As a startup using HubSpot's new Lead Scoring feature to create a Contact Engagement Score, I've found the experience to be both promising and frustrating. The initial setup is user-friendly, with intuitive drop-down modals and easily weighted engagement properties.
However, the post-publication audit capabilities fall short of expectations, hindering our ability to refine and iterate on the score effectively. The ideal workflow for optimizing our engagement score would involve:
Creating a score with weighted engagement factors (bonus points if we were able to see the % of contacts that have taken a specific action/ or volume of action within the last 30 days)
Auditing individual scores in a test environment ( ie. Test 50 contacts to see if the score is prioritizing/ weighting correctly)
Analyzing which factors contribute most significantly to the cumulative weight
Iterating on the score or addressing underlying data quality issues
Visualizing individual scores on contact records, showing how contacts become hot, warm, or cold contacts.
Unfortunately, the current system lacks transparency in how individual scores are calculated, making it impossible to identify which engagement factors are truly adding value / which signals are within the score at all!
This opacity hampers our ability to refine our scoring model and ensure its accuracy over time.
To truly leverage the power of engagement scoring, HubSpot should consider implementing:
A detailed breakdown of score components for each contact
An activity feed showing weighted scores and their decay over time
Improved audit tools for analyzing score effectiveness across Contact segments.
I'm curious to hear from other HubSpot users about their experiences with the Contact Engagement Score feature. Have you found effective ways to audit and refine your scores, or are you encountering similar challenges?
I agree with some of your points but generally don't think that what you're mentioning stops this from being a powerful tool.
Creating a score with weighted engagement factors (bonus points if we were able to see the % of contacts that have taken a specific action/ or volume of action within the last 30 days)
The option to weight certain factors is built into the new scoring by simply awarding more points to some actions, fewer to others. If that doesn't work, can you give an example, why it doesn't? The second part is more a question about filtering through lists. Yes, it would be useful to have this directly in scores (as is the case with other points of yours) but this information is still accessible in other places.
Auditing individual scores in a test environment ( ie. Test 50 contacts to see if the score is prioritizing/ weighting correctly)
Whenever I need to do this, I simply create additional scores within the prod environment. Yes, it would be nice to have a test/dev separation but you can have different scores performing against each other and review them as long as you're not hitting your limit.
Analyzing which factors contribute most significantly to the cumulative weight
By definition, when creating a lead score, you're creating a black box to some extent. Again, it would be nice if the lead score surfaced immediately which factors contribute most. I typically solve this by creating a filtered contact view, sort it by score and adding columns for the referenced engagement factors, e.g. number of forms submitted, marketing emails opened, last seen etc. Not everything exists as a contact property but a lot of factors do and that allows for checking which factors seem to contribute most. It's an approximation but I find it useful enough as long as there aren't more sophisticated features: https://knowledge.hubspot.com/records/view-and-filter-records
Iterating on the score or addressing underlying data quality issues
See above. A filtered view would show you if there are contact where data points are simply missing. Iterating could be done through additional scores.
Visualizing individual scores on contact records, showing how contacts become hot, warm, or cold contacts.
Keep in mind that this new scoring feature is a relatively new product, I would expect to see updates to this if the product team keeps on receiving feedback like yours.
Hope this helps!
Karsten Köhler HubSpot Freelancer | RevOps & CRM Consultant | Community Hall of Famer
@Shane-Poyar adding to this thread, HubSpot just announced they're developing the below:
Lead Scoring
Preview Score Insights in Lead & Health Scores
February 6, 2025
Note
This record includes updates regarding current development efforts for our existing or new products or services. These updates are not intended to be a promise or guarantee of future availability of products, services, or features but merely reflect our current plans based on factors currently known to us. They also are not intended to indicate when or how particular features will be offered or at what service tier(s) or price. These planned and future development efforts may change without notice.
What is it?
Selecting the right rules and weights for a Health or Lead Score can be difficult. Today, it can even feel like a guessing game. Test a record helps you see what the score would be for an individual record, but what about for everyone else? Now, the score distribution preview samples thousands of records in order to show you what the impact would be across their scores.
Why does it matter?
Reviewing the average score and distribution of both the total score and threshold labels now makes score rule building an iterative and insightful process.
How does it work?
When creating a new score or editing an existing score, you can now access preview insights on the top right.
Insights will be provided based on the current rules configured:
For combined scores, you will be able to switch between 'Engagement' and 'Fit' criteria.
Who gets it?
Professional Customer Platform, Enterprise Customer Platform, Marketing Pro, Marketing Enterprise, Marketing+ Pro, Marketing+ Enterprise, Service Pro, Service Enterprise
Karsten Köhler HubSpot Freelancer | RevOps & CRM Consultant | Community Hall of Famer
The current score property being removed from HubSpot has been versatile in counting various metrics. For example, I use it to track the number of locations a client indicates they want XYZ in per deal. The new tool appears to focus solely on lead-scoring, potentially eliminating other use cases on different objects. What are your thoughts on this change?
I agree with some of your points but generally don't think that what you're mentioning stops this from being a powerful tool.
Creating a score with weighted engagement factors (bonus points if we were able to see the % of contacts that have taken a specific action/ or volume of action within the last 30 days)
The option to weight certain factors is built into the new scoring by simply awarding more points to some actions, fewer to others. If that doesn't work, can you give an example, why it doesn't? The second part is more a question about filtering through lists. Yes, it would be useful to have this directly in scores (as is the case with other points of yours) but this information is still accessible in other places.
Auditing individual scores in a test environment ( ie. Test 50 contacts to see if the score is prioritizing/ weighting correctly)
Whenever I need to do this, I simply create additional scores within the prod environment. Yes, it would be nice to have a test/dev separation but you can have different scores performing against each other and review them as long as you're not hitting your limit.
Analyzing which factors contribute most significantly to the cumulative weight
By definition, when creating a lead score, you're creating a black box to some extent. Again, it would be nice if the lead score surfaced immediately which factors contribute most. I typically solve this by creating a filtered contact view, sort it by score and adding columns for the referenced engagement factors, e.g. number of forms submitted, marketing emails opened, last seen etc. Not everything exists as a contact property but a lot of factors do and that allows for checking which factors seem to contribute most. It's an approximation but I find it useful enough as long as there aren't more sophisticated features: https://knowledge.hubspot.com/records/view-and-filter-records
Iterating on the score or addressing underlying data quality issues
See above. A filtered view would show you if there are contact where data points are simply missing. Iterating could be done through additional scores.
Visualizing individual scores on contact records, showing how contacts become hot, warm, or cold contacts.
Keep in mind that this new scoring feature is a relatively new product, I would expect to see updates to this if the product team keeps on receiving feedback like yours.
Hope this helps!
Karsten Köhler HubSpot Freelancer | RevOps & CRM Consultant | Community Hall of Famer
