Lead scoring can be temperamental, because if things happen in a different order than planned, someone could miss going through an MQL campaign, or could be scored before their data is enriched. I have a few helpful tricks that can help make your lead scoring process a little more reliable.

Track recent engagement with “Positive Behavior Score” field

Have you ever noticed in Marketo that (without anything custom set up) you can’t filter on positive score changes? That means that if you have any negative behavior scoring set up, you won’t be able to use the “Score was changed” filter to find people with positive behavior score changes.

To get around this, you can create a field called “Last Positive Behavior Score Date” which you update with the date in the flow of every positive behavior scoring campaign. (leave it off any negative scoring campaigns).

This is a useful field for creating smart lists of people who have recently been active, as well as making sure someone does not MQL from a negative behavior score update.

Helpful hint: If you add this field to your MQL model, make sure you set this as the first data value change of your scoring campaigns, so that it will be populated before the score updates and triggers the MQL model.

Demographic Scoring Controller

If there is a lot of negative demographic scoring in your instance, this one’s for you! Let’s picture this scenario:

Here’s a scenario to think about:

A lead fills out a form on your website, so they run through the scoring campaign and get a behavior score of +50.

Your MQL Model has a trigger for behavior score changing to 50 or greater, but there’s also a filter on there saying demographic score must be at least 10.

The lead got a demographic score of +10 for being in a target industry, so they run through the MQL campaign.

But their demographic scoring wasn’t complete yet! They just got -5 points for their job title because they are an individual contributor.

Now sales has just been assigned a unqualified lead and they are not happy.

It’s a scenario that I’ve seen before, and while there are a few different ways to prevent it, my favorite trick is to create a Demographic Scoring Controller program. It looks like this:

• Create a boolean field called “Demographic Score Complete”
• Create a smart campaign called “Start demographic scoring”
• In the smart list you’ll have triggers looking for any changes to score-able data such as Company Name, Country, Job Title (anything you are demographically scoring).
• In the flow of the “Start demographic scoring” campaign, you:
• Change Demographic Score to 0 (we need to reset the score because people can run through multiple times)
• Set Person Score (if you use it) to {{lead.Behavior Score}}
• Set “Demographic Scoring Complete” to “False”
• “Request Campaign” to request the first demographic scoring campaign, let’s say it’s for Job Title.

• The Job Title scoring campaign would have a trigger for “Campaign is Requested” and then in the flow, score based on job title, and request the next demographic scoring campaign.
• Keep requesting one demographic scoring campaign after the other until you are done, and then in the final campaign’s flow, set “Demographic Scoring Complete” to "True”
• Then, in your MQL campaign, you can use “Demographic Scoring Complete = True” as a filter, and as a trigger. That way we can make sure people go through the MQL model only once their demographic scoring has finished.

If you use “Demographic Scoring Complete = True” as a trigger in your MQL model, you might worry that people with old behavior scores will become MQL. Well that is where the Last Behavior Score Date comes in handy! Use it as a filter to say that they had to have gotten scored in the past 24 hours, 3 days, or whatever you feel comfortable with.

Demographic Score Field Completeness

If you want to nerd out on your scoring data, this one is for you! To track how many demographic fields that leads are scored on, you can create a Score field called Demographic Score Field Completeness.

This allows you to look at reports of completeness percentage (using a report level formula in SFDC) to see how much of your database is actually getting scored on all the possible fields. Then you can use that information to see where you may have gaps in data or even implement more complex MQL rules for those who are missing demographic data.

Here’s how it works.

Let’s say your demographic scoring scores on 5 different fields: Industry, Job Title, Revenue, Target Account and Company Size. In each of those scoring campaigns, put a “change score” flow step for the 'Demographic Score Field Completeness score and add +1 when a lead is scored.

Then if your lead is missing Job Title, but gets all the other attributes scored, they will have a Demographic Score Field Completeness score of 4, out of the possible 5. Now, you can run a report and see how many of your leads have been scored 5/5 for Demographic Score.

Now you’ll be able to determine how many leads have gotten scored to their full completeness. If a good portion of your database has less than 5 attributes scored, you may decide you need to get data enriched, or reduce the behavior score threshold to MQL for those with less than 5 attributes scored. The possibilities are endless!

That’s all folks!

I hope these scoring tricks are helpful! Does anyone else use any other neat scoring tricks to help account for different scenarios? Let me know on Twitter @christiemont @fwdthinkingb2b!