Bulletin 28 September 2018. What are analysts analysing, anyway?

Differentiating between philosophy and technology

Honour is understandable between thieves. After all, if I start stealing from you, what’s to stop you stealing back? And, similarly, few analysts go out on a limb to question or correct their peers. They might question back, and then, all hell would break loose. Of course it happens, on occasion: most often a smaller analyst firm will call out one of the industry behemoths, in the knowledge that, frankly, nobody is listening that hard anyway. It’s like attacking HMS Victory with a pea shooter. 

I’ve done it myself, on occasion, largely by virtue of being unable to keep my mouth shut than any deeper motive. Over a decade ago, when IDC announced to the world that we would run out of storage, I called BS: the very notion was impossible (and showed a deep misunderstanding of resource management). We haven’t run out of any such thing, for the record. Nonetheless I do feel vaguely embarrassed at having spoken up. Just like that time at EMC, and that time at… oh well, we are where we are. 

At the same time, I wonder if we do enough cross-scrutiny. The trouble with IT industry analysis is, well, where to start? but it comes too easily from losing a sense of objectivity. Technology products and services generally exist to solve a problem: the better we can define the problem, the easier it is to create a solution. For all the Three-Letter Acronyms, terms and  we churn out, therefore, each one offers a touchstone, a rallying call. We saw it with ERP, then CRM; with ORBs and ESBs, with SDN and NFV. What they mean matters less than the fact they create a space. 

Sometimes, however, this process goes wrong, most often when a trend emerges from the front lines rather than being triggered or catalysed by providers. We saw this with open source and SOA, then Cloud and BYOD; more recently we’re seeing it with DevOps. The models analysts use to define a space fall short, as do their philosophical basis: “What’s the market for cloud?” is one example of how traditional analyst thinking comes unstuck (answer: if you’re asking that question, you really don’t get what’s going on). 

This does lead to some fascinating disconnects. It’s the analyst’s job to be the trusted third party, to make the complex simple and help guide decisions. Yesterday, I was in a meeting with a pretty big computer company, talking about a topic currently close to my heart. “That position aligns with our thinking, but we’re not hearing any other analysts talk about it like that,” I was told. Again, the topic is irrelevant (as is the fact I know several analysts that could help out). More telling was the fact that the analyst community as a whole wasn’t helping. 

For sure, both me, the big company and a few other people I know could be completely wrong, but I don’t think that this the case. Rather, in this case, the topic (OK, it was DevOps) is seen as an end-point, and any products associated with it are a means to that end. Nothing could be further from the truth: for sure, it’s a good place to get to, but it is merely a point on a bigger journey. Let me put it this way: you won’t become any better at delivering innovation quickly by implementing DevOps tooling. That would be like buying some breathing apparatus, and becoming an expert diver overnight. 

Of course, it’s never a good thing to buy some tech and believe it’ll change things, but in many cases, it does get a result (remember getting your first mobile phone?). And often, tech does need a rallying call, to help explain why it’s a good idea; it’s also a lot easier to talk about something if it has a name. However, just because the industry works best with bandwagons, we shouldn’t just assume everything needs one and in some cases, it can be counterproductive. 

The bottom line is, where technology exists to support a philosophy, we should act very differently to where a new way of working falls naturally out of buying a particular piece of tech. Sure, there’s a spectrum, but we should be able to differentiate between the two. 


Smart Shift: Brainstorming the future

The next section in Smart Shift is now up. It covers Robert Boyle and the Royal Society, unicorns and VCs, smart belts and hype cycles. Feedback welcome, as always!

Have a good weekend and all the best, Jon

Also published on Medium.

Bulletin 28 September 2018. What are analysts analysing, anyway?

Leave a Reply

Your email address will not be published. Required fields are marked *