By: Nick Kurbatov, Digital Business Analyst
Data Quality – what do you think about when somebody mentions these words?
Have you ever imagined what would happen if your address book or files on your computer suddenly go mad, because there are no unique identifiers(IDs) in your data?
It would be a mess. You wouldn’t be able to easily find your colleague’s date of birth or the name of your favorite restaurant unless you have IDs for them (that is applicable for large amount of data of course).
In a previous article we started a topic about data that has been entered by sales people into Salesforce.com via a mobile device, but what exactly did they enter? Did they follow the rules of data quality or did they just enter the data the way they like? This is the topic that I’ll describe here.
I’ll share my experience in running a Data Quality project based on Salesforce.com CRM.
First of all let’s cover the rules of entering data into the system.
Next, I will describe how I set up these rules in the organization. After that I ran a simple report showing the number of opportunities per account. I found that a large number of opportunities are lacking the information regarding the number of order attached from ERP system, which is one of the main proof that a salesperson made a successful deal with a customer.
The salesperson entered only the amounts of opportunities (which was calculated from SUM of product lines). That information is looking good in terms of the pipeline and presented as a big number in dashboards, but is it reliable?
No, it isn’t unless the SUM of opportunities is equal to the SUM of orders.
The solution was the following: together with the stakeholders we decided to link orders from the ERP system to Salesforce.com and set the KPIs that will be directly connected to the SUM of orders which are linked to opportunities.
Another big Data Quality issue I identified was regarding the information that’s been entered in an account object. The problem was that salespeople were filling only certain fields and in the manner they prefer (e.g.: there are 5 different variations of an account name). Some of the most important fields such as website were omitted as they wasn’t mandatory.
The solution for this issue: I set up a strict account object naming convention (later we applied this to opportunities and contacts as well), and set up mandatory fields for information critical to business (the business point of view of course) in order to see the major information about the account in one single page.
To track all of these changes I created a Data Quality dashboard and I run it on weekly basis to track the data health. In a period of 6 months after I launched this Data Quality Project, the total amount of duplicated data was reduced from 30% to 10%. I consider it a success.
You can’t regulate everything as well as set up rules for each activity of a salesperson as it will immediately influence employee satisfaction from performing activities in CRM or any other system. You always need to have ~10% of freedom in choosing the way you want to work with systems.
To conclude, I would like to say that every rule you apply regarding the usage of a CRM system (or any other) should be dictated first of all from the business perspective – not IT or any other, because the first and foremost reason of existence of CRM systems (or any other) – is to help companies make more money, to be more efficient, to be more client-focused, and to predict customers’ behavior. Without answering “yes” to at least one of those, it would be useless to implement any Data Quality rules (or any other) as there would be no business sense behind it.
In this article we covered entering data and data quality. In the next article I will cover Data Security as this is one of the most current trends in our digital world.