The part of Friday the 13th’s Microsoft outage that particularly interests me isn’t that Microsoft has suffered a failure. Large, sophisticated technology systems occasionally fail.
It’s that some waste businesses appear to have discovered that their ability to receive customer orders depends on their ability to receive an email.
That should prompt some uncomfortable questions.
Most importantly, have we eliminated operational costs through technology, or simply exchanged them for dependencies we don’t properly understand?
Over the past decade or so, businesses have embraced digital technology with considerable enthusiasm, and understandably so. Online ordering, automated payments, electronic job sheets, cloud-based management systems and all manner of other innovations have allowed companies to handle more work with fewer people, reduce administration and make life considerably easier for their customers.
And I’m all in favour of that. There’s little sense in employing somebody to perform a repetitive administrative task when technology can do it faster, more accurately and at a fraction of the cost.
But I wonder whether, in our enthusiasm to automate everything, we’ve given enough thought to what happens when the technology stops working.
When efficiency creates vulnerability
Take a fairly typical skip hire company. A customer visits its website, chooses a skip, selects a delivery date and pays online. Nobody in the office needs to answer a telephone, take down an address or process a card payment. It’s a considerably more efficient way of doing business than the old-fashioned approach, and the customer probably prefers it too.
But what happens to that order once the customer has pressed the button?
Is it recorded in a system that the operations team can access independently of the company’s email service, or does somebody simply receive an email telling them that a customer has ordered a skip?
If it’s the latter, we’ve certainly made ordering easier for the customer, but we’ve also created a situation where a failure in an entirely separate communications service can prevent the business from knowing that an order has been placed. The customer might even have paid for something the company doesn’t know it needs to deliver.
And that strikes me as a rather extraordinary vulnerability to have introduced in the name of efficiency.
Of course, not every business affected by Friday’s outage will have suffered from that particular weakness. There are plenty of other ways that apparently independent systems can become dependent on the same technology provider. The important question isn’t which supplier failed, but why that failure was able to interrupt the business’s operations.
Many waste businesses now rely on cloud-based systems for practically everything, from allocating vehicles and issuing driver instructions to managing customer accounts, recording waste movements and raising invoices. The advantages are considerable, particularly for businesses operating across multiple sites, but so too is the potential disruption when one of those systems becomes unavailable.
And it’s not a problem confined to the waste industry. Any business that relies on technology to receive orders, organise work or communicate with customers faces much the same challenge.
What happens when the systems stop?
If the transport management system went down tomorrow morning, would the traffic office still know which vehicles were supposed to be going where? Could drivers continue working? Would customers still be able to arrange collections? Or would everybody be left waiting for a technology supplier to resolve a problem over which the business has absolutely no control?
I’m not suggesting that we should abandon digital technology and return to handwritten job sheets and filing cabinets. That would be ridiculous. Nor do I think the answer is necessarily to spend a fortune duplicating systems and employing people whose principal responsibility is to prepare for something that might only happen once every few years.
But there ought to be some consideration of how the business would continue functioning if its principal systems suddenly became unavailable.
It might be as simple as ensuring online orders are recorded somewhere other than an email inbox, maintaining an independently accessible copy of the following day’s transport schedule or having an alternative means of communicating with customers and drivers. None of these things is particularly complicated in principle, although they do require somebody to have thought about the problem before it happens.
And there’s another consideration that’s easily overlooked. It’s not enough to have a backup if nobody knows how to access it, or if accessing it requires the very system that’s stopped working.
An emergency procedure that depends on somebody remembering a password, finding an old spreadsheet or telephoning an employee who’s on holiday isn’t much of an emergency procedure.
I suspect a great many businesses have invested considerable time and money in automating their operations without ever properly considering how dependent they’ve become on the technology they’ve introduced.
We tend to evaluate these investments in terms of the savings they deliver. How many administrative hours can we eliminate? How many more transactions can we process? How much faster can we respond to customers? How much money can we save?
All perfectly reasonable questions, but perhaps we should be asking a few others.
What happens when the system fails? How quickly could we recover? And could we continue operating in the meantime?
Because there’s a difference between making a business more efficient and making it more resilient, and the two don’t necessarily go hand in hand.
Business continuity isn’t just an IT problem
There’s a tendency to regard these matters as the responsibility of whoever looks after the computers. If something goes wrong, we call the IT provider, explain that nothing is working and ask how long it will take to fix.
But the IT provider can only do so much.
They might be able to restore access to a system, recover information or establish an alternative connection. What they can’t necessarily tell you is which customer orders need dealing with first, which vehicles are already on the road, which collections are commercially critical or how the business should organise itself while the systems are unavailable.
Those are operational management decisions, and they need to be made by people who understand how the business actually works.
Indeed, I wonder whether some businesses have inadvertently become less resilient as they’ve become more efficient. In removing people and manual processes, they may also have removed the flexibility and practical knowledge that once allowed them to work around unexpected problems.
That’s not an argument for retaining unnecessary jobs or inefficient ways of working. It’s an argument for understanding what we’re removing, and making sure that the ability to keep the business functioning doesn’t disappear along with the cost.
There’s also a question of responsibility. If a company has become so dependent on a particular technology provider that it cannot function when that provider suffers an outage, is that entirely the provider’s fault?
Or should the business’s management have recognised the dependency and made some provision for it?
After all, you wouldn’t normally design a waste operation around a single vehicle with no alternative available, then express astonishment when the business couldn’t operate because the vehicle had broken down.
Why should we think differently about the systems that organise and control the work?
The eight-hour test
Here’s an exercise I’d encourage any business owner to try.
Imagine walking into your office tomorrow morning and telling everybody that the company’s principal computer systems won’t be available for the next eight hours.
Not that they’re running slowly, or that somebody is trying to fix them. They’re simply unavailable for the rest of the working day.
Could your business continue operating?
Could you still receive and record customer orders? Would you know what work had already been booked, which jobs were outstanding and where your vehicles and employees were supposed to be? Could customers contact you, could suppliers receive instructions and could you keep an accurate record of the work completed?
And when the systems eventually came back, would you be able to reconcile everything that had happened in the meantime without losing orders, duplicating transactions or creating an administrative nightmare?
I’m not suggesting that everything should continue at full efficiency. There might be some inconvenience, a few more telephone calls and a certain amount of improvisation.
But there’s a considerable difference between a business operating less efficiently for a day and one that’s effectively incapable of operating at all.
If you don’t know which category your business falls into, perhaps it’s worth finding out.
And if the answer is that operations would grind to a halt, I’d suggest that’s something the management team ought to be discussing rather than leaving it to the IT department.
Making the business better, not just more efficient
There’s a broader point here about how we approach business improvement.
We spend a great deal of time examining the cost of doing things, but perhaps not enough considering the consequences of the changes we introduce.
A new system might eliminate thousands of pounds in administrative costs, reduce errors and make life easier for customers. All good reasons to introduce it.
But if it also creates a dependency that can bring the entire operation to a standstill, that ought to be part of the investment decision too.
It doesn’t necessarily mean the investment was wrong. It means we need to understand the risks we’re introducing alongside the benefits we’re expecting.
Friday’s Microsoft outage will eventually become another one of those incidents that everybody complains about for a day or two before forgetting. The systems will recover, the emails will start arriving and normal service will resume.
But I’d be interested to know how many businesses will use the experience to question the way they’ve organised their operations, rather than simply being relieved that everything is working again.
Because the purpose of introducing technology is to make a business better at what it does, not merely cheaper to operate.
And a business that can’t function when somebody else’s computer stops working isn’t necessarily a better business, however efficient it might be when everything is running smoothly.
If your business depends on systems and processes that nobody has seriously questioned for years, perhaps it’s time to take a closer look. Improving a business isn’t always about doing more. Sometimes it’s about understanding what might stop you doing anything at all.
About Gerald Price
Gerald Price is a business change consultant and interim managing director specialising in business turnarounds, operational improvement and commercial performance, especially within the waste and recycling industry.
Having worked with operators across the sector, he regularly writes about leadership, strategy, waste policy and the commercial realities facing UK waste businesses.
More articles and insights can be found at www.gpcp.co.uk and https://www.linkedin.com/in/geraldprice/
