Conversion Insights Dashboard with Revenue Data—FAQ

Estimated Revenue Lift calculates the incremental revenue gain attributed to Yottaa's optimizations by calculating the difference between Optimized and Unoptimized Revenue Per Session and extrapolating across all Sessions within a given time frame.  The dashboard presents Estimated Revenue Lift when the site’s traffic split is skewed towards either optimized or unoptimized sessions, as examining Revenue differences are not meaningful in a skewed traffic split.

Estimated Revenue Lift = (Optimized Revenue Per Session - Unoptimized Revenue Per Session) * Total Sessions

Estimated Revenue Lift is also the difference between Optimized Estimated Revenue and Unoptimized Estimated Revenue where:

Optimized Estimated Revenue = (Optimized Revenue per Session) x (Total Number of Sessions)

Unoptimized Estimated Revenue = (Unoptimized Revenue per Session) x (Total Number of Sessions)

and where:

Optimized Revenue per Session = (Total Optimized Revenue) / (Total Optimized Sessions)

Unoptimized Revenue per Session = (Total Unoptimized Revenue) / (Total Unoptimized Sessions)

Under some conditions you might see negative optimization impact values. This can be due to a variety of factors, and does not necessarily indicate that unoptimized sessions perform better than optimized sessions overall.

Verify the values for Sessions, Orders, and Revenue values for a period of time between Yottaa and your analytics vendor; they should be within 5%. If the data doesn't match, see requirements for tracking revenue data. If data matches and the site is optimizing the majority of traffic, there may not be enough unoptimized traffic to reach a meaningful comparison.

For sites that optimize the majority of their traffic, a negative Estimated Revenue Lift can be caused by a few large orders on unoptimized traffic. Because there are fewer  unoptimized sessions, a few large orders can have an outsized impact on Average Order Value, Revenue Per Session, and Estimated Revenue Lift.

You can spot larger outlier Orders in the Trend Chart by reviewing "Revenue Per Session Comparison" or "Average Order Value Comparison" by day or hour, as shown below.

You can select a longer date range if historical Revenue data is available or you can work with Customer Support to determine an appropriate percentage of traffic to leave unoptimized.

Orders and Conversions are similar but have different definitions:

A Conversion is defined as a Session that culminates in a user taking a desired action in response to a prompt, such as making a purchase, signing up for a subscription, or engaging with a piece of content. It represents a successful outcome aligned with a site’s objectives. Each session can only have one Conversion. Conversions can be tracked using a page URL matching method, or by using the data layer method and matching a corresponding parameter on the order confirmation page.

An Order represents the total count of purchase transactions completed based on unique Order IDs. A Session can have one or multiple Orders and is represented as one total Conversion. Orders are captured through the data layer of supported vendors on the order confirmation page. Learn more about Revenue Tracking.

When Conversions are used to track purchases, the Conversion count should be closely aligned to the Order count where differences are caused by sessions with multiple Orders.

Common reasons for discrepancies:

  • For a given date range, Conversion or Order data may be missing for some period. Double-check the Trend Graph to ensure both metrics are tracking within the date range.

  • If Conversions are used to track desired actions that are not purchases, such as subscription sign-ups, Conversion and Order counts will be different.

  • Different tracking methods. If Conversions are used to track purchases using the page URL matching method, counts may differ. For the most accurate results and to match other analytics vendors, it is recommended to use the data layer method for tracking Conversions.

For the most accurate results and to match other analytics vendors and systems of record, it is recommended to use the data layer method for tracking Conversions. This will also ensure that Conversions and Orders are captured in the same way. Inaccurate Conversions will lead to incorrect Conversion Rates and misleading insights on the dashboard.

  1. Find and edit the Order Confirmation Page under Settings> Page Categories.

  2. In the Client ID field, add “order-confirmation”.

  3. If there is a value in the Page URLs field, change it to “Full URL Contains” and type “no-url” into the field.

    This ensures that conversions are not counted by the Page URL match in addition to the Client ID.

  4. Make sure the Order Confirmation Page is at the top of list of Page Categories, so it gets matching priority over other pages.

  5. If the above process is completed accurately, the Order Confirmation Page Category should look like this:

Finally, to verify that Conversions are tracking properly through the Trend Chart in the Conversion Insights dashboard, select "Conversions" or "Conversions Comparison".