Analyzing B2C campaigns

Oracle Unity allows you to investigate problematic campaigns, identify the groups of customers that are responsible for the poor performance, and re-target the problematic customers. After identifying a campaign is not performing as expected, you select it along with similar campaigns to use as a baseline of comparison. Once on the Campaign analysis page, you do the analysis by looking at how different groups of customers in the investigated campaign performed compared to the same group of customers in the similar campaigns. This analysis allows you to identify which customers in the investigated campaign performed comparatively worse. The final step is creating a segment of the problematic customers so that you can re-target them in a new campaign.

Important: To properly set up your Oracle Unity account for importing and displaying analytics data, refer to the following:

Campaign analysis involves the following tasks:

  • Step 1: Open the investigation. Select the campaign from the One time campaigns page. You will also need to select similar campaigns to get a baseline of comparison.
  • Step 2: Analyze the campaign. Analyze customers in the investigated campaign across different attributes to find the ones that are the cause of the poor performance.
  • Step 3: Create a segment. Based on the results of the investigation, create a segment of the problematic customers to re-target in a campaign.

Step 1: Open the investigation

After noticing a one time campaign has abnormal performance, you can start investigating it from the One time campaigns page. You will need to select the campaign you are investigating as well as up to three similar campaigns to use as a baseline of comparison.

To start an investigation:

  1. Click the Oracle icon Image of the application navigation button. Use it to access the different parts of Oracle Unity. in the bottom-right corner to open the navigation menu.
  2. Select One time campaigns.

    3. Note: If you are on the homepage, you can also go directly to the One time campaigns page by clicking Details in the One time campaigns widget.

  3. Hover your mouse over the campaign you want to investigate. The investigate icon An image of the Investigate campaign icon will appear next to the far-right column. Click the icon to start the investigation.

    4. An image of the Investigate campaign icon

    You will see the campaign added at the top of the page. You will be prompted to add up to three similar campaigns for comparison.

    An image of the campaign added for investigation

  4. Find up to three campaigns to use for comparison. For each similar campaign, hover your mouse over the campaign. The add campaign icon An image of the add campaign button will appear next to the far-right column. Click the icon to add the campaign as a similar campaign for the investigation.

    5. An image of the add similar campaign icon

    You will see the added campaigns at the top of the page.

    An image of similar campaigns added for investigation

  5. When done selecting campaigns, click Continue. The Campaign analysis page will open.

Step 2: Analyze the campaign

Based on the campaigns you have selected for investigation, you complete campaign analysis by comparing the performance of the investigated campaign against the baseline performance of the similar campaigns. You select different metrics and see how customers across several attributes in the investigated campaign performed compared to the same groups of customers in the similar campaigns. Attributes you select for analysis are displayed in the comparison matrix, which displays the relative performance of those attributes. The bottom of the page shows the breakdown of performance and customer count for all of the campaigns selected for the analysis.

An image of the Campaign analysis page

An image of the comparison of the metric value and customer count

You can do campaign analysis by reviewing and working with the different elements:

After completing your analysis and identifying a group of customers you would like to re-target in a campaign, you can automatically create a segment of those customers on the segmentation canvas.

Campaign analysis indicators

The Campaign analysis page has indicators that allow you to keep track of your campaign analysis, view the metric value and customer count of the investigated campaign, make changes to the metric selected, and review the attribute values you have added for analysis. This section has the following elements.

An image of the campaign analysis details

Callout number one Campaign being investigated: The name of the campaign you are investigating.

Callout number two Metric of investigation: Change the metric you are currently investigating in the comparison matrix.

Callout number three Metric value of selected campaign and similar campaigns: Based on the metric selected, the metric value of the selected campaign and the average metric value of the similar campaigns.

Callout number four Investigation values: Values on Performance gap, Customer count, and Customer lifetime value.

  • Performance gap: The difference in performance between the campaign being investigated and the average performance of all similar campaigns.
  • Customer count: The total amount of customers in the campaign being investigated. As you add filters for analysis, the value displayed will represent the subset of customers in the campaign being investigated that meet the filter criteria.
  • Customer lifetime value: The total amount spent by a customer over the course of their lifetime.

Callout number ive Similar campaigns selected: The campaign names of the similar campaigns selected for the analysis.

Callout number six Showing customers with attribute: The breadcrumb attributes that you have selected as part of the investigation. Clear all will remove all selected attributes.

Callout number seven Update: Click Update whenever you make changes to the comparison matrix to refresh the data.

Callout number eight Create Segment: After finding a group of customers you would like to re-target for a campaign, click Create Segment to open the segmentation canvas with investigation details pre-filled for the construction of a segment for re-targeting. See how to use the comparison matrix and how to create a segment from the comparison matrix. Learn more about Creating segments from the segmentation canvas.

To review and manage the campaign analysis indicators:

  1. Confirm the name of the campaign you are investigating and the similar campaigns that will be used as a baseline of comparison.

    2. The names of the investigated campaign and similar campaigns

  2. To remove or re-add a similar campaign from the investigation, click the checkbox for the campaign.

    3. An image of the Edit selection button

  3. To change the current metric being investigated, open the drop-down list for Metric of investigation.

    4. An image of the metric selection menu

    You can investigate the following metrics:

    • Average order value
    • Click rate
    • Click to open rate
    • Conversion rate
    • Open rate

      • To see descriptions of metrics, review B2C Dashboard metrics.

  4. Review the breadcrumb attributes under Showing customers with attribute. These attributes represent the filters selected at any point in the investigation. The metric values will be updated based on the customer subset created by the filters selected. Click X in the attribute to remove it from the filters or click Clear all to remove all attributes

    5. An image showing the attributes selected for the matrix

  5. Review the Investigation values.

    6. An image of the metric value and customer count

    The Performance gap displays the following values:

    • Based on the metric and attributes selected for analysis, Performance gap displays the difference in performance between the customers in the investigated campaign and the average performance of the customers in the similar campaign. In the example below, female customers with an income between $151k and $250k had an open rate 5.5% lower in the End of Summer Sale compared to the average open rate for the same group of customers in the similar campaigns.

      • An image of the percentage difference in metric value

    • The value in parentheses displays what the values were at the start of the investigation before any filters were selected. In the example below, female customers with an income between $151k and $250k had an open rate 5.5% lower in the End of Summer Sale compared to the average open rate for the same group of customers in the similar campaigns. Before any filters were selected for investigation, the performance gap was 7.3% lower.

      • An image of the metric value for the investigated campaign

    The Customer count displays the following values: 

    • The number of customers within the selected filters in the campaign being investigated. In the example below, the total number of female customers with an income between $151k and $250k in the End of Summer Sale campaign is 100.

      • An image of the customer count for the filters selected

    • The value in parentheses displays what the values were at the start of the investigation before any filters were selected. In the example below, the total number of customers in the End of Summer Sale campaign before any filters were selected was 1.2K.

      • An image of the total customer count for the investigated campaign

    The Customer lifetime value displays the following values:

    • Based on the attributes selected for analysis, Customer lifetime value displays the customer lifetime value for the selected customers in the investigated campaign. In the example below, female customers with an income between $151k and $250k in the End of Summer Sale have an average customer lifetime value of $5.2K.

      • An image of the cusotmer lifetime value for the attributes selected for analysis

    • The value in parentheses displays the average customer lifetime value for all customers in the investigated campaign. In the example below, the average customer lifetime value for all customers in the End of Summer Sale is $4.8K.

      • An image of the customer lifetime value for the investigated campaign

The comparison matrix

The comparison matrix shows you how different groups of customers in the investigated campaign performed compared to the same groups of customers in the similar campaigns. This comparison is based on the following: 

  • The metric you select in the Campaign analysis indicators section.
  • The attributes of the investigated campaign you select for analysis.
  • The attribute values that represent different groups of customers in the investigated campaign.
  • The relative placement of attribute values showing if that group of customers in the investigated campaign performed better, worse, or similarly to the same group of customers in the similar campaigns.

The campaign matrix section has the following elements.

An image of the comparison matrix

Callout number one Attribute legend: The attributes you can add to the matrix. Click attributes in the legend to hide and show them in the matrix. You can add up to eight attributes to the matrix.

Callout number two Y-axis: Displays the difference in performance between the investigated campaign and the similar campaigns.

Callout number three Line of similarity: Groups of customers in the investigated campaign on or near this line performed equally to the same groups of customers in the similar campaigns.

Callout number four Investigated campaign performance: The line that represents how the investigated campaign performed on average compared to the average performance of the similar campaigns. This line will update based on the filters selected for investigation.

Callout number five Attribute values: Visual representations of the number of customers targeted in the investigated campaign. Larger circles represent a higher number of customers. The attribute names are listed underneath each group of attribute values. Hover your mouse over attribute values to view more information about them.

Note: The comparison matrix will only show data for each attribute type that has a value. If a customer has a null value for the attribute type then that customer will not be included in the matrix. This may cause a disparity between the total customer count and the combined count of an attribute's individual values. For example, there might be 1,000 customers in the campaign, but the attribute for Age group might add up to 970 customers for all individual age groups. The 30 "missing" customers from this attribute will not appear on the comparison matrix because they have no value.

Callout number six X-axis: Displays the customer lifetime value, which represents the average lifetime value of the customers within each sub-group.

To use the comparison matrix:

  1. Analyze the matrix by comparing the placement of attribute values in relation to the line of similarity. Attributes that are on or near that line represent customers in the investigated campaign that performed similarly to the same customers in the similar campaigns. Attributes that are above the line represent customers that performed better than the same customers in the similar campaigns and attributes below the line performed worse. Also consider the size of the circle and their impact to the total customer count of the investigated campaign. You can then select attribute values you want to investigate further.

    2. An image of the baseline average in the comparison matrix

  2. Manipulate the matrix view as needed by doing the following.

    3. Use your mouse wheel to zoom in and out of the matrix. When zoomed in, click and drag the matrix up/down and left/right.

    Hover your mouse and click Marquee zoom. Then, drag and draw a rectangle with your mouse to zoom into that specific area of the matrix.

    Click and drag the horizontal and vertical scroll bars to move the matrix.

    An image of the scroll bars

  3. Select multiple attribute values and add them as filters to the comparison matrix. After clicking an attribute value and clicking Update, you will see it listed above the matrix under the heading Showing customers with attribute. To remove the attribute value as a filter, click X.

    4. To select multiple attribute values, you can do one of the following.

    Hover your mouse and click Marquee select, Then, drag and draw a rectangle around the attribute values you want to select.

    Hold the Ctrl button while clicking the attribute values you want to select.

    Example: You notice that females in the investigated campaign performed considerably worse than the same groups females in the similar campaigns. After selecting the females filter for analysis, you also notice that females aged 25-34 and 64+ performed considerably worse as well, so you select them for further investigation. This allows you to see how specific group of customers performed in the investigated campaign compared to the similar campaigns.

    An image of the attributes in the comparison matrix

    An image of the filters in the comparison matrix

  4. Continue investigating attribute values and compare performance until you have identified the optimal group of customers you would like to re-target in a campaign.

Individual campaign breakdown

The bottom of the page shows a comparison of the investigated campaign and each individual similar campaign. You can compare the metric you have selected for analysis as well as the size of each campaign.

An image of metric values for campaign analysis

If a campaign has zero customers, then it will not be displayed in the comparison of the metric or the size of the campaign.

If the metric investigated or customer count of one of the similar campaigns varies greatly compared to the other campaigns, you may want to remove it so that you can get a more valid baseline of comparison. Click Edit selection at the top of the page to remove it from the investigation.

An image comparing the values of the campaigns

Step 3: Create a segment

After identifying the group of customers that you want to re-target in a campaign, move on the next task below so you can create a pre-filled segment with the conditions needed to target that group of customers. You can configure and save that segment on the segmentation canvas.

To create a segment:

  1. Click Create segment.

    2. An image of the Create segment button

    The Create new segment dialog displays.

    An image of the Create new segment dialog

  2. Enter the details for the segment.
    • Name: Enter a unique name. The name must be 1 or more characters, up to a maximum of 128. Other than underscores (_) and hyphens (-), special characters are not allowed. The first character cannot be a space. You can use characters from all languages supported in the language settings.
    • Description: Enter a description. This field is optional, but it is highly recommended to add descriptions for any entity created. This helps all other users get additional context when using and navigating Oracle Unity. The description can have a maximum of 512 characters with no restrictions on characters used. You can use characters from all languages supported in the language settings.
    • Base Object: Select the base object you are using to create the segment. You will view Master entities first followed by other data objects. The Instance admin user role can manage the list of available base objects by clicking Hide An image of the view hidden items icon for base objects that shouldn't be available for selection.
    • Tags: Enter applicable tags to organize the segment with keywords.
      • Tag names must be 30 or fewer characters. Other than underscores (_), special characters are not allowed. The first character must be a letter. You can use characters from all languages supported in the language settings.
      • You can add up to five tags to a segment.
      • You can search for tags on the Segments page under the Tags filter.
    • Click the checkbox to add the segment to your list of favorites.
  3. Click Continue. A new tab will open on your browser with the segmentation canvas.
  4. The pre-filled segment will be based on the attributes and metric you selected to investigate and will target the customers that did not perform the action defined in the metric. It will have the following attributes and conditions:
    • Campaign ID that matches the ID of the investigated campaign.
    • Profile attributes and conditions based on the attributes selected in the investigation. For example, the pre-filled segment may have the attributes Gender and Marital status, the operator Matches for both, and the values Male and Married.
    • Behavioral attributes and conditions that target the customers that did not perform the action defined in the metric.
      • Open rate: Includes customers that were sent a campaign email but did not open it.
      • Click rate: Includes customers that were sent a campaign email but did not click the link in the email.
      • Click to open rate: Includes customers that opened campaign email but did not click the link in the email.
      • Conversion rate: Includes the following customers.
        • They didn't purchase an item from the campaign email.
        • They purchased an item from the campaign email but they canceled or returned it.
      • Average order value: Selecting this metric will not create any additional behavioral conditions for the pre-filled segment.

    Review and confirm the pre-filled conditions.

  5. If needed, configure the conditions with additional criteria. Learn more about Creating segments.
  6. When done, click Save.
  7. Return to the browser tab that has the Campaign analysis page, click Clear all to removed all the breadcrumb attributes selected, and work on identifying additional groups of customers you would like to re-target. Then, click Create segment.
  8. When you have completed the campaign investigation, click the arrow button under the name of the campaign being investigated. You will return to the One time campaigns page.

You can manage segments you have created from the Manage segments page. Learn more about Managing segments.

After creating the segment, you can export it so that you can re-target the customers in a new campaign. Learn more about creating Creating export jobs and Creating campaigns.

Learn more

Managing B2C dashboard widgets

B2C Campaign performance

Creating segments

Managing segments

Creating export jobs

Creating campaigns

campaigns, campaign performance, campaign metrics, campaign analysis, B2C campaigns, B2C campaign performance, B2C campaign metrics, B2C campaign analysis