In the early afternoon of September 3rd 2020, Movistar, a telecom operator in Spain had an outage, reasons to which are still unknown. Other telecoms and operators within the same parent organization (Telefonica) were also affected.
As a result of this outage, GitHub and reportedly other popular online services (such as Firebase, some Google services) were unavailable to some undetermined, but large portion of customers. According to some reports, the issue was limited to Barcelona and Madrid, while others say it was an outage that affected all of Telefonica (parent organization of Movistar) customers, peer ISPs etc. Details are sparse, unclear, and honestly not that relevant to this article.
Spain is not a small country by any measure - 46 million people live here. As you might imagine, quite a few of those are engineers, developers and others that use GitHub and other affected services on daily basis. Most of them use those services for work as well. Situation being as it is and huge amount of people working from home, the impact of this issue seems to be quite widespread.
Once the issue was noticed by those affected, in came the complaints to Movistar customer support Twitter handle. Again, no surprises here.
Handling of these support requests by Movistar customer support left a lot to be desired. See for yourself.
Note: Sans the first one, most of these are in Spanish or Catalan. But the content is pretty much the same.
As you can see from the screenshots, every single response that followed from Movistar customer support was a canned response by the support team asking for client details in private - fixed phone number (yes, we still have to have these with our installations, apparently), national identity number and contact phone number.
The last screenshot has an interesting “explanation” basically stipulating that for customer support to be able to (!!!) pass the query to technical department they need this information. Which would be fine if, you know, ones internet connection would have gone out. Not when a barrage of people are reporting a global outage.
Wait, but it gets better than that though.
It looks like in a Tweet that I failed to notice at the time and has been since deleted, Movistar support asked for someone involved to actually contact GitHub to report the issue to them.
How should this have been handled
The outage in question affected customers of another Spanish telecom - O2. Both Movistar and O2 belong to the same company group, parent of which is Telefonica, and both use the same infrastructure - thus the issue propagated to customers of both.
Naturally, being part of the same organization you would expect the response to be somewhat similar. Interestingly, nothing could have been further from truth than that.
Not just that, O2 customer support actually did followup once the issue was resolved. Something I see Movistar customer support never did - at least as far as I am able to tell.
Seems like Movistar support policy is seldom aware of the negative effects that poor service recovery can have on both customer satisfaction and retention. They seem to also be largely oblivious to what service recovery paradox is and how important it is to consider and have a proper crisis response policy, again, in terms of customer satisfaction.
Lesson learned? Personal approach to solving customer issues goes 1000 miles above a canned inpersonal, and inflexible response. Even if the personal response is vague and not really solving anything.