Lead scoring models for Communities?

Level 7

Lead scoring models for Communities?

Hi Marketo pros,

My company has been developing what is now a very robust customer community (I'm posting about my community on Marketo Community...so meta). I'm getting Munchkin code up on all the pages, and the next step will be scoring actions within the community. Of course, the end goal of scoring these leads is to determine when ours customers are ready for a cross-sell and/or up-sell conversation. Anyone else out there have experience with this? Marketo, do you score our interactions in the Community? ​ - do you know?

Any input would be appreciated. I'm excited about this project, and I'll let everyone know how it goes.

Level 10 - Champion Alumni

Re: Lead scoring models for Communities?

David Gaible

This is a really interesting use case.

My initial take is that you don't really want a lead scoring model for the community in itself; rather it sounds like you are talking about an Up-Sell/Cross-Sell scoring model, which would take into account engagement with your customer community as an important signal.  

In developing this model you would follow the same process as building a new customer scoring model, e.g., talking to your sales team about behaviors or attributes that make a customer a good prospect for up-sell, looking at your data to find those common attributes, and so on.

A v1 of this model might include standard things like email engagement, program successes (live events, webinars), social engagement (if you can track it) and community engagement (initiate a post, respond to post, or whatever else you can track).

You may need some custom instrumentation in the community beyond standard Munchkin to properly track all the behaviors that are of interest to you. I also think that because your community is new, it will take some time to determine how to properly score engagement there, as you will likely need to gather data on how these behaviors correlate with up-sell potential.

Once you have demonstrated some value, a v2 of this model might add in product usage signals from your platform. These signals can also allow you to check for negative behavior/inactivity, which is useful for predicting churn and defection.

One other tip - make sure to instrument any links in the community pointing to your website with tags and create tracking programs for them so that you can give the community source attribution for any touchpoints that occur as a result.


Justin Norris | Perkuto​