Understanding Multiple Predictive Scores

  • Updated

Multiple predictive scores allow you to score your accounts according to their fit and interest for different product offerings. For example, your company may offer multiple products, each with their own go-to-market strategy. Setting up multiple Qualification or Pipeline Predict Scores let you score an account’s interest and engagement, qualify your top-of-funnel accounts, or identify new accounts for each of your products.

Multiple Qualification Scores

When setting up multiple Qualification Scores, you can train each new score from your past pipeline and customer accounts while also taking future possibilities into consideration. If you’re looking to move from mid-market to enterprise accounts or planning to sell to new industries, you can set up new scores that build ideal customer profiles from a list of aspirational accounts that represent the new market segments you’re targeting.  

Before setting up multiple Qualification Scores, review the following articles:

You should also take the following prerequisite steps before setting up additional scores:

  • Identify the products for each new score.
  • Identify a set of accounts to use to train each new score. The model evaluates other accounts based on their similarity to the example accounts you provide. You can provide accounts that have become pipeline or customers for a particular product or aspirational accounts for the new market segment you want to pursue.

Multiple Pipeline Predict Scores

With multiple Pipeline Predict scores, your sales and marketing teams are not only able to identify and prioritize accounts that are soon to become net new pipeline opportunities, but also which type of opportunity they are likely to become. Your sales team will know the right product areas to focus on before each prospect call and your marketing team will be able to build target account lists for campaigns with product-specific messaging.

Before setting up multiple Pipeline Predict scores, review the following articles:

You should also take the following prerequisite steps before setting up additional scores:

  • Identify the products for each new score.
  • Identify the activities that differentiate interest in each product, such as webinars, campaigns, site pages, and keyword searches. The model uses these activities as scoring criteria.
  • (Optional) Create activity segments for each of your products. When setting up new Pipeline Predict scores, you can use the Engagement activities section to add activities that show interest in your products. For each new score, you can do one of the following:
    • You can leave the section blank and let machine learning identify activities. This is the easiest option but the activities are not as specific.
    • You can input the activities you identified for each product. This option provides more specific activities but is not reusable.
    • You can create activity segments for each score you plan to set up. This is the recommended option as it allows you to have a more powerful and consistent segmentation tool across all Demandbase solutions by reusing the same segments for each group in analytics, reports, advertising, and personalization.
      Create Segments and the following Activity Segment Example for more information.

Activity Segment Example

Here’s an example of an activity segment we created for the Demandbase Advertising product. To define which activities show an interest in Demandbase Advertising, we created a new Segment named Product Interest and a Segment Group for each of our products, such as Advertising

Our Advertising segment group includes activities that show an interest in our Advertising product such as Keyword Intent and Visited Web Page.

After you create an activity segment (such as Product Interest), you can add it as a Selector for Engagement activities in the Pipeline Predict Setup page when setting up a new score. 

Tip: To use the activity segment, in the Engagement activities section, search for its name (in this example, Product Interest) and select the Value for the segment group for the score (in this example, Advertising).


Was this article helpful?

2 out of 2 found this helpful