The latest trend (or over-hyped term, if you like) is "consumerisation of IT." As with cloud computing, the term is somewhat ambiguous and is applied to a number of things that are recognisably related, but which differ in details.
Consumerization of IT is usually contrasted with "enterprise IT, " which carries connotations of interminable rollouts, bewildering interfaces, obscure functionality and high prices. The poster child ordinarily cited for "enterprise IT" is SAP, which seems to raise particular ire in commentators.
Consumerisation of IT, on the other hand, is associated with ease-of-use, attractive interfaces, intuitive functionality and low prices. Apple is ordinarily referenced as the exemplar of this type of computing. (Apple may not be known for low prices, but you get my point.)
Admittedly, Apple may not be known for low prices, but consumerised IT as delivered by the company can be delightful. I received a Google phone when attending Google I/O a couple of years ago. To get it configured and connected to Google's own email service required me to configure ports and various settings, none of which I knew off the top of my head. Between looking up the information and configuring the phone, the process required 20 minutes, most of which was consumed with me inputting stuff that I had no idea was correct. The iPhone, by contrast, required my email address and password. Within 10 seconds I was accessing my email.
Why was one easier and the other harder? Apple achieved its superior ease of use by designing its configuration to assume standard defaults and automate the configuration process. Only if the default configuration fails would the user then be forced to drill down into configuration options. After all, it's fairly uncommon that someone uses unusual ports for email access (although it can be done), so why not implement a configuration flow that assumes the typical mode as default and allows customisation if necessary, instead of requiring everyone to configure their system as though custom?
Certainly, there's no question that apps designed with ease-of-use as a primary objective are much simpler and more satisfying to use. I will note, however, that Apple is not perfect in this regard, despite what its multitude of enthusiasts believe. The latest version of iOS has discarded iOS's annoying screen-based notification system in favor of Android's superior pull-down notification mechanism.
In the corporate IT world, this move to "consumerised" IT has been described as the penetration of employee-purchased mobile devices like the iPhone, iPad, and Android phones and tablets. This phenomenon is going to swell to greater and greater dimensions. If you read my prediction blog last week, in which I forecast the enormous, gigantic growth of special purpose devices, you know that I think corporate IT will face that growth from here on out. IT organisations are going to face more and more pressure to support the BYOD (bring your own device) world.
However, challenging as this is--and it's plenty challenging, make no mistake--viewing the consumerisation of IT as an extension of employee use of internal IT applications seriously oversimplies what the trend really represents. Consumerisation of IT isn't about employees using consumer devices; it's about consumers becoming the primary users of internal IT applications. Dealing with this challenge will be significantly more difficult than supporting an employee BYOD environment.
Simply put, we are reorganising the way companies do business. We have moved past the analog age and are truly in the digital age. What does that mean? Consider how business was done pre-digital. A lot of interaction happened on an analog basis: sales meetings, mailing a brochure, fielding a telephone call. Lots and lots of human and physical interaction. Finally (one hoped), it resulted in a digital business transaction. A sale. A re-order. A claim. At which point, someone from your company interacted with your internal IT systems to enter the transaction. Subsequent handling of the transaction was more analog. A phone inquiry to check on shipment status. A query regarding progress of a claim. Easily over 90 percent of the value chain was non-digital.
Today, companies are moving to all-digital, self-service customer interaction. Learn about the company's products (i.e., be marketed to) by browsing the product Website. Place an order via the company's mobile phone app. Check the status of a transaction by logging in or using the automated phone tree. Companies are getting out of the human interaction business as fast as they can, which means that the volume of digital interaction--the stuff of IT--is skyrocketing. Today, a company's internal user base is far outnumbered by external users, who may be customers, partners or suppliers. This is the consumerization of IT, and it poses perhaps the greatest challenge to IT organisations ever.
Let me explain.
First, if you think your users had a variety of devices, wait until you see the rest of the world. Every device under the sun is in the hands of users, and more pop up every day. The notion of an "approved device list" is laughable. Instead of accepting a few and excluding the rest, IT has to develop a strategy that makes it possible to accept everything from everyone. This requires making it easy (and possible) to integrate an app on a device into corporate IT systems.
And by the way, the driver for this device explosion isn't frivolous users with time to waste on their smartphones. It's your business delivering its own apps to give customers a way to interact with the company on their terms, at their convenience. It's your business making it easy for business partnerships to spring to life by saying, "We can work together; it's easy. Just integrate your app with ours via this mechanism."
Solution: You need well-defined, convenient integration points that do not require a project team to get the handshaking complete. The buzzword today is APIs, but this need is the fulfillment of the SOA movement of a decade ago. The difference between now and then is that before it was a "should," i.e., "We should have a well-defined integration layer with standardized interfaces because that's the most elegant way to do it." Today, it's a "must," i.e., "To support ongoing partnerships and manage the plethora of devices that accompany them, we must have a convenient integration point that does not require custom engineering on our part or the partner's part."
If you're in IT, you're now a software supplier, and you need to offer and document straightforward interfaces with which partners can integrate without joint engineering projects.
Second, application load will be much more variable. Corporate IT was usually quite predictable. A known user population. Common use patterns: Log on at 8:00 AM, big surge at 8:30, slowdown around lunch, higher volumes through mid-afternoon, tailing off in the late afternoon. When mobile apps are in the hands of external parties, it's hard to know when they'll be moved to interact with you. It could be upon waking at 6:00 AM, during the day, as their schedules permit, mid-evening after dinner, or even in the middle of the night because progress on the order is keeping the person awake. Applications don't have downtime in this world, only varying levels of use around the clock.
Solution: It's not enough to implement virtualisation. You need a solution that is elastic, that can expand and shrink according to load. This is where cloud computing and mobile come together nicely. Just be sure your cloud is really elastic and not just warmed-over virtualisation with limited flexibility of application topology. When a highly variable load meets a highly static application infrastructure, bad things happen, career-wise.
Third, application load will be much higher. Not only does the availability of mobile make your user population's use profiles more variable, you're now subject to their use profiles, and that can drive enormous traffic to your systems.
Let me offer an example. One large US retailer is finding new life as a fulfillment backend for other companies' Websites. This is a great way for the company, which has struggled to remain relevant in today's retail environment, to obtain business from other retailers and even businesses that don't appear to be retailers (think domain enthusiast sites) who have close relationships with an entirely different consumer base. The US retailer provides an easy way for other retailers to drive transactions for its goods.
However, it also means this retailer is subject to the vagaries of those companies' promotions and user bases. It only takes one mention of that partner company's mobile app on Oprah (or America's Got Talent or some other mass media show--remember, this is all about consumerisation) to potentially drive huge volume and transactions back to the retailer. And, by the way, the retailer might not have any idea that the partner was going to drive this much traffic. In fact, the partner might not know it's going to drive that much traffic. That's what consumerisation of IT means.
Solution: You definitely need elasticity to be able to respond to high volumes. But you also need to ensure that your elasticity works end-to-end. If your database can handle tens of thousands of transactions per minute, but your load balancer can't support more than one tenth of the received traffic, you've got a bottleneck. The only way to expose and fix bottlenecks like these--before the crush of traffic takes down your system--is to load test it, fixing each problem as it arises during testing. Again, this is where cloud computing comes in. It's this kind of situation where "the illusion of infinite capacity" is important.
If you take one thing away from this post, it's that the consumerisation of IT raises the importance of load testing apps.
Fourth, your systems need to be easy to use and the functionality has to work, always. In the old days of enterprise apps, if users were unhappy, so what? They were employees. Even if they griped, nobody resigned because the application was clunky. And even if they did, there was always someone else who would take their place. Today, your users are fickle. They'll try an app, and if they can't get it to work right away, they'll go on to another. It's a "live by the sword, die by the sword" scenario. If people can't do what they want with two or three clicks, you're going to end up with a gaping wound in your company's forecast.
Solution: Learn from the best out there. Examine the most popular business mobile applications. I use The Weather Channel's and FlightTrack Pro's apps all the time. They're elegant, single-purpose applications that I rely on during my travels. Find your own application exemplars. Learn from them. Your applications should deliver value on the opening page--without someone having to do more than enter some simple information (like the iPhone email configuration example cited earlier). Your applications should make even more valuable information available via progressively deeper interaction. Find people who have never used your application before and see how they get on. If you make it easy for first-timers to get going, you'll be on the right track.
In summary, the consumerisation of IT is far more profound than slapping a pretty interface on a decade-old enterprise application. That's just lipstick on a pig. It's even more than adopting an easy-to-use SaaS application. It's all about recognising that the boundary between your company and the rest of the world is getting blurry--and that's a good thing. Letting end users engage with your systems can transform your business relationships and your economics. Just be sure you're ready for the real consumerisation of IT.
Bernard Golden is CEO of consulting firm HyperStratus, which specialises in virtualisation, cloud computing and related issues. He is also the author of "Virtualisation for Dummies," the best-selling book on virtualisation to date.
Join the CIO Australia group on LinkedIn. The group is open to CIOs, IT Directors, COOs, CTOs and senior IT managers.