How to Get Started With Lead Scoring in Dynamics 365 CRM

Predictive lead scoring in Dynamics 365 CRM is a helpful feature to increase efficiency for salespeople, but it’s often underutilized.

As a Dynamics 365 solution specialist, I’ve helped many organizations use predictive lead scoring to effectively and constructively streamline incoming Leads to their Sales teams. I also warn our customers against some misuses of lead scoring to avoid.

Lead scoring can seem intimidating because it could include many variables, but in reality, it’s straightforward to set up, and you can configure it over time to adjust it to your business needs and your typical customers. Ultimately, you oversee how lead scores are generated, and it is up to you to establish the parameters for your organization.

Table of Contents

• How to Turn On Predictive Lead Scoring in Dynamics 365
• Manually Adjust Lead Scoring
• Multiple Models
• Lead Scoring Best Practices in Dynamics 365

Note: There are technically five different Dynamics 365 apps that serve various CRM needs, including Dynamics 365 Sales, Dynamics 365 Customer Service, and more. In some contexts, you may also see these products referred to as “Customer Engagement” or “CE” apps.

How to Activate Predictive Lead Scoring in Dynamics 365

To turn on lead scoring in Dynamics 365 CRM applications, you’ll need to have the System Administrator role applied to your User. You will also need to have Sales Insights enabled, which could vary based on your license types.

Another prerequisite to get started with lead scoring is at least 40 Qualified and 40 Disqualified Leads within the past 2 years.

Step 1

Navigate to the Sales Insights Global Settings in the navigation panel on the left-hand side and you’ll see Lead scoring under Predictive models of the Sales Insights Settings.

If you don’t see the Lead Scoring option, you may need a higher license type. Reach out to your Dynamics CRM Partner about licensing.

Step 2

Once you are in the settings for Lead Scoring, you can change the following values, but the highlighted boxes in the screenshot below are options specific to your organization if applicable. Then click “Get started.” If you are turning this on for the first time, it could take a while for the model to finish the training/setup process.

1. Enter a New model name
2. Leave Business process flow as is. (If you’ve created a different Business process flow for the Lead to Opportunity process, you can change the selection.)
3. Set the proper Qualified and Disqualified value. These should usually be set to the default, out of the box options, but if your organization has modified these values, make sure you select the proper status that indicates a “Qualified Lead” and “Disqualified Lead” in your system.
4. Filter Column and Train with leads can be set to the default options of “None” and “2 years.” You can set your scoring settings to filter out Lead records based on a column value. I don’t see a strong use case for this unless you want to apply the scoring to only a specific type of Lead.
   I would also recommend setting the duration to 2 years, because that the max time it can go back, more data the better for your scoring algorithm. However, if there was a data standardization or purge that took place, it can be set to a shorter timeframe.

Step 3

When the training is complete, it will give you a breakdown of the current model. If the data cannot render specific performance criteria, it can come back with a status: “Not ready to Publish.” This does not mean that you can’t publish or turn the Lead Scoring functionality on.

It means the dataset that was used to train the model for a Predictive Lead Score does not have enough to data to accurately display a confident Predictive Lead Score. You can see more details on what these metrics mean in Microsoft’s explanation.

Step 4

Navigate to the Performance tab to see more metrics. You can see more details on your model results there.

Step 5

Go back to the Configuration tab and review the bottom section of the Lead score grading. This is the organization criteria and assessment of a Lead Score based on letter grades. You can Edit these ranges to modify what the proper assessment looks like for your organization.

Step 6

Once you have reviewed the data and the metrics of your trained model, you can publish the model. Before you Publish the model, make sure you toggle “Retrain automatically”. This will allow your model to auto-update and train based on newly created and qualified/disqualified Leads on a rolling cycle of last 2 years.

The score analyzes values in not only out of the box but also custom fields. All data is incorporated into how a Lead was Qualified or Disqualified.

Step 7

Once it’s published any new Leads created will have a section for the Lead Score with tidbits that allow you to see any indicators that can improve the Lead score.

How to Adjust Lead Scoring Manually

You can go into the lead scoring parameters in Dynamics 365 and decide what factors will tick a lead’s score up or down based on your organization and what you’re looking for. Here’s how:

Step 1

Turn off “Retrain automatically” and you should see the “Edit Model” and “Delete Model” buttons as options next to Publish.

Step 2

Click on Edit Model and you will see a better breakdown of fields contributing to the Lead score. You can also set an option to ignore empty values of certain fields.

When you make any changes to the model, you will have to retrain it. Manually select each field that you would like the model to train for your lead scoring algorithm. When you click Retrain model, you will get a prompt regarding the grade ranges. You can aways come back to the grade ranges after the retrain if you need.

I personally like to leave “Retrain automatically” on for each model so that I can get a general basis of what the Lead score looks like within my organization.

I would come back to edit a specific model only after Predictive Lead Scoring has been turned on for some time within my organization, so I can better assess and understand the metrics of the data we are feeding the model.

How to Have Multiple Models for Lead Scoring

You can have multiple lead scoring models in Dynamics 365. This is helpful if you have a different subset of Leads you want to track a different Lead Score for.

You can only build up to 10 Models in an instance. Also, the system will warn you if you try to create a similar model. If you plan to have more than 1 model, I recommend setting a filter for the first model that’s different than the second. For example: you can train the scoring model to cast a subset of variables for Leads that are ready to buy now vs Leads that need additional nurturing.

Lead Scoring Best Practices in Dynamics 365

What Predictive Lead Scoring Should Be Used For:

1. Additional Insight: It can provide valuable insights that you might not have otherwise. For example, if you notice that leads with a score of 60 or higher have an 80% win rate, that may lead to considering what you can do to get more leads that score 60 from the start.
2. Efficiency: It can boost the efficiency of your team by providing small but significant metrics that help in fine-tuning your approach to leads.

What Predictive Lead Scoring Should Not Be Used For:

1. Measuring Team Member Success: It should not be used as a key metric to evaluate the performance of your sales team members. You would not want to use something AI-built to define which of your sales team members are effective or not.
2. Determining Sales Effort: It should not be the primary factor in deciding how much effort to invest in a lead. A low score does not necessarily mean a lead should be ignored. Different marketing campaigns and sources can skew the scoring, so it’s important to understand the context behind the scores before making decisions.
3. Predicting Revenue: As mentioned above, a high lead score doesn’t guarantee the lead will get qualified. This score should not be used to determine future revenue or loss of revenue.

Note that different campaigns from the marketing side will tend to drive different amounts of data into your system for a lead. Make sure to consider why leads’ scores are high or low before making important decisions based on them.

Just because the lead score is a 20 doesn’t necessarily mean that you shouldn’t follow up. The context of the score matters.

Lead Management Best Practices

Watch this recorded webinar to learn how to optimize your Dynamics 365 CRM processes with effective lead management strategies. Discover how to set up lead routing rules and utilize predictive lead scoring to streamline your workflow and improve efficiency.

Improve Your Sales Efficiency With Dynamics 365

By using predictive lead scoring in Dynamics 365 wisely, you can enhance your sales team’s effectiveness without over-relying on AI-generated metrics.

You can find additional information about Predictive Lead Scoring from the Microsoft Learn page.

As a Dynamics 365 CRM Partner, we at Encore provide support, training, implementations, and integrations tailored to your unique needs. If you’d like advice on how to best implement and leverage the features of Dynamics 365 for your organization, please contact us. We’d love to talk.