Salesforce orgs are constantly changing, with new data being added by the second and new metadata customizations made by development teams every day.
While most businesses recognize the difference between the dynamic data management needed for Salesforce orgs compared to static systems, that same line of thinking hasn’t always caught up when it comes to backup.
How Do We Usually Think About Backup and Recovery?
When it comes to backup and recovery for any system, organizations’ concerns normally fall into two categories: “How much data have we lost?” and “How long will it take to get back to business as usual?”.
To answer these questions, the effectiveness of your backup and recovery can be benchmarked by two key metrics:
- Recovery Point Objective (RPO): This measures how much time has passed since you last ran a backup at the point you experienced data loss. So if you back up once a week, seven days would be the longest RPO you could expect from the point of data loss. Effectively, RPO measures how much data you can lose.
- Recovery Time Objective (RTO): This measures how long it takes for you to recover your data. It considers the time that passes before you notice you’ve lost data, the time it takes for you to plan out your recovery, and how long it takes you to implement those steps and recover your data. The shorter your RTO, the less downtime and costs you’ll incur.
In the event of data loss, RPO and RTO work together to measure the time passed and data inputted between your most recent backup and the restoration of data.
Improving RPO and RTO has been the best practice for businesses since the rise of business continuity planning in the 1990s. However, many organizations underestimate the need to protect their Salesforce production org and the heightened risks associated with data loss in a dynamic environment.
Why Is Backup and Restoration Different for Dynamic Systems?
Your Salesforce org is constantly changing. Consider how much new data – like new accounts, communications tracking, and purchasing information – is created in your CRM weekly, as well as how many customizations your development team is deploying. Losing a week of your business-critical Salesforce data has huge consequences. By contrast, losing a week’s worth of data for a static system may only translate to a small amount of data lost and a brief recovery period.
You also need to consider how the increased number of developers and end-users in your Salesforce org increases the chance of data loss. Salesforce orgs are the site of massive collaboration, with multiple users simultaneously interacting with the system, be it through deploying faulty metadata changes or end-user errors. Having more people interacting with your org increases the likelihood of accidental deletion or modification. Static systems, on the other hand, may only have a couple of active users which lowers the risk of data loss.
Finally, since Salesforce orgs are highly customizable, each business’s environment may be different. This means that a one-size-fits-all suggestion for backup and recovery, which may work for most teams using static systems, can’t be recommended for all organizations’ Salesforce orgs. You’ll need to create a more tailored backup and recovery approach if you want to succeed.
How Should You Recover Data in a Dynamic Environment?
Here are the best practices for restoring data in dynamic Salesforce environments to help you recover successfully when data issues occur.
Back Up Frequently
Salesforce orgs change rapidly, with new data being added and regular changes being made to existing data. This means you’ll need a far shorter RPO. Having a long RPO in fast-moving Salesforce orgs means that you still lose a considerable amount of data, even if you restore successfully from your latest backup run. Increasing the frequency of your CRM backups should be the first step to improving your data recovery.
Enable Automatic Data Notifications
How long it takes you to spot data loss is often a blind spot for organizations, and can often be responsible for a long RTO, with some teams taking months to realize there’s been a loss. To overcome this, you’ll want to implement a robust notification system that automatically provides real-time alerts for unusual activities or potential data threats.
Have a Disaster Recovery Plan
You don’t want to be left scrambling when disaster strikes. Deciding on a disaster recovery plan in advance means the key players are ready to jump into action if a data incident does occur. Once you’ve established a robust recovery plan, it’s important to practice the plan frequently and in a variety of common data loss scenarios. This can help you to decrease your recovery time, as everyone should be well-versed in their roles and responsibilities for recovery.
Restore Your Metadata First
Metadata gives “shape” to your environments and your data is somewhat meaningless without order and structure. By restoring metadata before data, you’ll make sure that your data is restored to the right place in its full context.
Restore Your Data in Sections
Once your metadata is restored, you should be ready to start restoring data. But you need to make sure you’re not losing useful data that’s been added to your org since the last backup run. This is why you shouldn’t just restore from a backup run in its entirety. Instead, work step by step to make sure you’re only making the right changes and not overwriting important data.
Summary
Data recovery for Salesforce orgs poses unique challenges and often has higher stakes. But teams shouldn’t be concerned that they’ll have to adopt an entirely novel process just for their Salesforce orgs. Instead, they’ll just need to raise the bar for backup and recovery when it comes to these dynamic environments.