Once your team has created a scoring algorithm, how do you determine if the scores are meaningful as you intended, and if they are not, how do you improve the algorithm so they are? While adjusting your algorithm may seem to be the obvious and simple choice, it is not without its complexities. First you need to find out if the scores are meaningful – Are they telling you and your Sales Team what you expected?
A tried and true approach to measuring the effectiveness of your Lead Score is to obtain an extract from your CRM system for a statistically significant period of time, and measure the results against the Lead Score. There are two elements of this analysis that are relevant: the time frame of the CRM extract, and the timing of the Lead Score against which the result is measured. We’ll discuss the first of these elements here, with the other in a subsequent posting.
When gathering data to measure, you need to ensure that the timeframe of the data extract will be meaningful for the lead score being compared. For example, if your sales cycle is historically six months, and you are looking to measure Lead Score effectiveness based on Opportunities that have been successfully won, the data you gather needs to be over a time period long enough for the measurement to have meaning. It would not be relevant to implement your Lead Scoring program and then three months later try to measure its effectiveness, since you know it takes, on average, six months for the leads to result in closed deals. On the flip side, if your sales cycle is two weeks, you might be able to gather enough relevant data in as few as eight weeks to make your sample significant from a statistical perspective.
Bottom line – make sure you have enough data over a relevant period of time to ensure your analysis will meet statistical guidelines for relevancy. It would be a waste of time to try to analyze your data without enough statistical relevancy.