Key performance indicators (KPIs) are measurable metrics that show how well you're doing in a particular area. If you want to know how much time is needed to recover after an incident, you can determine the average amount of time it takes to recover from the perspective of incident management.
Large-scale enterprise migration programs typically last several months, if not years, and involve entirely relocating from a data center or moving more than 100 servers. Due to this timeframe, it's crucial to establish KPIs early in the migration process to track migration progress and identify risks and issues.
There are four primary reasons why KPIs are valuable in large-scale AWS migrations:
A wide range of teams, such as security, infrastructure, operations, business, finance, and application teams, will be involved in various stages of large-scale cloud migration. Each team may have different reasons for moving to the cloud, but focusing on the desired business outcomes is critical. Sharing and discussing KPIs throughout the organization promotes alignment among teams towards the common objectives and enables tracking of their contributions. Taking this further, you could turn the KPIs into team goals to encourage the desired behaviors.
For instance, if your main objective is to lower your total cost of ownership (TCO), develop KPIs that relate expected application spending to actual spending.
Large-scale migrations necessitate a wide range of decisions. Many of these decisions will be made by application teams rather than the centralized program because they will better understand their application requirements. These decisions, however, can significantly impact the overall migration program and its benefits. Using the TCO example from earlier, application teams can choose target Amazon EC2 instance types (e.g., m5.large) or AWS Lambda function memory allocations (e.g., 6144 MB) for their applications. If the instance types or memory allocated to Lambda functions are significantly greater than required, AWS will charge you more. Defining and measuring KPIs ensures the program stays on track to meet the desired business outcomes. Furthermore, it aids in identifying when the program is trending negatively against a KPI so that corrective actions can be implemented following a thorough review.
Capturing and analyzing critical data points about a migration allows you to make data-driven decisions. Even if no data supports your decision, you can test a new approach and track its impact on your KPIs. If the new approach has a negative impact on your KPIs, you can quickly adjust your strategy based on the most recent data. It would take longer to notice the negative impact of the decision if you weren't actively monitoring the KPIs. Centralized migration tracking, including recording the dates assets were migrated, will be required. For example, your company may need to leave a data center by a specific date to avoid incurring large costs associated with renewing contractual agreements with a data center operator. You can forecast if you're on track to migrate the remainder of the estate by the deadline if you track migration velocity as a KPI (i.e., the number of servers or applications migrated per month). This will assist you in answering critical questions such as "How many servers are we migrating per month?" and "Are we on track to complete the migration within the required timeframe?" If you discover you are behind schedule, you can take corrective action and track the impact on migration velocity.
Once KPIs for the migration are defined, tracking and reporting program health can also become data-driven. For example, suppose you need to migrate 1,000 servers in 2 months. In that case, you can use data about your current migration velocity (number of servers migrated per month) to estimate whether you will likely meet the deadline. In the preceding example, the project must migrate at least 500 monthly servers to meet the required goal. Other program health metrics examples include:
Understanding your business-level KPIs requires working backward from your desired business outcomes. Because your migration may have more than one desired business outcome, you will likely need multiple KPIs. Socializing the KPIs to align your organization's leaders on the goals is recommended.
For example, if you're migrating to AWS to improve operational stability, you should identify specific KPIs that measure operational capability. This could include, among other things, comparing service availability to agreed-upon service level agreements (SLAs), the number of unplanned outages per quarter, and the mean time to resolve issues.
To measure program-level KPIs, you must first understand the constraints under which your program operates. A typical example is that data center consolidation or exit must occur before a specific date. As a result, you must migrate all assets before this date. Measuring achieved migration velocity against required velocity indicates your program's health.
Understanding how your KPIs will be measured once they have been established is critical. The goal is to create a fully automated machine that can collect the necessary data without human intervention. Furthermore, using automated data sources and creating dashboards increases the data source's integrity because it is not dependent on humans remembering to update the data.
For example, ingest and visualize the AWS Cost and Usage Report into Amazon Quicksight to create automated dashboards reporting cost-based KPIs. Alternatively, using the Cloud Migration Factory on AWS Solution, you could present your current migration velocity using the data captured.
Throughout this post, we discussed why KPIs are important, how to determine KPIs for your migration, how to measure KPIs, and outlined KPIs that are recommended for large-scale cloud migrations.
We recommend devoting time to determining the KPIs for your migration because it ensures that you measure what matters. Furthermore, once established and socialized, these KPIs will influence the behavior of team members assisting with the migration. Investigate methods to collect and present data automatically with little or no human intervention. This will reduce overall effort and eliminate the possibility of human error.
