• Home
  • Ideas
  • Setting Scorecards via plus or minus rules instead of absolutes

Setting Scorecards via plus or minus rules instead of absolutes

Related products: None
I'm boldly putting a big idea out here for feedback: For those of us using 0-100 scoring paradigms, I've often wished we could adjust Scores up or down by a given number of points based on certain criteria.





For example, I'd like to write logic that states that for every Overdue CTA, my Engagement score should have 20 points deducted. Or that for every 9 or 10 in the NPS survey that my Customer Satisfaction score have 10 points added.





Currently, I can sort of get to this point by writing a long series of Actions with corresponding criteria.





0 Overdue CTAs: Engagement Score is 100


1 Overdue CTA: Engagement Score is 80


2 Overdue CTAs: Engagement Score is 60


3 Overdue CTAs: Engagement Score is 40


4 Overdue CTAs: Engagement Score is 20


5+ Overdue CTAs: Engagement Scores is 0





That sort of works, but requires me to write 6 Actions with corresponding criteria. It doesn't, however, lend itself to multiple scoring scenarios. For example, I might want to deduct 20 for each overdue Risk CTA and but only 10 for each overdue Opportunity CTA. Soon the logic becomes too much for a list of actions because the criteria can quickly sprawl into too many variations to easily document (and there are some limitations on number of Actions in a rule, yes?)





I like the plus/minus approach because it's easy to understand, and it's easier to write one piece of logic.





It's not a "must have" and quite an adjustment in scoring theory from what Gainsight supports today. Even so, do others have a similar interest?
As of late as we have been coming up w more scoring criteria to the point the list is growing so quick w so many variables I was actually thinking about this as well. I think it would be a great addition.
If I understand correctly you want to only push these changes up or down to one scorecard? if that is the case you can break up your scorecards into two engagement scorecards. In the near future I think we will have the ability to have a scorecard that we can then have multiple scorecards under that one that rolls up into the scorecard of our choice that then rolls up to the main health score. 
I don't think that what Mathew is saying. I believe what Matthew is saying, and correct me if im wrong, is he's looking for a simplified way to score. Currently if you create a rule you are forced to set the score to a specific number. What we'd like to see would be assigning a +/- value to a rule instead. So you wouldn't need to make 5 copies of the same rule for 5 different scores. For example:





If having no engagement risk CTA's = 100 for engagement


and if a risk CTA can be assigned a value, in this case we'll say -20


then if there are 4 risk CTA's (-80) your total engagement score would be 20





Currently you'd have to create a long list of rules to get the same effect like Matthew mentioned, meaning that if you ever want to make changes you'd have to go through the tedious task of modifying a group of rules as opposed to changing a single value for specific parameter.





That was my interpretation of his ask. I'm not really clear as to how you'd get the same effect by creating multiple scorecards like you mentioned Andrew. Maybe you can elaborate to help me better understand if you can achieve the same effect based on what you described.
I see what you mean. Yes if there was logic that would allow you to write one action for +/- a count of overdue CTAs that would be nice. If you do not separate the scorecards though you will never know if it is and overdue CTA or any other factor pushing the Scorecard down. You can get around this somewhat with writing to the comments section for whatever action meets criteria but if you have two actions that meet criteria one will always have to overwrite the other with comments and scores. So the best action would be to separate the scores. In the near future I believe we will have the opportunity to have say one Customer Sentiment score and under that specific scorecard have a number of "sub"-scorecards that roll up to that Customer Sentiment scorecard. So what you could then do is write the logic in your action for each scorecard and anytime criteria is met just put in the Customer Sentiment Logic to "refer to sub-scorecards" and you can then have your comments write to that sub-scorecard so anyone coming in can quickly see exactly why the Customer Sentiment score has dropped. Hope that helps. 
Thanks Mathew and others. I think this is a great idea. The idea which you have proposed is very similar to several lead scoring algorithms out there where certain activities are given points and certain activities deduct points and the final overall lead score is driven by this logic.





There are pros and cons to this approach. 





The problem with points based approach is that it is very tricky to determine the number of points you want to add/deduct for a activity and with the example you have proposed above , the importance of a activity/measure need not be linear always.


Also , as Cummins mentioned above , if you are tracking Engagement Score as a measure and adding and subtracting points , then you would not know the reason why the score is high/low unless you track Overdue CTAs and Oportunity CTAs as individual measures.





We are planning to introduce scorecard groups which will help you track Engagement Score as a group and Overdue CTAs and Opportunity CTAs as individual measures under them so that will help you understand the scores better.





We will think through this and see if we can incorporate some elements from what you have suggested here.
Terms & ConditionsCookie settings