Product Q&A

To whom it may concern, I am part of a group of master students of the Technical University Munich currently researching the business model of Celonis. The company and how it operates is tremendously interesting, but it is difficult to find details on certain aspects of the company. Here, we are trying to reach out to you whether we could come in contact with someone who can give us some insights into how certain parts of the business operate. This would of course not be any sensitive information, more in regards to topics such as how revenue is generated across the variety of Celonis' activities and how Celonis conducts itself with customers and partners. Please let me know if anyone would be willing to talk to us or have an email conversation.  Thank you in advance,Luc

Hi peers,I try to set up new components using CALC_THROUPUT. The two following components show different results. This is presumably caused because the OLAP tables have multiple lines with single CASE_KEY though the process pattern only occurs once (and it should since we use FIRST_OCCURENCE).  It won't change when we use Attribute selection overhead to limit to single KUNNR or give KUNNR in Dimension. Why OLAP returns several line items? Single KPI (Number)<KPI>CALC_THROUGHPUT(FIRST_OCCURRENCE['Create Sales Order'] TO FIRST_OCCURRENCE['Create Delivery'], REMAP_TIMESTAMPS("_CEL_SD_ACTIVITIES"."EVENTTIME", DAYS)) OLAP (identical with single KPI)<KPI>CALC_THROUGHPUT(FIRST_OCCURRENCE['Create Sales Order'] TO FIRST_OCCURRENCE['Create Delivery'], REMAP_TIMESTAMPS("_CEL_SD_ACTIVITIES"."EVENTTIME", DAYS)) 

Hi all, I‘m desperately looking for a way to start a data job directly via a Python statement within the ML Workbench (IBC). I succeeded in starting a data model load via the api_request() function providing the reapective pool and model ID‘s but couldn‘t implement a similar logic for a data job execution on the ML Workbench. Could you help me out with the correct function and schema on how to do the call? Thank you very much for your help! BR David

Good afternoon everyone, We've been looking at our Delta Loads and Full Loads runtimes to try and optimize our Data Consumption and we've noticed that the Delta Loads always take more time than the Full Loads. Is there a reason for this? Is there something we can do to have them be faster? In the below Image you can see some examples of /COCKPIT/THDRV and CDPOS tables which are quite large  Please advise Thank you in advance for your support

I have two tables "Ta" and "Tb" in join directly, Tb is the master case table and the relation is "Tb" 1 -> N "Ta" . In formula Editor inside the number component COUNT_TABLE("Ta") = 1305 and COUNT_TABLE("Tb") = 7144 and all is correct. When I create a new number component with COUNT_TABLE("Ta") / COUNT_TABLE("Tb") I retrieve the number 1.2219..... I don't Know Why! I expect 1305/7144 = 0.1826. Components are without filters. How can I use multiple COUNT_TABLE in the same formula?Thanks,Alex Giorgini

Is it possible to set delta loads from SAP in such a way, that when new column is added, the extraction won't fail, but will just add the new column to the list of the columns and treats it as unselected?

Sustainability Base Camp Sustainable business execution is more important than ever. We believe that the EMS has the capability to help create more sustainable processes. To get ideas rolling, we've created the Sustainability Base Camp, a 2-hour training track, which aims to give you a broad overview of what matters in terms of environmental, social and economic sustainability.  Begin your journey today to get the conversations started! Use this discussion forum to exchange your thoughts on the questions posed throughout the track.  Remember that feedback to the track itself is best handled via the feedback form on the Academy :)  

Good Morning : Preparing to install on perm Extractor , and the we would like to use Windows Server. The documentation refers to 2012 R2. This appears to be out of support , will 2016 R2 work as a data base?.

Hi, I am looking into setting up a dashboard to monitor contract leakage in the form of paying a higher price for a specific item than agreed in a contract. 1.) I would like to obtain a count of purchase order items that were purchased outside of an existing valid contract.2.) I would like to obtain an indication on the cash value (difference between PO price and contract price) In our data model EKPO consists of both, purchase orders (EKPO.BSTYP = F) and contracts (EKPO.BSTYP = K). Now, I would like to compare the material number (EKPO.MATNR) for all EKPO.EBELN where EKPO.BSTYP = F with the material number for all EKPO.EBELN where EKPO.BSTYP = K. Whenever there is a match:1) Check if the contract number (EKPO.KONNR) in the PO (EKPO.BSTYP = F) equals the contract number (EKPO.EBELN) of the respective contract (EKPO.BSTYP = K) If not, the item was purchased outside of the frame agreement. If EKPO.KONNR empty, the item was also purchased outside of the frame agreement

Hello team, Our certificate is going to expiry today at 18h15 GMT and we need a new certificate. Can you help us with this? Best regards,Diogo

We are trying to access data model for an analysis and are using python script to connect and access data from backend. We are using Celonis cloud environment ‘TryCelenois’. We have created an apikey in user profile and using same ‘apikey’ during the login process. We have installed few sample applications from ‘EMSStore’. These applications has preconfigured analyses that we are using for our analysis. We are not using any custom data connector and instead using Celonis provided demo data model to provide data to these analyses. We can view data and perform all kind of analysis from the UI on these analyses. However, when we are not able to access datamodel object when trying to access it from external application through ‘pyCelnois’  We are using ‘pycelonis’ package to establish connection and extract data from Celonis. We are able to successfully connect with the Celonis system and able to extract the required analysis from the analyses collection. Though we are able to extract most

Hi,Is there some kind of limit how many columns can be visible in the column chart? I have a table, in which I wanted to visualize activity count 13 types of cases per date. When I create 10 of them, the column bars are barely visible. If I add next kpi, then the bars disappear. Is there some option to show all 13 kpi at the same time in column format? I know I can change some of them to line axis, etc. but would prefer columns for all of them

Hi Team,  I have a conceptual question here. We are importing data through flat files from 3 different souce systems namely : MDG, ECC and PLM. Need to identify a field which can be taken as a common reference which will further help in defining Case key for the process. The process is around Order to Activation of SKU in SAP..Any help would be much appreciated, or someone who would have dealt with similar use case. Eg : in a generic P2P process, #PO Items could be used as 'Cases' across the flow, trying to identify something similar in my current process.

Can task templates be shared with multiple data pools?

Dear Celopeers, On December 8th at 5PM CET, we are hosting a webinar with our product experts to dive into the technology behind automating sales order entry with 100% data accuracy. At the Celonis World Tour event in October, Celonis & Conexiom released a product offering to empower companies to mine, automate, and optimize their order management processes. Join this webinar to meet the technology behind achieving 100% data accuracy in sales order entry.  Register here!

I want to try root cause analysis in ML on my data. I have followed the same way as here https://celonis.github.io/pycelonis/notebooks/01_Root_Cause_Analysis_Tutorial.html. All steps work fine until it reachs  the row rca_result = RCA.apply()​The reported bug says this ： UnboundLocalError: local variable 'za' referenced before assignment​Details listed below:0it [00:00, ?it/s]---------------------------------------------------------------------------UnboundLocalError Traceback (most recent call last)/tmp/ipykernel_740/325769791.py in <module>----> 1 rca_result = RCA.apply()​~/.local/lib/python3.8/site-packages/pycelonis/root_cause_analysis/root_cause_analysis.py in apply(self) 60 selected_tables, search_list = self.filter_case_key() 61 chunks = self.create_chunk_list(selected_tables, search_list)---> 62 za = self.calculate_kpis(chunks) 63 zb = self.postprocess_kpi_frame(za) 64 return

Can we connect our locally installed SQL express under data model ? If yes, please help with detailed steps 

Is there a way to change the email or name in the training platform?

Hello,I created a action engine skill and its going very well. Since I am using approver name as a connection for the process automation to send email. I dont want that the approver will be bombarded with multiple email if he/she has many invoices that needs action. Is there a way to remove or filter approver with the same name only regardless of the other columns he/she has?Thank you.Example:

Hi there, I am getting a run time error creating a table and I am wondering if this concerns only the editor or also a scheduled transformation job?  Kind regards,Sasa  

How to get the connection details used by the extraction. We have the below scenario. There are numerous connections being established with different sources, and some are duplicated connections were created depending on the need of different credentials with different of permissions of a particular source. Now the permissions have been sorted by the DBA, need to clear out Data Connections which are no longer in use. Stating the IBC is being treated a PROD and any fail schedule will be treated as a downtime needed to identify the connections which are in used by the current extraction process.

Hi Team, I need to create some dashboard which shows the value added by Celonis in the organization based on Procure to Pay data. Do you have template/analysis which should help the organization to understand the value addition from Celonis. This will help to get the idea about what I can implement and show it to management. Thanks, Aakash

Hello Team, Im downloading the Process Flow from the Process Explorer as a PNG Output. Unfortunately the left most part of my PNG output is getting truncated. Are there any settings whereing I can specify the width, so that the PNG download is not getting truncated. Attached is the leftmost part of my process flow which shows the TP time as 81 Minutes. But in the PNG download, it shows only the Connection line with only minutes without 81. Kindly advice. Warm Regards, Jayanta Borkakoti

How could be a filter realised with the following condition: Activity_1 was performed by Division A and a following change Activity was performed by Division B We started with: FILTER (ACTIVITIES.ACTIVITY = Activity_1 AND ACTIVITIES.DIVISION = Division A) AND (ACTIVITIES.ACTIVITY LIKE %Change% AND ACTIVITIES.DIVISION = Division B); Which resulted in a empty OLAP-Table. Does anyone have an idea how to realise this filter?