Small business services – lessons from the BA IT fiasco
You probably don’t have hundreds of thousands of customers, it’s unlikely your chief exec is grilled on the BBC and when your company gets it wrong Twitter might not creak under the weight of invective, but that doesn’t mean that a serious IT failure such as that suffered recently by British Airways can’t cost your company a fortune in money and reputation. Ensure your small business services are robust and well-protected and that you understand best practice. Here we detail the questions to ask your hosting company and the staff managing your IT services day-to-day.
Ask your DC about power failure procedures
BA has put out a barely credible reason for the outage: a power surge at a data centre followed by a failure of the back-up power system. Data centres are designed to withstand power surges and have strong back-up power systems. Problems can still occur, but it beggars belief that a company handling the volume of data, time-critical services and financial transactions that British Airways sees on a daily basis wouldn’t have cast iron measures in place to protect its power supply and back-up power generation. It’s astonishing to consider that BA would leave itself exposed in this way.
For the small company, the lesson here is to ask your hosting company about how it handles a power failure – the classic digger-through-a-vital-cable scenario. Ask it to explain to you – in language you understand – until you’re happy that short of asteroid strike, your services will stay up.
But let’s say that a power surge and failed back-up power service has indeed knocked your website, email and other utilities offline. Small business services are potentially more vulnerable to this than a huge global company as they won’t be sitting on the same dedicated services as a large company. British Airways was heavily criticised for not communicating with its customers; Alex Cruz, British Airways’ chief executive, explained this away by saying the messaging services were also affected by the outage. A company with the resources of BA has all its eggs in one basket? Staggering. A basic strategic error. Of course BA has data protection to consider when it contacts customers and has to use secure and encrypted channels for this, so I’m not convinced about the suggestion I heard to fire up a GMail account and email everyone. A security breach would have been the last thing BA needed on top of the outage, but again, a company with the infrastructure of BA surely could use services sitting in another data centre to communicate with customers. All data held should be backed up and held in multiple locations, so all customers should have been contactable.
Consider how you communicate with your customers: do you have contact data backed up? Do you have an alternative channel to communicate with your clients if your services go down? Back ups are very important but unfortunately their value is often underestimated until things go wrong and that data becomes critical.
Virtually every company uses some form of social media these days, and while it may not be the right channel for communicating sensitive information, make sure your clients know how to make contact with you in the event of a problem. Put your social media contact buttons or URLs in your email footer and on any paper invoices or other communications you may send out. Not knowing what’s going on ups the ante for your cllients very quickly and makes everything 10 times harder to deal with. If you can let your clients know that you’re aware of the probem and that you’ve set in train your recovery plan, you’ll make the aftermath less heated. Once everything is back to normal, get in touch with your clients to give them an update. Remember that other companies may view you as you view other small business services and expect a similiar response from you as you do from your hosting company.
Reliable IT staff
Whatever the truth of it, accusations have been aimed at British Airways that it made redundant its best (and therefore expensive) on-site IT staff and that the delay in restoring services was in part attributable to having to use remote contractors. Small business services are especially vulnerable in this situation as they rarely employ in-house IT staff and are entirely reliant on the procedures and expertise of the hosted platforms they sit on, so ask your hosting company what its emergency plan is. What’s your SLA – everything restored in four hours? Within 24 hours? What resources does it have access to? How much redundancy is in place? Redundancy is effectively ‘spare’ services and capacity that step in automatically to keep everything working when the main service is having problems. Ideally neither you nor your customers should notice that anything happened. Ask your hosting company too whether you’re charged separately for this type of support and what compensation is offered.
Doing the day job
BA is an airline: it flies passengers all over the world. This is and should be its priority, not dealing with IT problems and their very public fall-out. What would happen to your business if you had to spend three days dealing with IT problems instead of doing your real job? Would you lose money, miss opportunities, upset valued clients, have to cancel appointments, spend time and money reassuring and perhaps compensating people? These things have lasting consequences. To minimise the likelihood of a problem, the time it lasts and the fall-out you have to deal with, create a disaster recovery plan. We wrote an article about this a while ago and the advice stands. Sometimes things go wrong, but in the words of that article, plan to reduce it from a disaster to a nuisance.