This article will help you:
  • Segment the results to see how different segments of your total traffic might behave differently
  • Use default segments or create custom segments from Audiences or Dimensions

Not all visitors behave like your average visitors. When Optimizely declares statistically significant results, you may see Audiences or Dimensions that are important segments for your business. You'll probably want to look at results for these segments separately on the Results page.

Watch this short video about segmentation:

 

Optimizely allows you to segment your results by Audience or Dimension so you can drill down into the behavior of certain groups of visitors. 

 
Note:

When you create a new visitor segment, it will apply to all your visitors going forward, but not retroactively.

How you might segment

Depending on your business and industry, there are a number of ways you may want to segment your results using your experiment's Audiences or Dimensions. For example, here are some common segments to explore:

  • Browser type
  • Source type
  • Mobile vs. desktop, or by device
  • Logged-in vs. logged-out visitors
  • PPC/SEM campaign
  • Geographical regions (City, State/Province, Country)
  • New vs. returning visitors
  • New vs. repeat purchasers
  • Power users vs. casual visitors
  • Men vs. women
  • Age range
  • New vs. already-submitted leads
  • Plan types or loyalty program levels
  • Current, prospective, and former subscribers
  • Roles (if your site has, for instance, both a Buyer and Seller role)

Segmenting your results gives you valuable insights about different groups of visitors, instead of simply analyzing the results across your entire visitor base. Figuring out what experiences work for different audiences will form the basis of your personalization strategy.

Default segments

Optimizely captures certain segments by default for you to use in your results analysis.

  • Browser Type: Firefox, Google Chrome, Internet Explorer, Safari, and Other

  • Source Type:

    • "Campaign" source type contains users that arrive on a URL containing a 'utm_campaign,' 'utm_source,' 'gclid,' or 'otm_source' query parameter. If the URL contains one of these parameters, the visitor will count as "Campaign" traffic even if they arrived through search.

    • "Search" source type contains users that visit through a referral URL containing Bing, Google, Yahoo, or Baidu but NOT a specific campaign parameter (otherwise, they would be bucketed as "campaign").

    • "Referral" source type includes all users that come from another URL that doesn't count as Campaign or Search.

    • "Direct" source type includes all users who do not have any external referrer in their URL

  • Mobile: Whether the visitor is using a mobile device -- please note that this includes visitors who use tablets.

    Here is the logic behind how we determine if a visitor is on a mobile device or not. 

    • If the device type is detected as a "mobile" (phone) or "tablet" then it's mobile.
    • Else if the device ID is one of the ones we look for specifically (currently only "iPhone" and "iPad"), then it's mobile.
    • Else if the platform ID is one of the ones that we consider to be mobile (currently "android", "blackberry", "ios", " windows phone") then it's mobile.
    • Else it's not mobile.
  • Campaign source types (Enterprise Plan only): Visitors with the parameter utm_campaign will capture the parameter value. This will let you drill down into campaign keywords such as “holiday_sale”. There is a 20-character limit for values, and longer values will be truncated.

 
Note:

The “source type” dimension, with the exception of campaign, is based on the document.referrer value in the browser. Your visitors’ “source type” values may change as they navigate your site.

For example, if your site’s flow includes a step where visitors leave the page, go to a 3rd party site or different subdomain, and then come back to your site. In this case, they would have an original value (campaign, search, referral, or direct) until they leave your site. Once they come back, they will have a "referral" value because Optimizely detects that they went to another site.

What this means for you is that, on the Results Page, you may see the same visitor's conversions appear in different dimension values. For example, any conversions that visitor made before going to the third-party site or subdomain would count in "campaign," "search," or "direct." And any goals triggered afterward would count in "referral." This is because the visitor has left your page and come back, so Optimizely now considers the visitor a "referral." The conversions from before the visitor left the page will not be affected.

See below for an example:

Create a custom segment

Optimizely also allows Enterprise plans to create segments from specific Audiences or Dimensions.

You can choose up to 10 audiences and/or custom dimensions to use when you filter your results. These 10 slots can be made up of audiences, dimensions, or both. Dimensions are automatically added for segmentation, so if you add 10 dimensions, you will not be able to enable additional audiences.

To enable an audience for segmentation on the results page, go to the Audiences tab on the Home page, show below: 

Now, create a new Audience (or select from the Audiences you have already created) and click the checkbox for Segmentation.  

 

 
Note:

Audiences that are enabled for segmentation are persistent. This means that if a visitor qualifies for such an audience, she will continue to qualify for that audience every time she visits the page. If she visited multiple times, sometimes meeting the conditions and sometimes not, she’ll keep seeing the experiment.

This also means that when she sometimes qualifies for one audience that's enabled for segmentation, and other times qualifies for another audience that's also enabled for segmentation (which might have conditions that are the opposite of the first audience), she will show up in both segments on the Results page.

 
Important:

Custom segmentation is not compatible with audiences based on custom tags. To create a similar custom segment for your results, dimensions can be used instead.

Segment results on the Results Page

The UI limits the Segment drop down to 200 segment values. This limit prevents the UI from crashing.

You can filter your entire results page by an audience or dimension by using the Segment drop-down at the top of the Results Page.

You can also filter individual goals by an audience or dimension by using the Segment drop-down within the goal module. Note that you must do this while you're adding the goal in a new custom view (see below).

This means that if you wanted to compare your overall results by two different audiences to gain insights on how different visitors behave, you could create a custom view that shows the same goal twice, but segmented by different audiences: