Use "double drop-down" in dimensions of tables for a simple drill-through analysis


Celonis offers a very powerful function which is the drop-down function to replace a dimension in tables graphs etc. We noticed quickly that many tables we were setting up were repetitive. Or trying to have two parallel tables to to a drill down to product and supplier.
This was really inconvenient for us.
For example many KPI / PPI are product dependent. So every company will say that a global KPI is wrong as a comparison since the have a complete different product portfolio. You are comparing apples with pears!
This criticism is valid and should be avoided.
We now as a standard use double drop-down in tables.
You than can switch in these two columns to exactly the drill down you want.
For example when we want to compare company and Rework or Touchrate for the same product, we take in the first column the product hierarchy and in the second drop-down the company code.
Also to understand where we still have problems concerning PPI , this breakdown is a must.
It is brilliant and simple to implement.
Now we can easily show were the weak spots are and show show results which are having the correct basis…

Have fun!

Best regards,


Hi Hans,

Yes we have been thinking to use it as well - and some Celonis use cases besides O2C Monitor (e.g. P2P, Logistics, Inventory, etc.) are using it.

I think it is very helpful to have this additional and easy to understand visualization!

What I would like to learn more about though is:
Who is comparing apples with pears? People who understand that it should not be compared, or people who do not understand it and would like to make themselves important :wink:

Our take is: if somebody wants to compare apples and pears (in Siemens it’s always oranges), then let them. These kind of people are a dying breed anyhow. However, we still should be able to answer to them and there is no better companion than standardization and 100% visualization of reality. Whether this is meaningful, is decided by humans. Humans derive value - or not.

Furthermore it would be interesting to understand better who your audience (users, receivers of data, etc.) is: in our case the users usually have anyhow only their own data in scope (e.g. a person from Country X sees data of country X). The people who have extended authorizations like audit, IT, etc. must be careful in their assessment of reality. E.g. if they say “Country B has 10% more rework of Country A (and they read it off the Celonis platform), then it is correct. If they say Country A has a better process because everything else being equal, they have less rework, would be wrong! I think the fear of comparing apples and pears and oranges come from the fact, that people who understand are just sick and tired of having to explain. They probably do understand the nuances and that many things are in the eye of the beholder and that the devil is in the details :wink: .

In any case, I would always favor 1000 manual rework steps with 1 rework activity over 50 manual rework steps due to 25 rework activities.

Let me conclude with one of my favorite poems:

The Road Not Taken


Two roads diverged in a yellow wood,

And sorry I could not travel both

And be one traveler, long I stood

And looked down one as far as I could

To where it bent in the undergrowth;

Then took the other, as just as fair,

And having perhaps the better claim,

Because it was grassy and wanted wear;

Though as for that the passing there

Had worn them really about the same,

And both that morning equally lay

In leaves no step had trodden black.

Oh, I kept the first for another day!

Yet knowing how way leads on to way,

I doubted if I should ever come back.

I shall be telling this with a sigh

Somewhere ages and ages hence:

Two roads diverged in a wood, and I—

I took the one less traveled by,

And that has made all the difference.


Hi Thi,

Thanks for your remarks and questions.
The double drowns are actually a necessity for us.
We have about 12 entries in each to choose from ( e.g. company, company type, sales organisation etc., product, customer, process type, supplier, etc)
The reason is simple: our cockpits contain data of 50 sales units and 8 factories. We have many process types from classical Third party deliveries, sales from stock, Return Process, debit and credit memo, all service related processes and even MTO. We took everything from the VBAK and downstream from SO related Purchasing and Production (MTO) or upstream to the service order into the maintenance process.
To avoid many different analysisis which needs a lot of maintenance, we use these drop-down and according filters.
It gives the user a possibility to see the trees again in the wood. For us (cockpit designer) it is all about a efficient process discovery and simple root-cause process. Our user get lost in the wood otherwise.
For us this was one of the essential steps to understand the many millions of SOI cases and find the real interesting hot spots.
We do also the same for activities based tables. We see immediately the hot spots for example to understand that one figure is actually meaningless without the right context.
For example 70% of automation across all companies sounds great. With the double drop-down we see a huge spread for example based on process type. You can slice further and find the real hotspots to focus on (or even conclude not to do anything). It thus accelerates the process discovery.
A nice by-effect: all drill through data is in one table which you can export easily to excel which is extremely handy.
But of course when you already can see the trees clearly there is probably no need.

Important Info:since Version 4.3 it is much more easy to set up these drop-down…