Best Practices for MuleSoft API Measurement
For many companies, even those with seemingly comprehensive and sophisticated application strategies, API measurement is an afterthought at best. Or worse, non-existent. Without data, it can be hard to prove that your investment in your MuleSoft implementation - and your continued re-investment - was worth it, even if you’re confident in the value that it provides.
As Gartner points out, though, API measurement isn’t a “nice to have” element of your API strategy but an imperative to integration strategy success. As Gartner’s research report, How to Use KPIs to Measure the Business Value of APIs, states,
“Not measuring the business value of APIs results in unproductive investments, missed opportunities and lack of governance. Application leaders responsible for API strategies must establish, measure and track key performance indicators to demonstrate and increase business value.”
Clearly, Gartner's point encompasses the business value of your API-led strategy and the APIs themselves. But API measurement isn’t just about the value the APIs bring - it’s a means of proving out the investment in MuleSoft itself. It offers a path to showing what MuleSoft does for the business and acts as a leverage point for expansion and further investment in the platform.
Measurement goes beyond just the business value, too. Metrics for your APIs and the associated KPIs are critical for operational success, adherence to process, security, and efficiency.
Gartner isn’t the only organization extolling the benefits of using KPIs for API and best practices. In a recent two-article series, Stephen Fishman, MuleSoft Global Practice Lead - Customer Success Strategy & Architecture, revealed MuleSoft’s perspective on data-driven organizations and the 11 best practices necessary to achieve that objective.
Big Compass certainly agrees with the sentiments of the series and especially appreciates two of Fishman’s noted best practices:
- #3 - Always remember that the purpose of a KPI is to aid a person/team in the act of making a decision, and
- #6 - Use KPI dashboards to teach decision making and systems thinking.
Of course, getting from little or no API measurement to a comprehensive and valuable set of metrics and KPIs can be challenging, even at the conceptual level. Following a set of Big Compass best practices can give you a roadmap from rudimentary application measurement to business value.
Best Practices for API Measurement
As with all best practices, these guide rails keep you from getting overwhelmed or off track as you develop your metrics and KPIs. It’s important to adapt these recommendations to your organization’s unique goals, operations, and strategies.
Invest time to learn the MuleSoft APIs and resources that can support your KPIs
MuleSoft has devoted time and resources to develop many APIs that can be used in your KPI program. The MuleSoft Developers Portal lists APIs that are readily available for inclusion into your measurement system. The use of these existing APIs can shorten your time to implement your first iteration of reporting.
An even faster implementation can be achieved by using a prebuilt dashboard like the one Big Compass offers. Information about this solution can be found on the Big Compass website.
Adopt KPIs that focus on business outcomes
Organizations choose MuleSoft and an API-led strategy for a lot of different reasons:
- Cost savings
- Accelerated innovation
- Faster development and responsiveness
- The ability to be more agile in the marketplace
- Closer alignment with the business
- Platform stability and reliability
The reasons you chose MuleSoft are unique to your business. Your KPIs need to reflect that focus to see where the platform and your strategy are succeeding and where it might be time to tweak (or overhaul) your existing applications.
It isn’t just these high-level goals that are important. Your APIs and related groups of applications serve the business, and your KPIs should reflect that. Focusing KPIs on business outcomes creates a common language between IT and the consumers of the APIs.
Start by identifying the business objectives for the APIs. Is a rapid response to the market at the top of the business’s list? Do you need to provide reliable APIs for your customers to use to interact with your business? Or do you need to measure revenue generated from these APIs?? You should have a set of KPIs that answer if the application is reaching the desired business objectives.
Once the KPIs for goal tracking and business outcomes have been established, add in those for ecosystem monitoring and API adoption. From there, align your metrics to those KPIs.
Some KPIs may require a combination or aggregation of metrics. For instance, if you have a KPI to reduce time to market for new application features, that may need to combine metrics like development velocity, API reuse, number of deployments, and number of incidents. You may be cranking out new code, but if it’s unstable or you have frequent outages, you may not be meeting the spirit of the KPI.
Group API Metrics and KPIs By Audience or Purpose
Grouping your KPIs by the audience that will be most interested in them can steer you toward which API metrics will be relevant for which group. Understanding this can also help you define and design reporting and dashboards for clarity.
Your API audience groupings may look like this:
- Business stakeholders
- Product managers
- Development teams / DevOps
- Infrastructure and operational groups
You may also want to consider grouping your KPIs and APIs by purpose. Here are some examples:
- Operational
- Adoption
- Revenue
- Business impact
Include API Measurement Design with Application Design and Enhancements
You should treat the implementation of your API measurement strategy the same as you would any other project or process.
Start by defining the process. Who should have input into the KPIs that are tracked? Who makes the final decision on the KPIs that will be tracked? Whose responsibility is it to define the metrics that will answer the questions that the KPIs work to answer? Who will implement the monitoring process? Who is responsible for reporting (dashboard or otherwise) and the cadence for periodic review?
Once you’ve defined the process and assigned responsibilities, document the system, and operationalize it. APIs should not be designed or developed if they won’t be part of the API Measurement system.
Baking metrics and measurement into the design and development process ensure that you aren’t adding unnecessary technical debt at the end of application development. Technical debt may be unavoidable, but that doesn’t mean it shouldn’t be managed. A well-heeled API metrics and measurement strategy will minimize the burden of technical debt, especially debt that gets added due to poor or uninformed decisions.
Leverage Your API Inventory to Define High Priority Applications for Measurement
It’s probable that you already have some APIs in place that are not monitored or not measured at all. Effectively implementing your API measurement strategy will require going back and defining the data and KPIs for those existing applications.
Start with your API inventory (if you don’t have one, now might be the time to create it). Using that list of existing APIs, involve the development teams, ops, security, and the business to define a prioritized list for the APIs that should be included in your API Measurement system.
You’ll probably need to balance competing priorities across these groups. As you set priorities, consider the API’s business value, usage, and visibility.
Use The Data Collected to Make Better Decisions
The information collected will help make decisions about the APIs you’re measuring. Your KPIs will help define what next steps might be needed to increase the value of an application, enhance it, or encourage reuse.
However, your measurement strategy should be a factor in making API-directed decisions going forward. The metrics you implement, and their story will make future development and API adoption smarter as you continue to leverage your MuleSoft platform. What’s working? What’s not? How do you avoid those pitfalls in the future? Where can you innovate in other areas to take advantage of the gains you’ve seen with existing implementations?
Using past performance as guidance for future development can keep new applications on track with organizational goals and accelerate the rate at which you achieve the benefits that an API-led strategy offers.
The data can also help in expanding or optimizing your MuleSoft footprint or adding in additional platform features. For instance, do you have the right number of vCores? What’s your utilization? Can you explain to leadership, in concrete terms, why you need more?
How about adding investing in connectors? Is there a case to be made for purchasing a certified connector? What do the metrics for current workarounds look like?
Conclusion
Best practices are a valuable set of guideposts to use when developing and expanding your API measurement strategy. But remember that they are just guides - you need to define a path that adapts these best practices to your particular needs and goals.
If you’re struggling with how best to get started with or advance your KPIs for your APIs, or if you just want expert guidance to create and implement a highly effective API measurement strategy, Big Compass is here to help. Our expertise in MuleSoft API measurement combined with our accelerated KPIs for APIs Dashboard implementation will have your organization measuring the value of your API strategy quickly.