Product Q&A

I would like to add a column to case explorer from another table that is not the case nor the activity table. I am not using the cloud version of Celonis and I could not find a way to pull columns from other tables. There is a table that is linked to my case table, from where I want to pull a specific column which has comments. Is there a way to do this?

Hello friends!Will you be so kind to help with one issue:I have standard activity table: CASE_ID, ACTIVITY_NAME, EVENT_TIMEPlease find in attachment a sample of table.I want to make diagram chart (or OLAP Table) to show how many cases are being delivered at a certain month.For example -The activity Start of delivery for case #15 takes place on 05.01.2020 and the **next **activity End of delivery takes place on 10.06.2020. That mean that case #15 is being delivered on January, February, March, April, May, June.The activity Start of delivery for case #16 takes place on 07.01.2020 and the **next ** activity End of delivery takes place on 12.04.2020. That mean that case #16 is being delivered on January, February, March, April.So the diagram chart (or OLAP TABLE) should show followingMonth Case CountJan 2Feb 2Mar 2Apr 2May 1Jun 1Will you please clarify which formulas or functions I have to use.Thank you

HI, I found out there is a huge difference between using ISNULL(table.column) = 1 and different type of writing this formula (table.column) IS NULL. Do you have know from where this difference comes from ? Here is example from analysis: ISNULL(BKPF.KBR_KONT_FLAG) = 1 BKPF.KBR_KONT_FLAG IS NULL

Hello community, In my datamodel, the case key is BSEG.SID||BSEG.MANDT||BSEG.BUKRS||BSEG.BELNR||BSEG.GJAHR ||BSEG.BUZEI The LFB1 and the BSEG tables are linked (1:N) with the fields SID, MANDT, BUKRS and LIFNR. I have filtered my analysis as below: FILTER LFB1.SID =FC3; FILTER LFB1.MANDT =130; FILTER LFB1.BUKRS = CN01; FILTER LFB1.LIFNR = 0000115406; There are a total of 20 cases: COUNT_TABLE(BSEG) = 20. I am doing a PU_count of the BSEG cases to the LFB1. My understanding is that I should also get 20, but I get 29 ?!? PU_COUNT(LFB1, BSEG.SID||BSEG.MANDT||BSEG.BUKRS||BSEG.BELNR ||BSEG.GJAHR||BSEG.BUZEI) = 29 Do you know what could explain such discrepancy ? Thanks in advance for your help ! Clment

Hello,I am trying to export to excel data which contains a saved formula in some columns. I have downloaded Celonis APP for AP: Payment behaviour. I have noticed that when I change from saved formula KPI to formula directly, then there is no issue to export to excel. If I refer to saved formula in column value, then I get error Failed: Code 500.Is there a possiblity to export data to excel which includes saved formulas in column values?Thank you in advance for clarification

Hi all, I want to calculate the time difference between two activities in an OLAP table. Based on the amount of the time difference I want to use another unit. Is it possible to create a variable which gives you a result based on the row in the OLAP? For example: Event From | Event To | Time between Post Invoice | Pay Invoice | 30 Days Post Invoice| Cancel Inovice | 2 Hours This code in the variable will be only calculated once. Therefore the variable unit give me in the second row 2 Days instead of 2 Hours. CASE WHEN MEDIAN(DATEDIFF(ss, source(WFMT_EVENTS.TIMESTAMP), target (WFMT_EVENTS.TIMESTAMP))) < 120.0 THEN sec WHEN MEDIAN(DATEDIFF(ss, source(WFMT_EVENTS.TIMESTAMP), target (WFMT_EVENTS.TIMESTAMP))) < 7200 THEN min WHEN MEDIAN(DATEDIFF(ss, source(WFMT_EVENTS.TIMESTAMP), target (WFMT_EVENTS.TIMESTAMP))) < 86400 THEN h ELSE Days END

Hi Community. I am trying to put average of CS(Customer Satisfaction) on PE. The CS is based on Cases. I set Custom KPI like below. avg(case_tbl.cs) avg(source(act_tbl.cs)) They returned average based on activities. I thought PU functions would work and tried them, but they return the same value. avg(PU_MIN(case_tbl, act_tbl.cs)) avg(PU_MIN(case_tbl, source(act_tbl.cs))) avg1353839 98.7 KB Is there any way to calculate average based on cases in PE ? Thank you.

Hello everyone,I want to analyze how many vendors were just created for one order.What did work?I created an OLAP table with vendors and a simple case count.What is not working?I am not able to count the number of vendors with one case. So, I would like to have a KPI Number of One-Time-Vendors and it should show for example 150.Can anyone help me with this issue? It should be basically equivalent to an Excel countif.Thanks in advance.Best regards,David

Dear Colleagues,can you please let me know if it is possible to have the distinct string PU function applied in Celonis?In my specific case I need to know what currencies the vendors have used in the past. I can easily do the Count Distinct, which gives me the number of currencies, but I also wish to know what were these currencies. With PU_STRING_AGG I get the full list from all the documents, but I only need a string of distinct values - in my case USD and PEN. Can it be achieved?Of course I could create a separate Olap with currency as dimension, but the point is that this table needs to be used in Action Engine, so that is why I need it in single OLAP.Thank you in advance for any suggestions.image1560826 19.9 KB

Hi all, I want to get the sum of all processes waiting inside a specific activity over the time. I tried to solve this with Process Analytics as well as with Transformation Center but was not able to measure this. How can I do that? Kr, Mara

HI there,Can anybody help with the component filter to apply in order to focus the Process Explorer to include all pathways including the Change Price node.Many thanksChristian

Dear Colleagues,a simple question regarding function ADD_MONTHS. Is there a way to include in it leap year? I am calculating an expiry date for certain financial documents using formula:ADD_MONTHS ( ROUND_DAY(TO_TIMESTAMP(RBKP.BLDAT,YYYY.MM.DD)),3)So expiry is after 3 months of document date and if it is 29.11.2019, then the result should be 29.02.2020 instead of 28.02.2020. In excel we have function EDATE. Can we achieve something similar in Celonis?

Hello,Im trying to set up Action Engine skills using standard deviations from KPIs to trigger the firing. However, its not the Action Engine part Im having trouble with, its the standard deviation calculation.Essentially Im trying to calculate the standard deviation of client abcs cancellation % from all clients cancellation %.Ive experimented with the STDEV function but because the cancellation % function already has AVG (an aggregate) it doesnt work.Cancellation formula:AVG(CASE WHEN AuthMgmt_Tasks.auth_outcome IN (Cancelled after work,Cancelled prior to work)THEN 1.0 ELSE 0.0 END)Hoping someone has experimented with this before, appreciate any help!

Most frequently asked questions from Snap users: Q: Is Celonis interested in the data I upload? A: NO! Celonis has no interest in the data you upload. Learn more in our Trust Center. Q: What is the main difference between Celonis Snap and Celonis IBC Enterprise? A: These are the main differences: In Ceonis Snap you can upload ONLY up to 500MB.In Celonis Snap there is NO continuous integration with Event Collection as in IBC Enterprise.Features like Transformation Center, Workflows and Machine Learning Workbench are NOT enabled. Q: Can I use Celonis Snap for my own projects? A: Yes, as long as you dont exceed the limit of 500MB. Q: Can I invite other people to my team? A: Yes, you can invite your colleagues and work together in the same analysis. Q: Where can I learn more about pricing and editions? A: Click here to find out more.

Hey Celonis Community, If I want to show the average TPT between different steps in the Process Explorer the TPT shown should be around 1 to 30 days, instead it is now around Billions and Millions of days (see attachted file). (Same applies for the median TPT and the trimmed Median TPT as well as if I calculate the TPT by myself.) Is anybody else having this problem or can help me out? Thanks a lot! Jonas Process Explorer displays wrong TPT values945807 48.6 KB

Hi, The division of customers into different classes only works for one case. Only the customers from Europe. For the customers in America and Asia there is exactly only one class. Each customer is classified as a C customer. I have not changed the code. Only the sales organization has been changed. If I remove this filter and look at it on a global level, the division works into different classes but not within the intended limits (B starts at 90%). The abc function is configured as following: CASE WHEN ABC(SUM(VBAP.NETWR_CONVERTED), 0.6, 0.20) = 1 THEN A WHEN ABC(SUM(VBAP.NETWR_CONVERTED), 0.6, 0.20) = 2 THEN B ELSE C END

Hello, Is it possible to filter an OLAP table column for specific KPI values? Specifically, if we have an OLAP with some table attribute as the dimension and a KPI which calculates numerical values, is it possible to filter the KPI column for KPI(" ") > 0? Thanks!

Hi, I have a table with these two columns: Name and Total Net Value. The dimension is a simple column (NAME1) from the KNA1 table. The KPI Total Net Value is calculated as follows: PU_SUM(KNA1, PU_SUM(VBAK, VBAP.NETWR_CONVERTED) ) Now, i should get all customers with the total net value. This should be fine. Next step, I want to create a custom KPI which should give me a cumulative sum of total net values from all customers. So, every row should have the same cumulated number in the third column. I want to archieve this with the global function. But unfortunately, this function can not handle PU_X function as aggregation How can I solve my problem here? Thank you in advance!

Hello, I have loeaded my data model successfully but when i open my analysis there is no data (even the poucentage of cases). This error appears in all components : too many stages in execute_like Do you have any idea what causes this issue please an how can i fix it? PS: there is no filter applicable on the whole analysis and i kept the data model very simple. Thank you in advance

Running a standard transformation, Im encountering an occasional source data error (e.g. an alpha in an INTEGER column). Of course, the script halts immediately. The source environment is highly customized and there will be a few of these. DataException: [Vertica]VJDBC ERROR: Could not convert R03B802 from column table.numeric_field to a float8 This error message does not reveal a row number. Im loading over 20M records, so going back to source to locate and fix each data anomaly and then going through a long reload each time ugh. I cant get to the juicy part of this great tool because of the time spent returning to source data, one element at a time, cleansing and reloading. Im new to this, but it seems like Im missing something simple here. Is there a way to patch data post-extraction?

Hello all, I spent quite some time exploring all possible functions but Im unable to solve this.I need to do a running last based on the table as below. I will then use this to minus off the actual quantity. Ive explored the MOVING_X function, PU_X function and it doesnt seem to give me what I need. Any ideas? Appreciate the inputs!a869462 55.6 KB

Hello Community,I am trying to create a data range for my analysis.I have the below equation which works:FILTER CASES.NetDueDate BETWEEN ADD_DAYS(TODAY(),-6) AND TODAY() ;However, i would like to replace the 2nd TODAY with the last day of the prior month - 6.Can you please let me know if this can be achieved in Celonis?Many thanks,Anna

Hello Team, Can you please help how to move an analysis from one workspace to another with all the variables, filters and other stored details.

At this moment we have a filter on the extraction of the EKKO and EKPO which filter all orders and items based on the start date 31-07-2018.However, we want to add a second filter or expand the filter that fetches all orders starting with 37 which are in the EKKO and EKPO and started before before 31-07-2018.All other orders not starting with 37 must be retrieved via the previously set filter of start date 31-07-2018.The result must therefore be that we all pick up orders starting with 37 and the other orders only if they have been created as of 31-7-2018.How should we set up this filter?

Dear community, I would like to create an OLAP table same as cases to go in a process explorer. My data model is single eventlog which is consisted of process group, activity, and time stamp. If I could, Id like to show in the ratio as pie chart. Thank you for your support. Sugano