Many people and companies position new products as "<old version>-killers". Sometimes it's true (or close to true); over time, touch tone phones replaced rotary dial, CDs have more or less replaced vinyl records, and DVDs have replaced VHS tape (and are in turn being nudged out by streaming video). But sometimes it's just change for change's sake, which doesn't always end well for the new offering.
Consider: New Coke was supposed to kill old Coke. Client-server was supposed to kill mainframes. BlackBerry 10 devices were supposed to make the older devices obsolete. The command prompt was supposed to die with Windows.
It is becoming increasingly evident that the old saw, ‘everything old is new again’, is alive and well. And that is, to me, rather funny. People have short, selective memories. Why else would we keep making the same mistakes over and over again, and hoping for a different result?
Of course, in many cases, companies repent of those mistakes, and then try to figure out how to put a good marketing spin on their backpedaling. It's all they can do, when the market tells them in no uncertain terms that they blew it with their new approach.
Think New Coke and Coke Classic. For those who don't remember the fiasco, in 1985 Coca Cola decided to redo the formulation of their beverage. Initial research showed that the new, sweeter version was accepted, and even enjoyed, in many areas, but that there was a core of drinkers who resisted the change. Coca-Cola released the new product with great fanfare nevertheless, and it was met in some regions with hostility. People either hated the product, or hated the idea that Coke was messing with their beverage of choice. Even bottlers began to rebel. Public pressure was so powerful that after a mere 77 days, the original formulation was reintroduced under the name Coke Classic.
So much for change for change's sake.
Microsoft ran into a similar problem with the command prompt, aka (to PC veterans) the DOS prompt. From the day it introduced Windows, it did its best to wean users away from typing commands, trying to get them to use point and click graphical user interfaces (GUIs) for everything. Sometimes it took longer to perform a task using the GUI, and was more difficult, but it was ‘modern’.
Users yawned, and continued to type. Finally, at the Windows XP launch, the CEO ceremonially ‘killed’ the command prompt, much as Dave killed HAL in the movie 2001: A Space Odyssey.
Users didn't fall for it though. Through XP and Vista, and especially in various server versions, the command prompt remained quietly in the background, as users and administrators insisted on typing commands when it was more convenient. Finally, Microsoft got the hint, and in Windows Server 2008, it even included a minimalist Server Core edition that (surprise) dispensed with the GUI entirely, in favour of the command prompt.
One could argue that the company did the same thing with Windows 8. The Metro, aka Modern, UI had more detractors than supporters, so in subsequent iterations, Microsoft has been backing away from the features that annoyed users the most. Windows 10's technical preview is becoming more and more familiar to users of the much-loved Windows 7. That's deliberate: CEO Satya Nadella said during the company's latest Windows 10 update that Microsoft's goal is to "make Windows 10 the most loved version of Windows," and that means removing the bits that irritate customers and adding goodies that make them smile.
On the hardware side, we have a couple of tales: one cautionary, and one unexpectedly positive.
First, a painful tale of what happens when you mess with a successful product without providing sufficient new advantage to customers. BlackBerry's abortive attempt to become trendy cost it, bigtime. With the BlackBerry Classic, it has returned sufficiently close to the BlackBerry its core corporate users know and love, and with the Passport, it has returned enough of the critical features to its innovative design that it is appealing to diehard fans. Time will tell if it's enough.
IBM, on the other hand, decided not to mess with the underlying concept of one of its major successes. Those who remain firmly convinced that mainframes are dead need only look at recent news to see that apparently the company didn't get the memo. It just released a new mainframe.
Yes, despite the fact that the company has sold off its x86 server business, it has clung to its supposedly obsolete mainframes. Analyst A.M. Sacconaghi of Bernstein Research estimates that those dinosaurs, along with their software, services and storage, generate 25 percent of IBM's revenues, and 35 percent of its operating profit. That's not bad for something that was considered ancient by most people 30 years ago.
Ironically, the old mainframe concept is just the thing for many new use cases. The newly-announced z13 was designed to handle the huge volume of transactions generated by our increasingly mobile devices; one system can run up to 8,000 virtual machines. IBM spent $1 billion on its development, and generated 500 new patents, many in the security realm. It appears to be paying off – the New York Times quoted one early customer who said that it expects to cut its total cost of ownership by 50 percent by replacing its backend server farm with a single z13 mainframe.
With IBM, the key advantage was that it hung on to the mainframe concept, but updated the underlying technology (real time encryption, embedded analytics, faster processing, more memory, etc.). It didn't, as the saying goes, throw the baby out with the bathwater.
That's an example other companies would do well to learn from. It's possible – and very lucrative - to innovate without totally alienating loyal customers. New doesn't need to be killer of the old.