Recently a user contacted support about a problem they were having. They had created a new Percent aggregation type Measure, and wanted to set up a Fixed Target.

In the past they had been setting up Additive Aggregation type measures and had good results with their Fixed Targets using the Fixed Target with Aggregation feature. This allows you to set a fixed target and to have that target automatically weight properly regardless of how many rows of measure data are aggregated. So a target of 75 units at a branch level automatically is multiplied by the number of branches, so you would get 75 x the number of branches in a grouping. That gives you proper, balanced targets against your actuals when using Fixed Targets.

So naturally they assumed they should use Fixed Target with Aggregation feature for their new Percentage aggregation type measure. But when they ran their reports, everything looked wrong. The percent targets were being added up! They contacted us.

Little did they know, if you’re using Fixed Targets, it really only makes sense to use Fixed Target without aggregation for Percent and Ratio type measures. That’s because percents and ratios automatically balance, due to the nature of the way they are aggregated. So you do not need to aggregated the Fixed Target values; they automatically make sense without aggregation being applied. If your target is to resolve 60% of your Call Center calls in one day, a Fixed Target means you want that 60% to carry all the way up the dimensional chains, regardless of how many Call Centers you are tracking.

Of course, if you want each Call Center to have a different Target, you know that you don’t used Fixed Targets at all – you load separate targets for each Call Center.

So why don’t we force all Fixed Targets for Percent and Ratio aggregation measures to only use non-aggregating Targets, and all Additive to use only Fixed Target with Aggregation? Well, we could, but it would mean we’d force all of you to work that way. The way we have it now, you are free to do it however you want, but of course you can make an error of having your Additive measure use non-aggregating Targets, when you might want them to aggregate, and your Percent and Ratio measures using aggregating Targets, when you don’t want them to. We leave the control in your hands, unless you beg us to take it away...

What do you think? Should we take the choice away to avoid errors? Can you think of any reason we should leave it as is?