The Bias of Innovation

This article extends my earlier thinking on the topic, Systematic Innovation Failure.

Bias in Happiness

People tend to believe earning more will lead to increasing happiness. If you’re making twenty thousand dollars doubling your income to forty will, by all accounts, actually increase your happiness. But doubling your income again to eighty and then doubling it from there, and so on, doesn’t lead automatically to increases in happiness; diminishing returns above basic economic necessity is pervasive. Researchers have realized that by having more to spend increases what we want to spend it on, and that we will always have insufficient to achieve what we think will make us happy.

Psychologists have, in fact, researched what truly makes us happy. The core findings include that not only we don’t know what will make us happy, we believe quite strongly that we do know what will, and we tend not to learn from experience (i.e., we make the same errors over and over again) in our quest for happiness. Before we even start, we begin from a biased series of assumptions which set us back. The dynamic is roughly as follows:

1. We are confident we know what will make us happy.
2. We are confident in our confidence about this.
3. We actually don’t know what will make us happy, and we don’t learn from our experience about it.

The research on happiness is hardly unique or alone in suggesting that there are major systematic biases in our understanding of the world. Daniel Kahneman won a Nobel Prize for his research on this, with the development of Behavioural Economics. Frustratingly, humans are rather poor at learning to overcome the biases, such that Kahneman reports he makes the same biased errors in the exact same situations he has demonstrated people make biased errors in.

Bias with Unlocking Innovation

Efforts at innovation likely follow the same type of dynamic:

1. We believe we have a solution to a problem that the market has.
2. We are confident in our belief that our solution will solve a problem the market has.
3. We actually don’t know what the market wants, and we are somewhat immune to effectively cracking this problem.

So What?

Working with early stage startups that are focused on creating and launching market changing innovations, it’s not uncommon to see presentations focused on non-stop feature descriptions that solve an imagined problem. There is a challenge to tac toward actual customer needs. Many teams think they have a value proposition, but they don’t, or they might but aren’t certain. Some spend years building so that they can eventually discover a lack of demand. While some people external to the startup seem to be able to quickly identify the lack of customer need being addressed, those putting in the time and energy seem to miss the mark. Many of the startups cause prospective mentors to wonder “what’s the value to someone who will pay?” While many potential customers just can’t see the value in buying what’s on offer, and never buy a thing.

This is demonstrably a broad based issue with early stage startups: something like 80–90% of startups fail, if we believe the work by CB Insights. CB Insights further suggests that bulk of these failures are due to some type of lack of value (i.e., no market need). Corporate innovation projects are also frequently challenged though this type of data is harder to come by.

The Top 20 Reasons Startups Fail by CB Insights

Despite seminal works like Lean Startup, innumerable blog posts providing advice, accelerators, innovation hubs, government programs, etc. we still seem unable to increase the odds beyond an 80% failure rate. That said, some (i.e., Y Combinator) may have found something.

Becoming Less Mired in Failures

If poorly defining what value is being provided to prospective customers is a problem, how do we avoid the problem or find a solution for it?

The Business Model Canvas Template

One area to chew over is how frameworks like the Business Model Canvas set-up innovators for failure. While the Canvas has clearly been useful for simplifying and clarifying key business elements, it partially ignores the core element of what makes an innovation effort challenging: the value proposition itself. It takes this as a given. Everything flows from this core element: pricing to channels, financials to distribution. For the early stage startup trying to deliver the value to a customer, nothing else really matters.

If you don’t have that core value element it simply doesn’t make sense to talk about anything else, full stop. It’s unclear why incubators and startup hubs do anything other than focus exclusively on this.

The Value Challenge

Some attempts to crack this challenge have been undertaken.

The Jobs-To-Be-Done approach is one. There are others. In most cases, it’s a labour intensive effort involving serious piles of cash. But the lessons from them are probably generalizable to startups/innovation efforts.

The earlier points around our inability to understand what will make us happy, our false confidence in our ability, and our challenge to even learn from experience returns here. Initially, the three ideas are:

1. We believe we have a solution to a problem that the market has.
2. We are confident in our belief that our solution will solve a problem the market has.
3. We actually don’t know what the market wants, and we are somewhat immune to effectively cracking this problem.

So we need a mindset that outlines:

1. We need to first understand that we don’t know what problem the market actually has, though our solution might be an answer and we have an idea that it might work.
2. We then need to dispel any confidence that our own solution might work.
3. We will need help to understand what the market wants, and need support to crack this problem.

So we talk to Customers?

To uncover what a customer actually values is a labour intensive, unrewarding task. It’s uncomfortable to avoid telling people you have a solution to the problem you imagine they have. Like conducting good science, you spend time testing a hypothesis by trying to disprove its validity. Like the scientist, the innovator works in areas of interest and actively tries to disprove a hypothesis all the while, the hypothesis is the cherished idea of the innovator. The incentives are seemingly incompatible with good science. Add to this the challenge that what the market/customers will value isn’t nearly as defined as natural or social phenomena.

Into this mix, we must tread carefully as we will quickly and confidently find fragments of information and data to support our pre-existing solutions and ideas. We will avoid learning from the feedback we get from talking to potential customers about what they value. Instead we naturally and quickly reinforce the narrative we already believe. Quickly we start building a business on a foundation of sand.

Add to the mix a tendency for penetrating inquiry and questions around the value to potential customers to be seen as combative arguments that can be ignored as such. It’s easy to rip things down, to assume others just don’t see the light. This can come from mentors advising the innovator, but also the potential customers the innovator questions to understand value. Layer in a drive to get some sort of funding, to demonstrate traction, from investors so you can pay yourself something as the innovator.

The mantra of “talk to customers” is, bluntly, insufficient. It’s rarely combined with tools and solutions to drive the innovator towards a success rate of anything more than 20%; a roll of the dice. It’s fine to listen to podcasts venerating successful innovators for their drive to uncover unmet needs, or delve into innovation frameworks that require consulting contracts worth hundreds of thousands of dollars. But how do you actually do this?

Value Discovery?

Conceptually it’s fairly straightforward to move through the first of the points that inhibit happiness discovery as well as the inhibit the innovator. Academically at least:

1. We can accept that our solution to the problem the market has may not actually be that fantastic.
2. We can then accept that our confidence in the solution we have to the problem is misplaced.

Once accepting the first two points, the third becomes the challenge: what can we actually do to change the odds of success from something close to 20% to something more? What is actually involved in building an understanding of what customers value? How do you actually structure ourselves for better success at innovation? Aside, that is, from hoping you are the undiscovered artist capable of revealing the needs of a customer and what they value, there is little to rely on to increase the chances of innovation success. For the startup innovator, a tool/solution/framework is needed to increase the chance of success.

A Structured Process

A decision support framework is a complicated way of basically saying “use a checklist.” Across a number of fields, the use of checklists has quite literally saved lives. Airplane pilots use checklists, though this doesn’t negate their training and expertise. Increasingly, medicine uses checklists, though this doesn’t negate a doctor’s training or expertise.

Entire careers and areas of expertise have been built simply focused on this, such as Farnam Street. In finance, top investment managers use very structured decision making processes. Principles by Ray Dialo describes exactly this. While Annie Duke’s approach to how to think through problems in a structured way, where the outcome is unknown, fall into this broad area.

In each of these cases, tools are being used to help the expert overcome natural biases and their challenges. To increase the odds of success for the startup innovator, the challenge is to build tools, frameworks, and decision support processes to reduce the 80% chance of failure.

It’s likely with this type of solution that the startup innovator can overcome the third challenge of innovation: we need help to understand what the market wants, and need support to crack this problem.