Salesforce.com is now an increasingly used enterprise CRM system, which can improvise customer relationships and give plenty of analytical insights to decision-makers. There are many aspects of customer relationships and other crucial business functionalities being handled by Salesforce CRM for optimum results. Whatever the case is, the mightiness of any Salesforce system largely relies on the quality and integrity of data it holds. Data in the Salesforce system is often managed as a strategic asset. The data points must be clearly specified and always kept updated as the relevance of any piece of data, i.e., the significance of a “contacted lead” and a “stale opportunity” could be based on when it was entered and how much relevant it is it remains. The analytical reports always try to compare the performance of these leads over time, which may go meaningless without data integrity.
In a previous article in this Salesforce data backup and recovery series, we discussed an 11-point checklist to audit data and metadata on Salesforce systems. Here, in this article, we will discuss data in more depth and identify which systems are going to contain junk data instead of valuable information.
Quick metrics to evaluate Salesforce database
This article will discuss some of the quick metrics used to evaluate the SFDC database for data completeness, integrity, and quality. It is not just any one of these metrics, but each of these has its own contribution to data corruption. In this specific order, some rules may look at every SDFC instance. We will also discuss why these matter in the standards SDFC use cases.
- The inactive owners own not more than about 5% of the SFDC object records. In the Salesforce ecosystem, a record can ideally have only a true owner, and if this ownership is left with the wrong person, then the right ones who are supposed to cannot do the edit or view it. So, the records owned by inactive users may further ignite the data permission errors, which should be taken care of.
- About 90% of the contacts which are flagged as valid may have some information that should be validated. In many cases, it will be the phone number of the email, but you should also consider the calculation that those who have a snail mail of an FB ID.
- About 98% of those valid contacts may have filled in the accounting field. Without specifying an account, the system may categorize the contacts as private, and no one else than the creator can view it. This may cause some user-experience errors, which may be left unnoticed easily. You can create an account with the name by tagging it as “unknown’ and “do not delete” and tag private contacts with such an account name. Now when all the contacts are made visible, users can do a cleanup of data.
- Every object which declares the record type should have zero records without any record type. Such an event may occur while the record types are enabled with the deployed system. In this case, the newly added record type may be left blank. This will end up in some loopholes in the overall workflow and validation rules and codes. Unfortunately, system UI will not enforce the population of record types and edits. So, first, clean up all these by using relevant administrative tools.
- Ensure that there are no more than 10% of Cases and Opportunities are overdue. Ideally, this should be maintained as zero percent, but you cannot expect such perfection in a constantly changing data environment. This being said, any overdue records are important indicators of problems related to people and processes. So, you need to ensure that the data related to these areas are clean. You may also want to create standard status picklist values and set the overdue records to these values. Also, crosscheck whether all your business and sales support processes use these “overdue” data in a meaningful way, which may often happen.
- All these records may indicate that someone is not using these Opportunities correctly or they are closing the Cases permanently. Ideally, it should be less than 5% of Opportunities, and about 10% of the Cases are left “insta-close.” This may be mostly moved to the closed status without delay. Doing this well will also help during the time of Salesforce data recoveryif such a need arises.
- Ensure that the percentage of events, tasks, activities, which are open for more than a year or so is less than 5%. This may be a source of noise and only can help create confusion among the admins and users. In fact, there is nothing wrong with using these tasks to remind the users about the follow-up actions, but if there are tons of such elements which are long overdue, it may create chaos.
- Lesser than 10% of the close-won Opportunities have been associated with the Activity records. Having this at a higher level is an indication of info-hiding by the sales executives. This can be considered as an abuse of the system and should be addressed immediately.
- Also, less than 25% of opportunities are left without any updates. For this, you may create a timestamp that is the same as a modified or updated timestamp. This is similar to the info hiding observation we made above.
- Ensure at least 90% of the active users are logged in the near past, say over the last one month or two. Also, at least 75% of the users must have created or updated their records within a month or so.
- The next consideration can be made as not more than six users have the complete admin privilege in the Salesforce system. In fact, this can be a data issue that may cause system governance and metadata problems.
- The overall system should have a minimum of 10% spare capacity for further files and data. It may also become a challenge if you integrate any marketing automation systems as the event records and emails can be generated automatically and may prove out to be prodigious.
While considering these above measures, you need to be clear about your organization’s specific Salesforce use case and take the measures accordingly.