The last two weeks I've talked about the battle between IBM and Amazon Web Services and how I believe IBM will beat AWS. The previous two pieces discuss turf and positioning, and how companies that try to move from where Amazon is to where IBM is either stall or go out of business.Why can't user-focused companies do enterprise business, or enterprise companies do consumer? Apple's servers failed. IBM had to sell its unprofitable PC business, which is now the world's top PC maker. Cisco Systems' consumer business (which it purchased) nearly took the company out. EMC sold its consumer business. Sony's attempts at enterprise sales failed (along with a lot of other stuff). Netscape imploded when it tried to expand to the enterprise.I've spent the most time watching Apple and Microsoft over the last decade. I was front and center for the Apple server and Microsoft Zune launches. I told both companies those products were dead at birth. Turns out I was right. I actually told the Zune guy he'd lose his job, and he did.Let's talk about why companies can't seem to do both consumer and enterprise technology at once \u2014 and how you'd build a company that could do it. It is possible, but it takes a unique set of skills, and conviction, to pull it off.Enterprise Tech Firms Making Consumer Goods Rarely Succeed \u2026 One of the earliest stories I heard about Bill Gates and Steve Ballmer, which folks inside both IBM and Microsoft at the time can confirm, recalls a trip to Armonk where they saw a bunch of employees working on IBM PCs that all ran 3270 emulators. There were expensive, slow-to-boot terminals and weren't really PCs at all. Young Gates and Ballmer realized that IBM just didn't get the whole PC revolution thing. (I can imagine Larry Page and Sergey Brin visiting Redmond in the mid-2000s and saying the same thing about Microsoft and mobile, though I doubt this ever happened.)Nonetheless: Microsoft to Make Leaps in Mobile EnterpriseThe irony is that both Microsoft and IBM achieved success by emulating Apple, which was, and remains, the most user-focused of the group. It's ironic because, when you focus on empowering the user, IT is an obstacle to overcome. Apple has had more success going around IT then it ever did trying to sell Lisa to IT departments.In those examples, IBM and Microsoft lost track of PC and mobile users, respectively. Rather than helping users overcome IT, they switched to serving IT. From the user perspective, this would be like Luke Skywalker suddenly realizing he wants to be Darth Vader \u2014 which is why the new Star Wars movies were so bad).This transition is painful, and often deadly, because trying to be IT's buddy after being the bane of its existence for so long doesn't go so well initially. IT sees you as you were, while the user customers who made you successful suddenly feel betrayed. You also have to learn compliance, licensing, discounting and how to protect folks who make stupid decisions (that's politics). So you bleed customers at one end and lack the skills for the new ones you're trying to win at the other end \u2014 and they really don't trust you.Now think of Amazon Web Services. How often does it (and other firms like it) come up in IT meetings as the problem to overcome? And isn't one of the key benefits of AWS that users can go around IT to use it? \nHow-to: 6 Tips to Help CIOs Manage Shadow IT\nMore: Shadow IT: Once Hidden, Often Hated, But Well Worth EmbracingOnce we label a firm as a problem, it takes a long time to shed that label. Microsoft has had a long string of consumer product failures, starting with Zune. Like IBM before it, Microsoft clearly didn't get users anymore. What was the Zune's killer feature? Secure file sharing between two Zune owners who both had Zune music subscriptions. At its core: Compliance, an IT concept. At the iPod's core: The ability to get music regardless of its source, even if it was stolen.Essentially, Apple was the Rebel Alliance; Microsoft was the Empire. You can't serve both masters effectively. If you try, as Anakin Skywalker did, you're simply screwed. \u2026 But Xbox Shows That It's Not ImpossibleMicrosoft did succeed, though, when its Xbox took on the Sony PlayStation. Granted, Sony's mistake was building a game system that was priced out of the market, but the Xbox still gained a following and eventually led the market. Xbox got there because Microsoft launched a well-funded, arm's-length effort to go after gaming, and the folks Microsoft picked to do it were allowed to execute as if they worked for a separate company.News: Xbox One Will have a High-performance Custom ChipSo a firm can do both enterprise and consumer tech. But you have to firewall off the fledgling effort \u2014 and, I'd argue, operate it under its own brand so there's no confusion about what the parent brand stands for. Car companies do this more than most; their professional brands carry professional messaging, while their consumer brands are about design, safety and status.Here's something to noodle on over the weekend: When you look at the new Xbox One, which was created after the group was re-integrated with the rest of Microsoft, do you see the Rebel group that created the Xbox, all fired up and rabidly gamer-focused, or do you see the Empire and the Zune, all about subscriptions and assuring content revenues? Remember, Xbox One is $100 more than PlayStation 4.AWS, IBM Should Target Each Other's Core BusinessesIn the end, if Amazon wants to go after enterprise business, it should buy or create an arm's-length company structured so it won't critically damage the business Amazon already has. If IBM wants to go after Amazon's core business, it should do the same \u2014 and to that end there's a relatively large number of affordable companies IBM could purchase.Analysis: Amazon Web Services Competitors Get Bad News From GartnerThe key to making this work is assuring that the efforts remain at arm's length but maintain adequate funding for success. Remember, too, that imitation is the sincerest form of flattery: Samsung copied Apple, then spent three times as much to market its products (and went to court to defend them). This kind of fight isn't cheap.Right now, Amazon is going after IBM's business, but it's not using an arm's-length strategy, and the result should therefore be predictably bad. It amazes me how many companies enter markets and battles firms despite being ill-prepared and underfunded.If you want to win, emulate a winner. Don't omit the parts you don't want to do, such as funding and assuring focus. If you do, you'll fail \u2014 and that lesson goes well beyond the scope of this column.Rob Enderle is president and principal analyst of the Enderle Group. Previously, he was the Senior Research Fellow for Forrester Research and the Giga Information Group. Prior to that he worked for IBM and held positions in Internal Audit, Competitive Analysis, Marketing, Finance and Security. Currently, Enderle writes on emerging technology, security and Linux for a variety of publications and appears on national news TV shows that include CNBC, FOX, Bloomberg and NPR.Follow everything from CIO.com on Twitter @CIOonline, Facebook, Google + and LinkedIn.