As of 9:00 a.m. on Thursday 30 July, the Pension Fund Administrators (AFPs) opened the process of applications for the withdrawal of 10% of their funds on their websites, however, they are registering high response times in the process, and others simply do not go beyond the home page.
Although today, Thursday, the official start of the application process, problems of availability of the sites have been occurring since the announcement of the 10% withdrawal, with a high number of concurrent users trying to consult their balances. There are more than 11 million people who can access the benefit in a period of one year, and due to the current contingency and the quarantine in several regions of Chile, users must make exclusive use of the AFP websites to carry out their procedures, collapsing the system. In addition, it was announced that this process can only be done remotely until 2 August, and the administrators' websites are already experiencing problems.
After a quick analysis carried out with the Atentus methodology and tools on each of the AFP sites, we have been able to detect high response times and unavailability to varying degrees in all of them, which can be explained by the high number of simultaneous users and websites that are not prepared for this.
Some AFPs are using the "virtual waiting line" as a containment strategy, but in the Internet world it is nothing more than a disguised unavailability, a way of saying "I can't help you now" or what is equivalent to having the branch closed. They provide the false assurance that the site is still available while users wait for the branch to be up and running again so that they can complete their transactions correctly.
Knowing that this would happen, the general manager of the AFP Association, Fernando Larraín, gave an interview with Cooperativa, anticipating that "the number of transactions and beneficiaries, which are more than 11 million people, may cause some failures during this period, but I can guarantee that we are in the best disposition".
What are the best practices for dealing with such events?
Knowing that several thousands (or even millions) of people would be accessing the websites of the different pension fund administrators, it is necessary to do work to avoid the unavailability of the sites and the high response times. work needs to be done to avoid site unavailability and high response times. This can be monitored in advance so as to know how their sites behave in the face of high concurrence of simultaneous users.
At Atentus we have specialised Stress Tests, which are in charge of stressing the sites to know how many users they can support before having failures, high response times or unavailability, not only at the level of direct load but also testing the processes or functionalities that will be used by them from the perspective of the user view and the application that supports it at the level of source code, databases, infrastructure and networks. Doing this would have helped to improve and prepare the platforms before they collapsed today. It is paramount monitoring of digital channels is paramount if we want to ensure a correct user experience and business continuity. In a process as important as the one we are going through as a country, the right decisions must be made with the people who depend on the different AFP service channels in mind, and these are not responding with the quality standards that users expect. This generates mistrust, uneasiness and anger on the part of their users. All of this is 100% avoidable with proper monitoring of their platforms.
We monitor the websites from the different internet providers in the country, receiving the information corresponding to the availability of their platforms and we alert in real time when they have an incident, thus being able to make improvements and fixes to the sites before users notice that something is not working normally.