Pokemon Slow: How the Game’s Server Issues Could’ve Been Avoided

    July 27, 2016

    By Luis Colon

    The release of Pokémon GO has been plagued by server outages across the globe. What could have Nintendo and Niantic done differently?

    Since Pokémon GO has become available to consumers, the highly-anticipated mobile game has had a bit of a rocky beginning. The augmented-reality app has been marred by a number of server issues, including crashing, freezing, glitches, and loading problems.

    After the game launched for U.S. users, the ensuing frenzy of activity soon overloaded the servers of game developer Niantic and resulted in a medley of connectivity-related issues. As Patricia Hernandez of Kotaku notes, encountering such problems after the release of an exciting online game certainly isn’t out of the ordinary, but what makes this situation unique is that the issues have persisted for so long. Social media such as Twitter is still awash with frustrated users voicing their Pokémon GO server woes.

    The official launch of the game in Canada resulted in server crashes according to the Globe and Mail, and servers went down in Australia post-launch as well. And the game still isn’t available in Japan (where The Pokémon Company is headquartered) due to inadequate server capacity in the country, as Gamespot reports.

    Server Problems a Widespread Issue

    The Pokémon GO server debacle isn’t the only high-profile server crash in recent times. In October 2015 Southwest Airlines experienced outages and flight delays due to technical system problems, joining the list of airlines temporarily derailed that year by such difficulties. While the airline didn’t release specific details regarding the cause of the crash, a Southwest representative made it clear that it was not the result of a cyber attack.

    Capacity problems have also recently beset companies in the online retail, financial, and healthcare sectors. In September 2015, Amazon Web Services underwent outages that affected various websites and apps like Airbnb, Netflix, Tinder, and Amazon Instant Video. Target.com experienced outages on Cyber Monday due to a surge in shopping activity, according to TechCrunch — the online shopping holiday also forced PayPal offline, leaving 70% of users unable to complete online payment transactions.

    While Nintendo and Niantic aren’t exactly suffering on the revenue-front as a result of these technical problems, as AppInstitute explains (although some suggest that the protracted capacity issues might curtail interest in the game in the long run, according to CNET), other companies experience more demonstrable losses from similar outages.

    Southwest’s technical problems only lasted for two hours, yet still managed to generate delays for more than 800 scheduled flights, not to mention the added stress and inconvenience for passengers. Amazon’s systems were down for 6-8 hours, and with estimates putting the company’s net income losses at approximately $1,100 per second, the impact of such an outage is far from insignificant, as BuzzFeed reports — even for a company as large as Amazon.

    IT outages in the healthcare industry can have even more dire consequences than just lost revenues. BJC HealthCare, the largest hospital provider in the St. Louis area, experienced a 20-hour computer outage in July of last year. 13 hospitals were unable to access medical records, email, registration, and scheduling systems, making transferring patients extremely difficult.

    Preventing Crashes without Overspending on Infrastructure

    Given how damaging these high profile crashes can be, it’s surprising that more companies don’t engage in regular capacity planning. Gartner estimates that somewhere between 5-10% of large enterprises do it, though this number is expected to rise as the cost of digital performance problems becomes clearer and more damaging, and the expectations around cloud settle into reality. But the underlying reason for the lack of capacity planning may be more insidious; experts have suggested that there simply enough capacity planning experts out there to go around. Fortunately, new technologies are making this complex science more manageable.

    One company that successfully prevented outages before a big event is Verizon Wireless. Preparing for the launch of the iPhone on their network, the telecommunications company worried that its servers wouldn’t be able to support all the expected surge of orders placed for new phones (AT&T had suffered from massive service issues when they previously launched the iPhone). With the help of TeamQuest software, Verizon was able to effectively right-size their infrastructure to meet capacity needs for the launch, resulting in record sales with no notable performance issues. It was a huge win for Verizon.

    Every organization that is planning for a major launch, merger or multi-channel marketing campaign should be doing capacity planning. The opportunity costs of not so are very real and potentially disastrous. Not every outage is avoidable, but with the right tools, you can be prepared for server issues and equipped to resolve them before they become disasters if and when they occur.

    (Image credit: Pokemon/Pixabay)

    Category: capacity-planning