What have we learnt that causes FP&A leaders to loose their sh!t over their Anaplan modelling?

Poor model design!!

After more model health checks than we can remember and dozens of interviews with prospective, current and future clients these are the things that really grind their gears when it comes to their Anaplan projects.

🔸 Excessive mappings tables
🔸 Over extended core hierarchies
🔸 Lack of dimensionality
🔸 Flat reporting
🔸 Counterintuitive modelling structures

This is how we have discovered over the course of designing, developing and deploying dozens of FP&A models powered by Anaplan how to overcome these headaches -

✅ We learnt early in our journey with the platform that by utilising shallower hierarchies, which we refer to as shadow hierarchies allows modellers to create more intuitive forecast input structures, calculations and dynamic reporting.

✅ We discovered that key to vastly reducing the volume and complexity and the mapping required across a model is to always use common codes when building lists which are part of a family of parallel hierarchies. This allows modellers to auto-populate mapping tables allowing data to be mapped across them with ease.

✅We learnt that by designing and building parallel hierarchies users are able more naturally manipulate model data and navigate reporting to better suit their needs. Input structures that are built using shallower hierarchies are better able to align to the process being modelled rather than the process being shoehorned unnaturally into a large and overextended single hierarchy.

✅ We found that clients prefer to sacrifice model resources in the form of excess sparsity ( where dimensions are combined to form structures where intersections of lists contain not data ) when reporting can be designed to be more intuitive, responsive and dynamic. Flat structure are often wrongly used in a misguided attempt to keep the model efficient but can leave the end user feeling frustrated when incorporated into the final reporting layer. We found that when we 'spend' our sparsity dollars in creating fully or partially dimensioned reporting our clients have a more intuitive and agile experience.