The shift from experts to experiments (Part 2)

Most successful business activities are the result of careful planning by experts: people paid to use their intellect to make a balanced recommendation or decide a course of action. While a seemingly logical and dependable approach to making business decisions; it is hardly infallible. Just look at how few economists and financial experts were able to predict the global economic crisis. There are plenty of examples in other industries too: the car industry that dismissed the potential of electric cars; the media empires who missed the social media revolution; the failure of Gillette to appreciate the profit potential of mail-order razor blades. These are just a few cases where meticulous analysis by industry experts got a call utterly wrong. In fact, many so-called authorities are often trumped by people who are decidedly not. How can this be? Why do experts routinely get it wrong? More importantly, if this continues, where should we turn for answers?

The declining value of expertise

Almost all fields rely on experts to perform very specific functions. If you need brain surgery, or plumbing fixed, or advice on tax — then the best expert in that field is who you need to seek out. Same goes in large, successful organizations ­­­– you can’t defensively operate reliably or safely without ensuring your team is full of high calibre expertise.

But, as our world continues to change and be influenced by digital and virtual breakthroughs, more and more, it appears that the reliability of expert opinions and predictions are waning, and aren’t necessarily what we should be hanging our hats on.

Consider that multiple industry experts — heads of technology companies, industry experts — predicted the iPhone would be an abject failure.[1] This same group also saw bright futures for Nokia and BlackBerry. IT experts were aghast at the lunacy of Amazon building their own network of data centres that ultimately became the Cloud. Blockbuster turned down the opportunity to buy Netflix. Kodak didn’t think digital photography was worth exploiting. Oh, and back to Apple — Steve Jobs thought letting anyone outside of Apple build apps for the iPhone was a dumb idea.

The real danger — relying on experts for complex problems

In last week’s blog, we discussed the differences between Complex and Complicated problems. Go have a read — it’s important. TL;DR — complicated problems are like fixing a Ferrari (you can become an expert) but complex problems are like knowing the impact of self-driving cars on alcohol consumption. Complicated problems need experts to work out the answer, complex problems need one to run experiments to better understand the problem.

The most significant problem we face in the banking industry is that we have shifted irrevocably from a complicated domain — where we could generally rely on subject matter experts — to one where we increasingly need a different process to supplement decision-making.

We can’t be certain an expert in consumer banking — who grew up with cheques and mortgages — will be best to advise how millennials will save for a home. We can’t rely on television marketing experts for how best to attract new customers who only watch YouTube and Twitch. We can’t be sure a usability expert will get it right the first time when speculating how customers are going to talk to their Alexa device to do their banking.

We can (and probably will) continue to ask our experts on what to do next. But — and this might be humbling — research suggests that sometimes we might do just as well asking a non-expert. The Smithsonian institute discussed this in 2012, noting that in many fields, an expert’s predictions were not much better than random guesses! They followed up with one startling observation that experts appeared to be slightly better at predictions when operating outside their area of expertise.

I’m certain we still need experts — existing knowledge and expertise aren’t irrelevant — but in truth they cannot be relied on with the same confidence as we embrace unpredictable outcomes of disruptive technologies. Likewise, traditional structures, approaches, and risk appetite prove challenging when building competitive offerings using emerging technologies, which have created a high pace of change and uncertainty. Industries undergoing drastic shifts, as a result, need to rely less on our experts and more on our ability to experiment, test concepts and improve our understanding to determine the best path forward.

Shifting from prediction and analysis to experimentation and experience

In a complex domain, one solves problems through learning. Experimentation is about learning; it’s about taking an idea with promise and relying on real-world feedback, rather than speculation. It’s an important business strategy to cope with disruptive threats and uncertain opportunities, providing a way to gain a better understanding of a murky domain or an idea, without betting the entire organization on something unproven. In today’s digital economy, with rapidly changing customer behaviour and many new offerings from which to choose, companies are turning to experimentation to understand the most pertinent needs of their customers and how best to meet them.

Most of us today work in dynamic industries with a high level of flux, where all the fundamental building blocks are changing underneath us. Changes in technology and consumer behaviour requires problems to be addressed not solely from a base of expertise or experience, but through keen analytical and critical thinking that considers many possible outcomes. To do so, we need to be prepared to challenge our preconceptions. And above all, we need to shift from relying on expert opinions to embracing uncertain outcomes, testing-concepts and, ultimately, learning.

[1] “There’s no chance the iPhone is going to get any significant market share. No chance.” — Steve Ballmer, CEO of Microsoft, 2007