When Marketo setup our instance they imported an example scoring campaign to grow & decay lead scores. It was spot on and very useful. But I noticed that they grow & decay both Lead Source as well as the corresponding Behavior or Demographic score... At first I thought the Decay score would only decay the Behavior & Demographic numbers which would leave Lead Score as an aggregate count of scores, but no, it decays as well. Does anyone have a use-case on why the other field is also there?
Lead Score is not an aggregate field of both Behavior and Demographic scoring fields. You must have both (Lead Score and Behavior or Demographic Score) in all your scoring smart campaign flows for the totals to add up correctly. Make sure you use your scoring tokens or it will be a beast to manage.
So you can edit your lead scoring model if it's not working for you--it's definitely not a 'one size fits all'. What issues have you encountered with your decaying lead score?
One other thing to consider is that a Total Score which is the sum of the Behavior and Demographic scores can be misleading. You can have leads that have low Demographic scores, but high Behavior scores that are tire kickers that download a lot of resources, and their Total Score is above other more qualified leads. Similarly you can have leads with high Demographic scores, but low Behavior scores because they only download one resource. A better solution may be to create a two dimensional lead score (e.g. A1 through D4) where the Demographic Score quartile is represented by the numeric character and the Behavior Score quartile is represented by the alpha character. The graphic below shows those in red being highest priority and those in yellow being lowest priority.
On a related note, I saw a neat presentation that multiplied the behavioural and explicit contact request scores to create a smooth exponential curve based on behavioural data and explicit contact requests. Demographic data was included in this process, but mainly used in the routing process of the leads/MQLs.
This particular client had a great problem of too many top of funnel leads. The exponential curve allowed them to could control the rate at which leads were routed through to tele-sales or account executive teams by altering the threshold score. This process also solved the problem of having a strangely large bucket of 'A1 leads', which should be reserved for only the best of the best!
I'd like to learn more about this multiply transform.
You have the ability to increase the score threshold that results in someone being assigned an 'A' or '1' behavioral / demographic rating. We try to set the score thresholds so that they result in a relatively even distribution across the 16 lead rating categories.
You have to score on numbers, and then have a smart campaign that converts the behavior and demographic scores into Alphanumeric and then have that alphanumeric score sit in a different field.
I don't have access to an instance at the moment, but essentially you need to set up a number of smart campaigns that say 'When Demographic score is between x and x' AND 'When behavioral score is between x and x' then Change Data Value 'Lead Rating to A1' for example.
I hope this helps.