A Step-By-Step Walkthrough for SAP Analytics Cloud compass (PART II)

[ChenNee]

This is part II of the two-blog series meant to give an overview of the typical steps taken while simulating with compass. This blog picks up from the part directly after having triggered a scenario run. Just as a recap, below you can see the structure of the two blogs:

Blog 1:

#1 – Starting a simulation
#2 – Configuring drivers
#3 – Simulating

Blog 2:

#4 – Interpreting the Results
#5 – Save Simulation and Scenario Comparison
#6 – Collaboration and Sharing

Let's dive in!

#4 - Interpreting the Results

Result components

Once the simulation calculations are completed, the probability distribution of the impact will be displayed in a graph, which is divided into 3 segments:  the optimistic, pessimistic and realistic case. By default, they are defined as such: Pessimistic (lowest 5% of the results), optimistic (highest 5% of the results) and realistic (90% of the results in-between). The definition of the percentages can be changed via the “case settings” configurations, but 5/90/5 percentage division is among the most common. The corresponding result ranges for each case is visible on the graph and also in the three boxes reserved for the cases below. You will also see a slider which is parked at the current baseline, which is the current value booked in the model for the target KPI, marked also on the X-Axis with “Baseline (current value)”

Understanding the results

Interpreting the simulation results require a look at all three main components together: (1) The graph, (2) the position of the current baseline including the percentages displayed, (3) and the results ranges in the boxes below.

In the example above, you can see that the current baseline is lying in the realistic area of the graph. This indicates that the current booked value in the model is within the attained results but not within the edge cases (the 5% highest or lowest value) and hence has a realistic chance of being achieved from a statistical point of view. The 16% on the right of the slider means that there is a 16% chance of achieving a value higher than the current baseline of 215.81 mUSD. But note that since the optimistic case (in the rightmost box at the bottom) only stretches to 246.91 mUSD, this means that there is a 16% chance to achieve a value between 215.81 to 246.91 mUSD. Similarly, for the 84% on the left side of the slider, it means that there is a 84% of achieving a value between 215.81 to 158.21 mUSD, which is the lowest value attainable in the pessimistic case.

Now some of you might be wondering why there is a slider in the graph. This slider actually allows the exploration of chances associated with a specific value. Let’s say if you are interested in what are the chances of achieving 200 mUSD under the conditions you have modelled in the scenario, simply drag the slider to the position of 200 mUSD or enter “200” in the input box. The percentages display on the left and right side will be recalculated and are to be interpreted as explained in the paragraph above.

#5 – Save Simulation and Scenario Comparison

The significance of saving a simulation

Once done with the simulation, you could proceed to save it in the file repository by clicking on the save icon next to the simulation or scenario name. Saving a simulation has two advantages: (1) It is possible to return at a later point in time and compare the results to the actual performance, enabling insight into the accuracy of the assumptions made and the results observed. If performed over time, this approach will promote understanding of the important drivers and how to simulate more accurately. (2) Saving a simulation allows the creation of multiple scenarios*, which can be compared against each other to understand the results obtainable via different assumptions.  
(* Working with multiple scenarios require the saving of changes in each scenario before moving on to work in another. This is to reduce the ambiguity regarding the actuality of the results when comparing the scenarios.)

Comparing scenarios

In the above example, you can see that it is possible to select another scenario via the dropdown in the header. The result of the selected scenario will appear at the bottom in the three boxes designated for the results of the pessimistic, realistic and optimistic cases. This is to enable the comparison of the results across different scenarios at a glance, giving immediate insight to how each scenario performs in each case.  

Helpful Hint

It is recommendable to simulate a first scenario with value ranges reflecting the uncertainties in real life to understand how hard the fluctuations would hit the business and where the company would stand in worst and best cases. After gaining insight here, it would make sense to create further scenarios with value ranges reflecting the assumptions of how the company could mitigate the risk with counter strategies. The scenarios could be then compared to assist strategic decision. Note that during scenario comparison, a scenario giving a worse result in worst case might deliver a better result in optimistic case, hence it is always recommendable to look at both ends of the result when assessing the impact.
Remember also to give each scenario a good description that describes the intention behind the value ranges given. As all scenarios could possibly share the same drivers, it will become difficult after some time to recall the intended strategy behind the value input for each scenario just by observing the numbers. If possible, try to start with a brief description first such as “investing in marketing in Asia”, and elaborate it as you move on. The description is going to be your best friend here; it is actually one of the most useful features, together with the graph and value ranges, to give clue to the strategy modelled.

It is also useful to keep in mind the following: Depending on the answers you are seeking using compass, the answer might be contained at times within the result graph, and at times within the value configuration in the driver list.
For example: In the first scenario with value ranges reflecting the uncertainties in real life, the answer you are seeking here is the possible risk distribution of impact, which is to be found within the result graph. However, when modelling counter strategy with further scenarios, the result graph of each scenario becomes more of a reference for the possible effectiveness of your assumptions. Here the answer sought is hence actually in the value configuration of the specific scenario with the most appealing result for you, giving insight to the ranges which should be applied for the chosen drivers to attain the acceptable risk context.

#6 – Collaboration and Sharing

Collection of assumptions across departments

When you are done with simulation, you might want to share your results with fellow colleagues or a decision maker. Or you might want to invite responsible colleagues to simulate together with you. With compass, it is possible to simulate with multiple colleagues together, or even across departments. The collaboration aspect is crucial, because more often than not, the knowledge of the driver uncertainties (i.e. the min and max value per driver) across different regions and products are not centrally located but is dispersed across different (global) departments. The collaboration feature here will enable a structured environment for collective enterprise simulation.

Private and Public scenarios

Maybe you have noticed by now that when you start your simulation, the scenario created is always a private scenario, meaning that it is only visible to you and no one else. This ensures discretion during the simulation process. Once you have decided on a scenario which you would like to show colleagues, you can do so by publishing a scenario AND sharing the saved simulation artefact with the intended colleague(s), just like when you would share a story or a model with them. Note that publishing a scenario is done within the compass UI (in the scenario header), and sharing is performed in the file repository. If the publishing icon is deactivated for you, it means that there are unsaved changes in your scenario. Simply click the save icon, then publish afterwards.

Colleagues, with whom you have shared the simulation, will only see the public scenarios, which is in a read-only state. If desired, they can create a private copy of the public scenario, add their value configuration, run the scenario calculation and then publish it again, either overwriting the current public scenario or publishing as a new one. Note that while it is possible to explore the public scenario results by e.g. moving the slider, these changes will not be saved.

---

This is all for now and I hope the step-by-step guidance could give you more insights and useful hints on how compass could be utilised to not only help you understand the risk at hand, but also how to mitigate it.

We are working constantly to improve compass! Do leave us a note in the influence portal to let us know what you feel could be the next enhancements. We monitor the portal closely for your voices!