Skip to content

Don't Believe The Hype

The second in a blog series: “Closing the Growing Gap in Analytics Capability and Effective Use

Cautious Optimist or Happy Cynic – Important Roles for Analytics Leaders to Close the Analytics Gap

Since analytics leaders live at the nexus of data, tech, and business they are often the best suited to tamp down the hype around analytics. Being the happy cynic or the cautious optimist can make you wonder if you’ll be accused of lacking smarts or bravery or ambition; after all everyone else is “doing AI in a fully deployed cloud with a squadron of ML engineers.” But your role as an analytics leader is not to deliver a technology or technique it’s to deliver an organizational capability that ensures the long-term positive impact of analytics at a large scale. And to secure long-term analytics success you need to deliver near term results. So, it’s you who must make clear what the hyped things are and where they live, and then bring them down to reality in the here and now. In this sense, at least you don’t need to play the role of Dream Crusher, more like Enthusiastic Realist.

Hype Is Often A Post-Dated Check (Maybe Drawn on Insufficient Funds)

In essence, the difference between hype and reality in analytics is in timing, scale, and semantics. There are not so many things that are likely to come as demands from your colleagues that will never come to be. It’s more likely that the hyped things are further off than folks expect, have a lower impact/higher cost than they expect, or are expressed with terms that are unclear. Most hyped areas have some elements of all three and some, like AI, have enough of all three to make them problematic. (More on that later).

In addition to the dimension of scale, time, and semantics most hyped things in analytics come in three categories - hype on roles, hype on technology, and hype on techniques. A few examples below illustrate how to reframe the hype and drive the reality.

A good way to start to show and practice your role as Enthusiastic Realist is to lead by example and admit where you have set the organization back by buying and selling hype. An area where many analytics leaders have done so is in believing that data scientists were the magic piece to their organizational puzzle. They were after all the ones with the sexiest job in the 21st century. There is no doubt, at all, that the skills brought in by folks who have the title data scientist have made significant impact in value adding analytics. As a field, we are better off because more people have embraced the skills put under the data science umbrella. Unfortunately, too many organizations didn’t look upon this new role as the addition of a critical team member when they hired data scientists in 2012-2015 but instead believed they were hiring “unicorns” and saviors. So really the data scientist hype fits clearly into the scale category, with a little semantic spice depending on who you talk to. We all should have known that half-joking references about “unicorns” were an indication that we were buying and selling hype.

Bridging the gap on technology can be a little trickier. Technology can be massively intimidating because it has a toxic mix of complexity, strange words, and the sense that falling behind in technology is akin to certain death. Technology often ticks all three boxes of hype inflation; semantic (how unhelpful has the term ‘cloud’ been?), scale (what can blockchain realistically solve?), and timing (when will I get my flying car?). To align technologists, data and analytics, and business folks focusing on the timing aspect, the notion of when a given technology will be right for your firm to pursue most aggressively, is often the best. This is largely because, when it comes to timing, you’re all wrong. Here you can leverage the history of technology, cleverly illustrated in this Washington Post piece to bring into perspective that it takes more time than we think for the most ubiquitous tech to become mainstream. This should not be used to defer trying new tech and put off playing aggressive catch up where you need to but rather to put the size of the effort and the time it will take into a broader perspective.

As mentioned earlier, AI hits the hype trifecta. Semantically, AI is a term that even AI experts don’t agree on. It’s best however to avoid starting here when taking the role to deflate the AI hype – after all the phrase “I don’t want to get into a semantic discussion” is a powerful conversation killer for a reason. From a timing perspective, regardless of your definition of AI, you’re likely to agree that AI has been promised as the next big thing for about half a century. Still there is enough of what folks believe to be AI in the here and now that philosophizing about the notion that the future never arrives is not likely to be very fruitful, fascinating maybe, but not fruitful. The most effective area of hype to unwind is the area of scale, specifically that AI can do all and do it effortlessly. And the way to do this is skip the semantics, park any discussions of time, and go straight for the challenge that your colleague is looking to tackle. Analytics 101 – what’s the business problem you’re trying to solve? Focus the needs, not the solution.

You’re not trying to save you colleague from an overblown sales pitch or a latent desire to be the keynote speaker at the next Adweek conference. Those things could well be driving your colleague’s passion for AI and you can use that passion without concerning yourself with what underpins it. You might never find out what drives it, as is the case with a client I advised. The client, a highly capable analytics VP from a CPG had a new CMO. When they first met this new colleague, it was in a room with a whiteboard where the CMO had drawn two circles, making a donut with a big middle section, so more air than donut (irony forthcoming). In the middle, the CMO wrote “AI” and said – “marketing will be done nearly 100% AI and the most important parts will all be AI.” In this specific instance the analytics leader didn’t have to deflate that hype, since the CMO was gone in three months. I sincerely hope it’s not often the case that people leave jobs so quickly, so I share the guidance we discussed which was based on interactions with several analytics leaders in the Analytics Leadership Consortium - all of whom have faced a similar AI Enthusiast.

  1. Match the enthusiasm. Greet the AI Enthusiast with “Yes! I am so glad you see the power of advanced analytics like AI, Machine Learning, and Predictive Analytics.” If they want to discuss terms, go for it, but more likely they believe these three are the same or close enough. Maybe they’re right.

  2. Drive problem understanding. Ask, “Tell me why these problems are so tough that you want to use AI to crack them? What have you tried before or are using now?” These questions not only help you repeat your enthusiasm for their ambition (“those are really interesting problems, looks like you guys are working really hard”), it also deepens your understanding of the problem and helps you understand what they think AI is, something you might have to deal with later.

  3. Repeat your commitment to solving the problem in a cost-effective way. In an environment of finite resources and fast-moving competition (which is the world of every organization), you as an analytics leader must commit to solutions that deliver on time and on budget. And your colleagues must as well. Alignment on the problem and the need to be good corporate soldiers should unify you enough to get past the semantics of AI.

While it’s understandable that you might view this approach as avoiding conflict, or even being a bit sneaky, I see it differently. I see it as a way to postpone any conversation about the power of AI or the magic of cloud or the genius of MLOps or whatever until you have walked parts of the journey together, have more common experiences, and see the picture more clearly. And if that sounds too kumbaya for you, consider the effectiveness of the alternative where you define terms and speak in abstraction in an effort “educate” people on what AI or Cloud or MLOps really is. This technique I know from personal experience will likely result in you educating people on what a Condescending Nerd looks like, and that’s a role you really don’t want. Nerd, yes, Condescending Nerd, not so much.

About Drew

Drew has close to 20 years of experience, having worked on both the business side of analytics, leveraging insights for business performance, and on the delivery side of analytics driving the use of enterprise analytics. As the lead of Analytics Leadership Consortium, Drew drives engagement with analytics executives and top analytics practitioners in the IIA Community to help them lead their firm’s journey to analytics excellence.

Before joining the IIA, he led the Enterprise Data Analytics and Governance function at IKEA’s global headquarters in Europe. He leveraged analytics in various leadership roles across the IKEA value chain in both the United States and Europe. He received his MBA from Penn State and his undergraduate degree from Boston University.

About The Analytics Leadership Consortium (ALC)

The Analytics Leadership Consortium (ALC) is a closed network of analytics executives from diverse industries who meet to share and discuss real world best practices, as well as discover and develop analytics innovation, all for the purpose of improving the analytics maturity of their firms and securing the business impact they deliver.

You can view more posts by Drew here.

Follow IIA on LinkedIn, Twitter and Facebook for more updates.