Data Quality is the most important aspect for any business to reap the right benefits as it is a layer for the analysis that help business make strategies.
Poor Data Quality means, you are simply running in the wrong lane. Well talking about People Data or called Human Resources data, it aint it a very good condition either. This is serious, as this data means about you, yes YOU.
This also intrigues a thought is business data more important than people data. Well, we might cover this in another article.
Unlike financial data which is supposed to be accurate and secured, HR data has its own privacy and sensitivities. Poor HR data may results in looking at a good picture in a bad way, what I mean here is, if Rewards and Recognition data is not captured or stored accurately, we will not be able to derive actionable insights.
Let’s say you want to find out if R is a driver for employee retention. Hence with incorrect people data, you might not be able to get right context.
Let’s find out the reasons which impact HR data quality:
Well, having said all of this the most important one is to find out how is the people data getting generated? Then where is it getting stored, post that how is it getting stored?
To start this, we will take recruitment process as an example.
The table below is called as SIPOC (Supplier Input Process Output Customer) which is taken from the area Lean Six Sigma.
· External Search
· Internal Hiring Team
· Employee Referrals
· Talent Mobility
· Social Media
· Performance Reviews
· Requisition Management
· Sourcing Response Management
· Screening & Assessment
· Selection and On-boarding
· Selected Candidate
· Requisition closure
· Quality of Hire
· Hiring Manager
· Business Unit Head
· HR Employees
· Hiring Team
If we understand this table, it shows the full recruitment process in a nutshell. From our perspective of hr data, it is prepared and generated for the customers to understand and take action from.
Now if we look at the input, there can be various issues. For example the organisation may be using a portal where candidate would go ahead and input his information to be eligible to the interview. The hr data quality issues start from here. Questions like how robust is the hr data capturing methods in the portal must be answered, such as a field like education, anyone can write anything in any format if it a free text. Why I am talking about this is, that we must make provisions so that we control the correct inflow of hr data in the system.
Is the people data stored in silos: Out here you need to be able to answer the top 2 questions:
a. Where is the people data stored?
b. How is the people data stored?
If the above 2 are positive, then you must check are the data definitions standardized?
This leads to the following important aspects:
Where are the definitions stored?
How old are they?
When were they last revised?
What percentage of the HR partners understand the data definitions?
Are they using the right templates or framework?
Now let’s take a very interesting example: We will tie this up with a human capital question.
Are our high performers leaving?
Considering this question is raised by the HR Head. We might rush to find who is leaving, tie up leavers data with performance ratings, look at what is the average tenure of the high performers, what are the top reasons of turnover, what is the proportion of voluntary and involuntary turnover of employee.
If you look all the above points I mentioned are data points or metrics. Hence it is very important to find where and how is this data being currently captured.
To improve the hr data quality, we must look at the 4M approach.
1. Assess Data
2. Define the baseline
3. Define metrics and targets
4. Define and build data quality rules
5. Enforce data quality rules
6. Monitor data quality against the targets
7. Review exception and gaps
8. Capture the errors
9. Refine the data quality rules
10.Automate hr data quality process