The way you can look at all the different elements of successful and failed requests by breaking them down and then discarding things over time, like just SSL time, time to first byte, or any number of those individual components, is extremely powerful. I was unable to find a competitor that was anywhere close to be able to slice and dice the data. I consider Catchpoint the best in breed for what we're using it for.
To get the best possible view into the performance of their systems, CardConnect set up Catchpoint on premises observers both inside and outside of their ﬁrewall. Those inside the ﬁrewall are set to test the performance of the CardConnect UI and the diﬀerent APIs on which it relies, while those outside are running on the Amazon Web Service (AWS) network in order to test the same experience that a customer would have when accessing the system.
The CardConnect operations team also runs regular tests of their DNS servers to ensure that they’re directing the users to the proper data center, and that they’re doing it in an acceptable amount of time.
Beyond just testing their system architecture, Catchpoint also allows CardConnect to utilize diﬀerent monitor types, and to slice and dice their performance data in many diﬀerent ways depending on the task at hand.
For server health, a TCP check tells them whether or not a server is up and reachable, whereas the FTP monitor is better suited for determining whether they are able to send and retrieve ﬁles in an eﬃcient manner. Meanwhile, the direct API monitors allow them to ensure that all of the applications within the UI are performing well.
All of these services are reachable quickly and easily within CardConnect’s custom dashboard. This dashboard allows them to move from test to test without wasting any time or setting up new requests. It also allows the operations team to compare ongoing performance data with historical trends, thereby determining right away whether there has been any change in the performance of any component.
Thanks to Catchpoint’s in-depth analysis and intuitive reporting features, CardConnect is able to save countless hours that would otherwise be spent researching data and running new tests whenever a problem arises. Instead, they’re able to rapidly drill down to the root cause of any issue and determine whether it’s a problem with their own servers or with a third-party provider. This is especially beneﬁcial when rolling out new features of the CardConnect service, as any problems that arise in either development or deployment can be detected before they becomes widespread, and often before any customers even notice.