How to validate the Good, the Bad and the Ugly of SuccessFactors Data
Imagine someone told you that some of the reports you get distributed for years are wrong. Getting anxious about explaining it to your manager? I guess you are afraid and getting paranoid it happens to even more reports. All the lost trust on your people analytics team and your data base. Gaining trust is the foundation of user adoption and business value of your data management program. How data works in SuccessFactors and recommendations how to validate your data and reports in this article.
Data is the most valuable asset we have – validate them!
Taking data quality seriously can be difficult in times of agility and speed-to-market. Losing trust in your own data can be a very painful experience and lessons learned are often costly. How do you prevent issues from occurring? After all, you may not have validation checks as part of your standard process.
Full data-quality frameworks can be time-consuming and costly to build and establish. The costs are lower if you institute your data quality steps upfront in your original design process. With SuccessFactors you at least do not need to worry how the data will flow into your database. Including all struggles around to build an efficient and trustworthy data warehouse. Some valuable exercise to review and overhaul your data quality is still important and gives you a better understanding of data in SuccessFactors.
SuccessFactors Data – the good, the bad and the ugly
As mentioned, customer do not need to worry of where and how data gets stored in SuccessFactors. With SaaS in the Cloud SAP guarantees a proper framework for your data. Therefore, cloud technology allows you to be on an innovative edge. Let us talk about the good, bad and ugly….
- All data are stored on SAP HANA 2.0 (don’t ask me for details if you don’t want to get fancy marketing buzz-words) which means data is stored efficient and enables you to access them very quickly.
- No ETL’s, SQL or programming needed. SuccessFactors provides an integrated “Data Factory” which allows to store and manipulate data. No time or invest needed.
- SAP SuccessFactors guarantees “quality” and consistency of the data base. In general, the database can be considered as very safe!
- Single point of truth in SuccessFactors. Most companies still struggle with the fact that data is stored in different place and maybe are not consistence or accurate anymore.
- In some modules (e.g. RCM, COMP) you need to define fields as reportable before you will able to use them. Yet, not all modules (LMS) will provide all data to report on.
- You will not be able to add additional information (e.g. tables) and for advanced analytics you often need Workforce Analytics (I would exclude all external tools like YouCalc Dashboard Builder, BIRT and SAP Analytics Cloud).
- You will not be able to correct historical data for non time effective data (missing for most of the talent modules). You have no possibility to correct/manipulate data in the row data base in general.
- There are no data dictionaries for most of the modules (just for LMS). It is difficult because implementation still can vary a lot.
You have some more points to add? Feel free to reach out to me or leave them in the comments below.
Why is hard to test / validate data
Validating data always sounds easier than it is. Most of the work are often hidden in details or complex report requirements. Here are couple of reasons why it is so hard to set of test criteria:
- Validating data and analytics is hard for developers, testers, and data scientists that are usually not the subject matter experts, especially on how dashboards and applications are used to develop insights or drive decision-making
- Data by itself is imperfect, with known and often unknown data-quality issues
- Trying to capture validation rules is not trivial because there are often common rules that apply to most of the data followed by rules for different types of outliers.
- Active data-driven organizations are loading new data sets and evolving data pipelines to improve analytics and decision-making
How to validate – what I recommend
Practical tips how to validate your data in SuccessFactors reports:
- Build Reports in your production instance. You are not implementing a new feature or change fields. You always create something new. Data quality and quantity (if you don’t have an awesome replication/refresh process established) is always better on production. It will help you building and secure a good report by knowing limits and being able to have better test cases. Also, you save time for export and import. Worst case is that things are different on the instances and it would affect your report.
- Create several test cases (minimum 3) to test single data points. For example, review single employees and timestamp, actions or values. The focus is one data set.
- Check reports against data source. A report often represents manipulated data through filters and different logics or calculation. Review your plain data and validate the data with Excel.
- Cross check aggregations. Controlling a single data point is awesome but don’t forget to validate mass data. Easiest way is a cross check with data aggregations. Ideally with a separate data source (report) to ensure you don’t proof your own mistake in creating a bad data source. Check out everything about the Report Center here.
- Leave enough time for post go live. Most reports for a running SuccessFactors module may not need it but especially the talent modules are a bit different. Especially Performance & Goals and Compensation running with forms and cycles. Often companies change something in forms or create a separate report for a new cycle. Leave enough time to see how data is flowing and what kind of scenarios can occur.
Some practical examples
We got all the theory above, but I also want to give you some practical examples where some of the recommendations will be useful.
Use Case: Migration of Employee Central Data. During your EC implementation you probably work with some test data. When it comes to the migration of your EC data it shouldn’t be enough to work with all the Excel files to validate your imported EC data. To validate for single data points (Point 2) you may just need to check some employee profiles. Using the standard export can be used to cross check against your import files. I would also use some standard EC Advanced Reports or quick reports to check aggregations of employees (Number of Employees active/inactive, in Division / department / Business Unit, Part Time/Full time and so on).
Use Case: Monitor Data and Data Quality. You run successfully one or more SuccessFactors modules. Time to check and monitor if your employees and key user use them as designed. In Employee Central you can check regularly if there are employees without a manager or monitor a headcount per Country, Division etc. In Recruiting you may want to check that every Candidate went through the Offer Step. There are many more examples….
Use Case: Check your Data. Before building a Report have a look at your data. I especially do not want to highlight data quality and quantity here. By browsing through Ad-Hoc report tables you may get to know the data structure and the one or other odd thing. For Example Recruiting: If a candidate skips some steps in the talent pipeline he will still have an entry in the Application History Table with the time in Status = 0. If you want to report an average time in a status for the entire talent pipeline we would get a bad result if we calculate: Average time in a status = SUM(Time in each Status) / Number of Status Candidate.
Subscribe, Share and stay informed – Thank you!
Please watch this blog closely and subscribe to get updates. We will continue working on a new article/interview with SAP to give you a clear understanding what is behind People Analytics, SAP Analytics Cloud in SuccessFactors and much more…
Please share this article to spread knowledge and give other SuccessFactors customers the chance to be fully aware of Reporting and Analytics Capabilities in SuccessFactors.
Never hesitate to comment, give feedback or ask: email@example.com.
Thanks for your time and interest!