I really wish this was written up as an article rather than a tweet storm. To that end: --- In @stratechery @benthompson offers this (subscribe as you should

I really wish this was written up as an article rather than a tweet storm. To that end: --- In @stratechery @benthompson offers this (subscribe as you should for entire analysis). I totally agree on the general analysis but want to dive into "no one gets fired for buying AWS". Intentionally provocative thoughts: https://pbs.twimg.com/media/DtRbxi9UcAAdEur.jpg https://pbs.twimg.com/media/DtRbxi9UcAAdEur.jpg IBM "invented" the idea of not getting fired for buying IBM. The key to all of this became known as "account control". Providing enough product, support, service, that there is no reason (arguably, impossible as practical matter) to look elsewhere. For decades this was viewed as an insurmountable moat. IBM did not just offer products and services, it essentially "dictated" them. Of course not by force, but by having defined product roadmaps and literally explaining to customers what they needed. The world was awash in confusion and complexity, but not awash in information. IBM exploited that asymmetry. At the same time IBM delivered transformative products and services, and profits, and so everyone was happy. This all unraveled essentially starting with the PC but really it was client/server computing and then the Internet that changed this equation. There were new technologies and new sources of information. The asymmetry was broken. IBM wasn’t the unique supplier it was. But this did not happen overnight. Little Microsoft in the early 1990s made a living selling MS-DOS and Windows to PC makers (half the company) and copies of apps (mostly on the Mac) one at a time at retail. It was a joke of an "enterprise" company. IBM tried to exploit that. Personal story. In Jan 1993 I went to my first "exec" offsite (I was just a kid working for bill). Our small breakout group was tasked with developing a presentation answering the question "How does Microsoft fill the void left by the demise of IBM?" Really that was our group. None of us were equipped to answer this question (I was a developer on C++!) though a) many read Father, Son & Co, b) knew IBM was ~bankrupt c) heard the phrase 'no one ever got fired for buying IBM'. We had about 10 hours to come up with a slide deck. The thrust of our presentation was about how to develop account control. The main tool of account control is providing air cover, a backstop when things go wrong, and comfort. A huge part of comfort is knowing that customers are not buying products but "investing". This means there needs to be more certainty in product roadmaps--primarily when are things coming, how pieces fit together, big changes that will take /, will something break. We spent a lot of time discussing "industry analysts" and how Microsoft "needed to figure them out". No one was really working with these parties but boy customers sure loved them. Customer comfort comes from third party validation not just first party. And so on. We concluded that there is a love-hate relationship with the idea of account control--both from the customer perspective and the product development perspective. But when customers are spending millions of dollars (directly) this stuff matters. Fast forward to today. I am not as clear as above that no one ever got fired for going with AWS. Yes AWS is a huge success, massive and growing, but the product strategy is decidedly different than traditional enterprise. AWS has many products that overlap, may or may not have long term support within Amazon, there are very different update frequencies. This is not about innovation—there's TONS of that. This is about "enterprise"— old school. Does Amazon look like Microsoft did in 1992? Yes to some degree. But it also looks like a retailer. Some things retailerS do are an anathema to "Enterprise IT". In retail you carry 20 products in a category and let brand managers duke it out. Having "n" storage products in the aisle is innovative but maybe confusing. At AWS:reInvent there was some shade thrown in the direction of Analysts. Been there, done that. The problem is it is not that 'no one ever got fired for buying IBM' but 'no on ever got fired for listening to guidance to buy IBM'. There's a difference. Maybe I’m overreacting. AWS has a colossal lead but now that cloud is "done" exploding, there's a new era in IT—"investing". The desire to be "multi-cloud" is a reaction to uncertainty from the leader, not a desire to hedge. AWS has to react to that but not necessarily with innovation. For AWS to truly become the "don't get fired" alternative, customers are going to demand roadmaps, support lifecycles, predictability, reliability, and above all prescriptive coherent guidance that pulls all this together. Annoying, boring, and maybe just reality. Of course Microsoft knows all of this as starting in the mid 1990s all these things started—the EBC, analyst relations, prescriptive guidance, MCS, architecture reviews, advisory councils, and more. Lots of mistakes, lots of flaws, etc. This is not about perfect. Today, in the zeal to compete in the cloud one could easily say Microsoft is adopting many of the same approaches AWS exhibits—what is an AWS objection handler/compete versus strategy, is something a future or a 'project’…Is this new normal? That didn’t work for PC servers. Prescriptive guidance means having a point of view on what technology is best and how to use it. That’s different than a bit of everything everywhere (as a retailer might do). Point of view == strategy == guidance. Risky. Hard. What analysts look at. Who is really doing that? Maybe this is a new era in how IT thinks about software because software is now so pervasive and information so readily available. I am not so sure. Why? Because all this stuff is frigging difficult and there are a lot of choices at every step. Ultimately whole businesses will bet their future on software—partner choices are existentially important, not less than the past. As Microsoft learned, being "best" is a combination of product, price, place, promotion not just innovation. There's a lot to winning. https://twitter.com/stevesi/status/1068579932610027520 https://twitter.com/stevesi/status/1068579932610027520