The difference between monitoring the performance of your Apps and the performance of your APIs is greater than a lot of people care to admit. Application Performance Monitoring services out there do a great job at high level analytics about how your App is performing as a whole: monitoring key data points like Server Loads, User Adoption, Market Share, Downloads, Latency and Error logging. However, they can only monitor what your app and servers are doing at any given time. If users aren’t doing a specific thing, or if what they are doing is working but sending, for example, too much data for a mobile app to handle, then they won’t tell you that you have problems until your users notice.
For example, if you have a service handling expenses and something goes down, it will only tell you that the service is down when people try to use that particular service on a Friday afternoon, consequently, the lag between a service going down and your customers finding out could be significant.
Likewise, if something has gone wrong outside of your server, perhaps in the cloud, or in a load balancing service, then from the perspective of your app everything might look fine, but your end users will still be unable to use your service.
One of ruxit’s dashboards. Source: ruxit.
This is where API Performance Monitoring fits. With an API Monitoring tool, specific critical API calls can be set up to ensure the correct responses are occurring with the response times you expect. Core functionality of your apps, web services and back-end business operations can be configured to run globally on an ongoing basis, sending alerts when something unexpected occurs.
With the API Ecosystem growing fast and becoming a multi-billion dollar market, don’t forget about thoroughly monitoring the APIs that keep your business running.
Not being able to meet the SLA you sold a customer based on Availability or Latency could be the difference in keeping those customers you worked so hard to land and seeing them go to a competitor.