How to create a Value Driver Tree ?

In my previous article we discussed, how a value driver tree can help you to better understand implications on your KPI's. This time I would like to focus on a common approach to establish such a value driver tree and how you can start thinking about a value driver tree for your own business or your customer.

As an example for this thought-process I would like to use the value driver tree, which I showed in this video: Value Driver Tree for a Mining business.

How to get started ?

The first layer of the Value Driver Tree is probably the easiest one. Here you need to ask yourself, what the main objective of your Value Driver will be. What is the key question that you are trying to answer and gaining more insights on ?

• Are you trying to understand your profitability ?
• Are you trying to gain insights on your Operations Cost ?
• Are you trying to show your production rate ?

In our example around mining, our first Value Driver Tree is focused on the pure hauling part of the operations, so the main goal we trying to show is the amount of tons hauled and what the influencing factors are.

In our example, the main KPI is the "Tons Hauled" broken down for the next level into the number of hours that we can operate and the tons per hour we can haul.

How to get to the next level ?

After we identified the first level, we now need to create the supporting elements and the supporting structure for our overall value driver tree and we need to include the key drivers - but how do you get to create this structure and how do you find those key elements ?

This is the part where you get to ask a lot of questions and where you need to identify the key elements of your business, but perhaps there is an approach we can take.

What if we break down the different parts into smaller areas and focus on those to identify the key elements ?

In most cases there will be some form of a production, which might not be a full manufacturing site, but some form of a production process, and sales, and operations elements - so why not break it down into these areas?

Also in most cases there will be costs and revenue / income and if you wanted, you could also break it down into fixed and variable elements.

In addition you can see the Important Factors and Constraints rows and we will come back to those in the next step.

So if we take that approach, how would this then look for our Mining / Hauling value driver tree ?

The main cost for the hauling part is the equipment (the actual trucks), the fuel, and the labor costs. Because we only focus on the hauling part, there is nothing related to sales or a selling activity, nor is there anything related to revenue.

But we have some important factors and some important constraints to think off, such as the number of trucks and truck drivers, and the volume per truck. If the main KPI we trying to outline is the overall hauling, then those are very critical factors as the volume that we can haul per truck directly influences our overall hauling measure.

How to get to the details ?

As you can see these elements are clearly not enough details yet to establish our value driver tree. So how do we get to the next level of details ? This is where you need to get into the details of each individual KPI to understand the influencing factors.

• We started off with stating that the "Overall tons hauled" is our key element that we are trying to outline in the Value Driver Tree.
• We already identified several influencing factors and constraints, as well as cost elements, which we can now drill into.

Example #1: Volume per Truck.

Clearly if the overall hauling is our main KPI, the volume we can transport per truck is critical as it has a direct impact on our main KPI.

But ..... (remember to ask questions....) is the truck always loaded to its maximum capacity ?

So you can see now, that by digging into the next level, you can start creating the next level of your value driver tree.

Example #2: Tons per Hour

We already identified the "Tons per hour" that can be hauled as a KPI for our Value Driver Tree, but what does "Tons per hour" really mean ?

• We know how many trucks and truck drivers we have
• We know the amount of working hours.

...but I think we can agree that it isn't just the amount of working hours x the number of trucks x Truck capacity.

How about:

• The time that we spend on loading the trucks ? We have to subtract that.
• The time that potentially the truck is waiting in the queue ?

So this then brings us to this:

As you can see we started with the KPI on the "Tons per Hour" and we started to break it down into more details by also breaking down our overall approach into smaller steps.

What are your KPIs ?

Now before we look into, how such an approach can be implemented in SAP Analytics Cloud, lets talk a little but about the KPIs.

When you start outlining your top-level KPIs for the value driver tree (and also for other type of analytics projects) keep the following things in mind:

• Start with the problem (not the KPI). What problem are you trying to solve or gain further insights about.
• What are the associated business goals and how can we measure the impact ?
• What are the goals (are there company goals ?) for the KPI that you identified?
• Are there timelines associated with those goals?
• Who is responsible for the KPI?

These are just some high-level questions to think about, when you start identifying the KPIs for your value driver tree.

I hope that this was useful and in the next step we will take a look how we can use the process and implement a concrete example in SAP Analytics Cloud.