Tag Archives: innovation

COVID-19 Is No Black Swan

Many commentators have labeled COVID-19 a black swan event. Nassim Taleb popularized the expression, defining such an event as impossible to predict, having a major effect and seeming obvious in hindsight.

Yet, there were clear warnings from credible institutions. Three examples stand out, showing the crisis should not have been a complete surprise.

Epidemiologists had warned in the 2019 Global Preparedness Monitoring Board report that the chances of a global pandemic were growing and that the world was unprepared for a fast-moving, virulent respiratory pathogen pandemic.

The latest U.K. National Risk register identified pandemic influenza as a top high-impact, high-likelihood event.

Then, last October, the U.S.-based Center for Health Security organized an eerily prophetic pandemic simulation involving a coronavirus similar to COVID-19.

Can we still call COVID-19 a black swan when all those warnings were missed?

If COVID-19 is no black swan

The grey rhino, a term coined by Michele Wucker, is a cross between black swans and elephants in the room. Contrary to the low-probability black swan, the grey rhino is a high-impact, high-probability event usually ignored for various reasons. Climate change is a typical example, which until recently was discounted by investors, policy makers, corporations and wider society.

The grey rhino theory has many attractions. Rather than focusing on hindsight, it asks whether we will do something about an obvious problem. The theory embraces the fact that many things that go wrong in business, policy and our personal lives are avoidable, if only we had paid enough attention. It recognizes that the issue is generally not a case of if but when

Your black swan may be someone else’s white swan

Having diverse, multi-disciplinary boards can ensure a less blinkered review of risks, especially if the organizational culture values the input of the Tenth Man (or Devil’s Advocate).

Groupthink allows statements such as “impossible to predict,” so a risk register review by external advisers is a good idea, bringing fresh perspectives. War-gaming and red-teaming are also useful techniques successfully imported from battlefield to boardroom.

The past informs our predictions of the future

An extensive historical review, going back to 1000 AD, underpins the taxonomy of threats behind the work of the Cambridge Centre for Risk Studies and its established Lloyd’s City Risk Index. Obviously, new threats (such as drones) need to be factored in.

Modeling allows us to go beyond the historical record. For climate change, general circulation models, in particular, are used to model the whole of the atmosphere and ocean system and are key in understanding how global warming is affecting the scale, intensity and frequency of extreme events like floods or tropical cyclones. Scenario planning is another useful technique to communicate the results of complex models to the public and decision makers.

See also: Risks Facing the Tokyo Olympics

The real black swans — the “unknown unknowns”

The best strategy to plan for these may be to maintain a constant state of preparedness, irrespective of the specific nature of the threat.

Increasing our threat-agnostic resilience could be a good investment, allowing citizens, corporations and governments to prepare for a national crisis without knowing exactly what the contingencies will be.

Recognizing that crises should be expected rather than the exception, robust business continuity plans are a sure way of improving resilience – designed properly, tested and updated regularly, they could prove versatile and may be your best insurance against the next black swan.

Learn to better spot grey rhinos

Grey rhinos are more common than black swans. Focusing on spotting them would promote a proactive rather than a fatalist approach to risk. A holistic approach to risk management and resilience is a good way of turning grey rhino risks into opportunities, and corporations can ask their chief risk officers to coordinate a cross-departmental approach. It should be cause for concern that most governments still do not have a national chief risk officer, alongside their chief medical officer and chief scientific officer.

OnStar: Next Step for OEM Partnerships

GM announced this week that it will start selling policies directly to drivers through its new OnStar Insurance program. The company will start with its employees in Arizona and plans to expand to the general public by the end of 2021. This move has gotten some attention in mainstream sources like the Wall Street Journal.

The policies are underwritten by American Family Insurance affiliates and issued on its paper. In that sense, this partnership is an extension of the insurer/auto manufacturer data-sharing partnerships that have proliferated in recent years. Insurers have inked these deals to create a new way to collect driving data that’s easier for the driver than installing a device or downloading an app.

These deals have typically involved the original equipment manufacturers (OEMs) sharing driving data with the insurer, which can use it to score drivers and offer discounts. In GM’s case, it’s also been a marketing partner—you can head to OnStar.com and enter your state to find insurers that might offer you a discount if you’re a good driver.

What’s different about OnStar Insurance is that the OEM offers the policy under its own (subsidiary) brand, directly to the driver. It’s the logical extension of the relationship for both partners. For GM, it’s another way to monetize its data stream, which OEMs generally have trouble doing. AmFam gets additional marketing muscle and a new direct sales channel. This development was more of a “when” than an “if.”

Telematics is here to stay, but bullish mid-2010s projections for telematics adoption and growth haven’t materialized. OEM/insurer partnerships already had the potential to open up market space and reduce the barrier to entry; OEMs offering insurance directly to drivers may expand and accelerate this.

See also: The Evolution of Telematics Programs

How these policies are sold will be a significant factor. Most drivers opt into or out of OEM data sharing when they buy their vehicles; if these policies follow the same pattern, it will be on individual dealers to push the insurance offerings. On the other hand, it’s easy to imagine OnStar alerting a driver who’d agreed to share data but hadn’t bought OnStar Insurance that the drive could save X amount by switching.

User experience will also be a major determinant. GM currently plans to use OnStar to coach its drivers. That’s potentially a major plus, but only if the coaching is done well and the user experience (UX) is slick. Consumers have high expectations around digital experiences, and half-efforts won’t go over well.

Finally, one potential drawback: It looks like a major draw to OnStar Insurance will be price, which further cements the connection between telematics and discounts. Reward- and engagement-based telematics programs have been successful overseas, but they haven’t taken hold stateside. Anecdotally, insurers have told me that U.S. consumers expect upfront discounts as an incentive for enrollment in telematics-based insurance. Offerings like OnStar Insurance could further entrench the idea that telematics is all about exchanging data for money.

I recently published a report on telematics that covers these and other issues. Feel free to reach out to me with any questions!

How AI Transforms Risk Engineering

In a year marred by crisis and uncertainty, the mature property and casualty (P&C) insurance industry has seen its workload increase in both volume and complexity. According to the Insurance Information Institute, insured losses from natural catastrophes in 2019 totaled $71 billion. That number is only expected to rise in 2020 with the onslaught of hurricanes and wildfires hammering the U.S.

Insurers must contend with a rapidly changing risk landscape. Falling interest rates, climate change, man-made risks and civil unrest are causing unprecedented destruction and business interruption. This is exacerbated by the COVID-19 pandemic, cyber security threats and global terrorism, causing the number of claims to skyrocket.

Traditional methods of risk analysis are slow and expensive. Risk engineers spend considerable time performing repetitive assessment and administrative tasks that do not add value to clients.

One saving grace is the global movement toward digital transformation and automation, including the adoption of artificial intelligence (AI). Changing client expectations have propelled organizations to rethink age-old processes. 

An artificial intelligence study by PwC said, “AI could contribute to the global economy by 2030, more than the current output of China and India combined.” The same report estimated $6.6 trillion would likely come from increased productivity alone.

See also: Stop Being Scared of Artificial Intelligence

How do you know if you’re ready to embrace AI, and what are some of the areas it could improve within risk engineering? Below are three points to consider:

The easiest way to get started, is to contemplate your market in five years’ time and consider what capabilities you will need to compete – McKinsey

1) Align Business and AI Goals.

A certain appetite and readiness for change is required on the part of the C-suite and by the risk engineers within your workforce. A real pain point must be met, and the implementation of AI must align with the overarching business goals of your organization. For risk engineering, the time is ripe for AI disruption. According to McKinsey, “Efficiency improvement is an imperative. The industry’s current trajectory is inefficient and unsustainable, creating the conditions for disruption. This would involve digital technologies, automation and data and analytics to not only reduce error-prone manual processes but also enable an agile way of working.” 

If account engineers and risk engineering consultants spent more of their time on risk verification and selection rather than aggregation and analysis, this would help underwriters speed up the time to assess and quote on a new bid and ultimately increase the chances of winning business. The first response to a submission wins over 50% of the time. 

Still, the question remains, whether your organization wants to be an early adopter, fast follower or follower. Will the AI solution you create in-house or via a third-party vendor disrupt the sector and provide you with a competitive edge?

2) Examine Internal Talent. Find Your AI Champions.

Another critical factor is talent. Are there champions within your company willing to take on the added time it requires to inform the user journey and customizations, perhaps even label the initial data and ultimately execute on the AI opportunity at hand? It is vital that there is a top-down and, equally, a bottom-up culture of adoption for AI implementation to succeed.

A global digital practice survey revealed that insurance companies are attracting less digital talent than other financial services companies such as fintech and asset management. In a recent survey, 80% of millennials said they have limited knowledge of the insurance industry, and 44% said careers in insurance sound “boring.” Orbiseed’s recent interview with a veteran risk engineer also revealed that the majority of senior risk engineers are close to retirement and may resist employing new technologies. “Indeed, perception can shape reality, and the current reality is that the insurance industry isn’t viewed as relevant or exciting to up-and-coming digitally savvy workers,” the report concluded. 

3) Partner With AI Vendors You Trust to Scale Quickly.

An AI firm should know your industry inside and out, have secure networks to help protect your data and enable you to scale your AI program fast. You will also need to consider whether to select AI integrations over ground-up builds. An integration will vastly reduce the time it takes to produce a working model for your business. A good software integration will also layer into the existing system you have rather than force your employees to learn an entirely new system.

See also: 3 Tips for Increasing Customer Engagement

Next Steps Toward AI Transformation in Your Organization

AI is fundamentally changing the way business is done in 2020. For mature industries that still rely on manual, labor-intensive processes, adopting new technology can make a measurable difference in efficiency and deliver significant competitive advantages.

Risk engineering seeks to manage risk: Adopting AI practices early will ensure that your organization hedges against the risk of falling behind the competition. Firms that effectively adopt AI early report significant performance gains compared with competitors, including higher revenues and reduced expenses.

Designing a New Employee Experience

As the pandemic has accelerated the digitization of interactions with customers, the insurance industry has begun to rethink the customer experience. But companies now need to start to redesign another experience, too: the employee’s.

We know that two highly effective vaccines should be available in quantity in April or so, which means that they could become a reliable line of defense against COVID by next winter or perhaps even next fall. That gives us roughly a year to design and implement a new work environment, a new experience for employees that incorporates the best of the old, in-person model and the best of the new, digital/Zoom experience.

What should that new environment look like?

A smart recent piece in Strategy & Business provides a systematic way to start adapting now to the new world of work, by taking the same sort of approach to the employee experience that companies routinely take to understand the customer journey. The piece promises that companies have a “once-in-a-generation opportunity to increase engagement and productivity.”

For me, the smartest suggestion in the piece is a variant of the test-and-learn approach that many companies have begun to implement in their formal innovation programs:

“Many companies conduct annual employee engagement or satisfaction surveys. Our advice: Throw them out, at least for now. What you need now is a steady series of short pulse surveys and conversations that ask employees to name their three biggest time wasters or other headaches. Focus on tools (‘Do you have what you need?’), authority (‘Are you empowered to make decisions?’ or ‘Is it easy to get approvals?’), and distractions (‘What pulls you away from the task at hand?’). Turn those answers into a Pareto chart, start working the list, and come back the following month to get new insights.”

The authors note that, while we focus on the change in our customers’ behavior and demands, we should also realize that internal customers have changed, too, because of work-from-home and increased digitization. So, we should use process maps and customer journey tools to understand the flow of work and make sure we’re serving internal customers as effectively and efficiently as possible.

The piece argues for starting now (if we haven’t already) to rethink the mix of talent we’ll need and suggests a key adjustment in management. Because the timing of work has to be less predictable, as people juggle kids at home, management needs to be more flexible. The focus needs to be on objectives, not schedule, the authors argue.

The authors say staff development poses a particular problem. So much depends on mentorship, which is far harder in a world of remote work; they recommend continual focus on finding other ways to build relationships that nurture talent.

Finally, they underscore the need for even more communication than normal. With so many working remotely, people can lose the thread. So, employees need to continually hear what the big picture is and understand how they fit.

The piece I’d add is that special attention needs to be paid to innovation, because it depends so heavily on the sorts of interactions that just aren’t possible in a Zoom world. When I worked on an innovation-related project at the Department of Energy a decade ago, I was struck that Secretary Steven Chu’s redesign of the national research labs spent a lot of time on schedules and the design of the cafeterias, because he wanted to encourage as much informal interaction as he could among his scientists. When my younger daughter somehow secured us a viewing of “Up” at Pixar headquarters a month before it debuted in theaters in 2009, people talked about how carefully Steve Jobs had designed the building, down to placement of bathrooms, to encourage informal interaction among all those creative folks. Kevin Kelly, the co-founder of Wired, says the new world of work should see people doing all their work at home and just coming to the office from time to time so serendipity can work its magic through chance meetings in the hallways.

Those chance meetings will become possible again, both in hallways and at in-person industry events, but not until the virus is well and truly contained. So, we need an interim strategy for innovation and all the other issues raised in the Strategy & Business article that will last us at least a year and perhaps longer, given how slowly the world will probably return to normal even once a high percentage of us are vaccinated. We should also take the time to pull out a clean sheet of paper and design the world of work that we want to inhabit once offices and travel become widely available to us again.

People have never been more open to change than they are now. Let’s seize the opportunity.

Stay safe.


P.S. Here are the six articles I’d like to highlight from the past week:

How to Outperform on Innovation

It is up to all of us leaders to advance diversity and inclusion. It’s the morally right thing to do, and it’s the only commercially smart answer.

Technology and the Agent of the Future

Technology promises to free agents to spend more time with clients and prospects, broadening and deepening relationships.

4 Initiatives That Unlock IoT’s Value

IoT has largely been used in tactical ways to solve specific problems, but there is great strategic value if it is tied to certain types of initiatives.

How COVID Alters Consumer Demands

Digital transformations that would have taken three to five years are now happening in under six months.

Opportunity Among Latinos in U.S.

It’s crucial to understand the distinctions in purchasing behavior when comparing the Hispanic market with the broader insurance market.

P&C Distribution: What’s Old Is New

There are eight different models or options for insurers to consider — but it’s fair to ask if these distribution models are really new.