will launch on the current page without any specific targeting, and the traffic will be equally split between your variation(s) and the original page.
- 3rd step: Choose your reporting tool. By default, only Kameleoon’s reporting tool is configured with your account, but you can add other reporting tools on your personal space.
- 4th step (optional): Simulate your test to verify its settings.
This step is very simple and consists in selecting the variation(s) to include in the experiment. Then, you can indicate if the experiment is a Multivariate test or not. For further information about multivariate tests, please read our article about MVT creation.
Two types of views are available: the list and the graph.
The original is automatically included in the test.
By default, the experiment will start on the current page, all visitors included, and the traffic will be distributed equally between your variation and the original page.
To change the traffic allocation, click and drag the slider of a variation by holding the click, or directly enter the percentage by clicking on the number displayed on the right of the slider. Any unassigned percentage of your traffic will automatically be considered as “Traffic excluded.”
During the second step of the finalization, you need to define the targeting. Choose who among your visitors will be exposed to the experiment. By default, all visitors who visit the page on which you created your experiment are exposed.
You can target by web page or segment. If you choose a segment targeting, select an existing segment or create a new one. If you choose to create a new one, the creation pop-in enables you to select the targeting conditions that will make up your segment.
To learn more about targeting, you can read our article “Managing targeting segments”.
Tracking and goals
To analyse your experiments, several reporting tools are available. Kameleoon can be integrated with all major web analytics solutions (Google Analytics, AT Internet, Webtrends, Adobe Analytics (Omniture), Eulerian, etc.).
Kameleoon also offers its own reporting tool, configured by default with your account. To use the Kameleoon reporting tool, select the first tab of the pop-in, then choose your main goal. To create a goal, click on the “Create a new goal” button or follow the steps as described in the next paragraph.
In the “Other tracking tools” tab, if your reporting tool is not on the list, please read our article Reporting tools and goals to add a new web analytics solution.
You must also configure one or several goal(s). By default, a first goal “Engagement” is automatically configured, informing you if the visitor visited other pages after the landing page. Engagement can be defined as the opposite of bounce rate.
You can configure new goals to analyze the access rate to a page, number of page views, time spent on the page and any other custom goal. To do so, go to the “Goals” page on your personal space.
You also can configure a click tracking directly from the graphic editor. To learn how to use this feature, see the article on click tracking.
Once you have created your goals, all you have to do is wait for enough visitors to access your website before you can read and analyze the results. To learn more, see How long does an experiment last?
This step is optional but highly recommended. It allows you to check the settings of your experiment before launching it.
Click on the simulation: the page on which you have created your experiment opens in a new tab. The lower half of the page displays the simulation panel.
If you make changes to your experiment, don’t forget to simulate again.
For further information about simulation, you can read our articles on the subject.
Upon finalization of each of these steps, a green “check” icon appears as well as a summary of the chosen parameters. To be able to run the experiment, all first three steps must have been completed.
Then click on the “Launch” button to launch your experiment.
Last step before the launching of your experiment: a recap is displayed, allowing you to check if everything is fine.
Click on “Launch” to confirm and launch your experiment.
A pop-in will confirm the launch.
Note: There may be a short latency time (up to 10 minutes) between the launching of an experiment and its visibility on the website. Don’t worry if your experiment does not appear immediately!
Access to results
To view and analyze the results of your experiment, log into your personal space.
On the left side menu, click on the “Activate” > “Experiments”.
On the line of your experiment, click on the “Results” button.
Understanding the results
The results of your experiment are displayed.
At the top of the results page, you will find the main results of your experiment. You will be able to see at a glance which variation of your experiment is the winning one.
The information displayed here depends on the results and progress of your experiment. You may encounter 4 different situations:
- It is too early to exploit the results;
- A variation is winning;
- A variation is recommended;
- The recommended variation is the original one.
The side panel of the results page enables you to view the settings of the experiment for which you are viewing the results and to edit them.
At the top of the results page, to the right of the experiment’s main information you can fnd the general actions menu:
- Refresh data
- Reset data
- Customize view
- Save display settings
- Create alerts
- Enable bayesian
The comparative view enables you to compare, on a spider web graph, the target improvement rates set on the experiment. It is only available if at least two goals are configured on your experiment.
The graph shows the evolution of the data over time. The x-axis displays the time scale and the y-axis the value of the data.
For more details, a table similar to the one below is displayed at the bottom of the page, and offers a detailed view of experiment results per variation or per goal. Ventilation tools, filters and customization of the display are available to facilitate the visualization.
You can click on the menu linked to each variation via a small icon.
There you can change the traffic allocation or redirect 100% of your visitors to the top-performing version of your page.
For further information about the results of your experiments with Kameleoon, you can read our documentation about the results page.