This past week has seen one of the UK’s largest (and certainly well-known) hosted telephony platforms experience a major services outage – leaving a huge number of businesses and sub-contractors without services for a significant length of time. As is the norm, many took to social media to air their ever-growing frustrations, but most, did that very British thing and said nothing and took it on the chin.
Anyway, what’s another service outage? They happen all the time these days.
In some sense, this seems par for the course, where customer expectations once again do not remotely align with the promises they believe they were given when they signed their service-provider contracts.
All of the above could be filed under ‘nothing new to see here’ if it wasn’t for the repeated comments by those affected about why their Call Divert capability/service hadn’t kicked in as they were told before they signed up.
But wait, wasn’t one of the big selling points of moving to cloud-based solutions their reliability? Leveraging those economies of scale. Where your cloud service provider had made all the big investments to ensure seamless fail-over of hardware and software systems, power supplies and other eye-watering overheads, so you didn’t have to? And you believed them!
The cynic in me would say that if this were analogised as a board game, there would appear to be far more snakes than ladders. Ironically, and sadly for the hosted telephony provider, this unfortunate outage happened just a day before the release of a new research report highlighting that businesses were four times more likely to leave a brand because of bad customer experience.
Don’t get me wrong, I’m not blaming the customer support team here. If anything, the support team were updating and responding very regularly to their clients. No, the customer experience error was committed way before the outage occurred – by promising something that couldn’t be delivered. That said, this of course could have been a totally unforeseen set of unique circumstances. Normally, where systems would be crashing about their ears, you could at least take some comfort in knowing that through the robust design and carefully curated resilient system architecture, you, the end user, would never know anything was amiss. As they say, Ignorance is bliss, just not this time.
To me, this smacks of buying into the sales pitch without digging into the contractual detail. Our long-standing and ever-patient Customer Services Director, Trevor Mockett spoke of these and many other matters back in May 2017 with two articles (here and here). Most notable of all his comments must be the following observation:
“Cold, hard proof.
Assessing these issues appears to require a leap of faith. You don’t know if the provider’s claims on any of these fronts are honest until you’ve actually experienced their service. By that time, of course, you’ve signed the contract and you’re past the point of no return.”
There are many that might see this outage example as yet another reason why the ‘cloud’ is still an immature choice for such a critical service as communications. An on-premises, internally managed legacy system is far more dependable, or at least a known quantity
But cloud-based telephony and unified communications are enjoying an enormous upswing in popularity these days. At a time when most business communication platform sales are either flat-lining or experiencing downward cycles, hosted comms is the only clear rising star of the industry. In comparison to its legacy alternatives, it promises low-cost, decentralised, omni-present, always up-to-date and reliable performance. And I would assert that the technology is not what’s at fault here. After all, your average hosted solution is really just a very big version of what was typically running on a customer site.
We inevitably end back at the ‘how you design, implement and maintain it’ part of the service provider discussion. Dare I mention that horrible cliché ‘failing to plan is planning to fail’!
I’ve written before about service reliability, particularly into the claims made by those stating 100% and 99.999% availability. As a quick reminder, the percentages below illustrate how much downtime gives you in relation to performance criteria:
99.9% = 31,557 seconds = 8 hours 45 minutes 57 seconds
99.99% = 3,154 seconds = 52 minutes 34 seconds
99.999% = 315 seconds = 5 minutes 15 seconds
99.9999% = 32 seconds
100% = have a guess?
These are very sobering numbers. Think about the last time you had a service outage. How long did it last? Does your current SLA match your current service experience?
Let me finish by re-iterating what ONI mean and make clear about delivering 100% uptime and Business Assured services. If you visit our website, it’s almost the first thing you’ll read:
“Our commitment to 100% availability underpins our data centre and managed service portfolio; providing our customers with peace of mind that they will always have access to their data and applications. It’s what we call Business Assured.
Since we first opened our data centre in 2011, we have never had a service affecting-outage. Ever. We put this down to our unique approach to data centre infrastructure management; optimising performance, power and cooling to ensure 100% availability.”
To know more about how we achieve and maintain this 100% uptime performance record or how we get businesses collaborating better, contact us at firstname.lastname@example.org