Estimating Account Potential

Once you have matched your commercial demographics establishment data to each of your sales accounts, you can tackle the much easier job of estimating potential. This can nearly always be handled completely by machine.

Some commercial databases will offer their own proprietary estimates of spending in a particular product category. In this case, you might think that the job is done: Just use their estimate for each customer account.

More often, however, the commercial database estimate is very rough and can be significantly improved. Your company sales and customer database, combined with some of the commercial elements, can be analyzed to determine which elements are most important in predicting sales potential. Regression analysis is the standard tool here.

For example, your sales may be strongly correlated with a customer's employee count. We often find that a sales rep's knowledge of account employment is more accurate than counts from demographic data. This brings up a very important point:

Validate Your Demographic Data

All data that you use in estimating sales potential should be confirmed or corrected if necessary by the account's sales rep. Taking this often-missed step gives the rep a chance to see what you have and to provide any input that might be helpful. Reps must trust the data you are using.

Another step you can take is to calculate account penetration (as a percentage of estimated potential) to see if any accounts are out of reasonable bounds. You probably shouldn't have many penetrations over 100%, or even many that exceed some high limit based on your market share. Outliers need to be checked carefully to see why the penetration is out of whack. Account potential is usually the problem here unless you have a sales data error.

A common source of such errors is a mismatch between demographics locations and sales locations. Sales data is often full of special ship-to locations that are actually part of a master location's sales. Sales booked against these related account ship-to's must normally be rolled up into the master location sales.

Potential vs. Penetration

Account penetration is one of the two foundation data items that we use for territory planning, the other being account potential. You must have both to plan effectively ...more...

 

Validation Is Critical

Almost every dataset will contain some number of errors. These may be far from evident prior to analysis and may be detectable only once some results are available. We have noted account penetrations of 100% or more as an example.

This is one of the most important reasons for getting sales or account rep input and confirmation. Doing so not only brings them into the analysis but helps establish credibility of the data.

They should be asked to highlight any sales potential and penetration estimates that appear to them to be wrong. You can then look into each of these to see if an error source can be identified.

Sales people have a pretty good sense for what is real and are often the best way to validate critical data like this.