The Cycle of Insight
To create something new requires entering a murky realm.
This picture, inspired by a drawing I saw on Twitter but cannot find, captures the process perfectly:
This applies to anything, but today I'm writing about it strictly in the business environment. And I mean literally anything:
New products
An accounts receivable management process
A marketing video
Developing corporate values
As a visual representation, it also scales from the individual to the organization.
Sony started as a company with the purpose of being a company. It's first product? Rice Cookers.
Pixar's masterpiece Up began as a story about a king with two sons, and ended in a drastically different place.
The process of bringing ideas into reality is fundamentally non-linear. But we humans control progress more easily when we act in a linear, sequential way.
Here's the problem: great stories, great companies, great systems emerge out of the murk. They depend on a non-linear process that doesn’t obey a rigid calendar or step by step system.
Here's an all too common scenario when a new company pushes to hard to force a new idea into the market, as told by the brilliant Steve Blank:
Webvan, the first online grocery startup to receive substantial investment funding, raised $800 million in the late 1990s.
In such a company, when the new idea fails to gain traction, leadership concludes that failure was not because of anything except failed execution:
Sales starts to miss its numbers and the board becomes concerned. The sales VP arrives at a board meeting, still optimistic, and provides a set of reasonable explanations. The board raises a collective eyebrow. The VP returns to the field to exhort the troops to work harder. Sales asks Engineering to build custom versions of the product for special customers, since this is the only way that the increasingly desperate sales force can close the sale. Board meetings become increasingly tense. Shortly thereafter, the sales VP is probably terminated as part of the “solution.”
But actually, the failure wasn't poor sales, it was to concretely understand the market problem.
The company ended in total failure.
In fairness to Webvan's leadership, the process of creating a new company (or product, or process, or marketing campaign, etc) should not be free from any constraint. Time and resources are necessary constraints. There must be some control.
The solution is to systematize the non-linear process and clearly establish expectations.
A great solution to a problem emerges from a cycle of understanding. Ideas begin vaguely, perhaps with a sparkly end state, but most often don't hold up to simple questions, like:
how does it work?
what is the sequence of events to get there?
Figuring out your customer, or your company values, or your strategy, is not an event. We must write our plans with a pencil, updating assumptions when new insights emerge, as they inevitably will.
The Cycle
1. Assemble a council of advisors.
This is easier if you are inside of a company with a shared set of goals. Theoretically, this is very easy alignment, but it takes sustained effort to build the kind of candor required to achieve insightful dialogue.
In Good to Great, Jim Collins' found that companies who succeeded in establishing a culture with a clear mission and compelling values spent significant time working on and defining their values. It took an average of 4 years to figure out what they were doing. They generally had a council of advisors of 5-12 people who routinely came together to debate and revise what they were working on.
2. Create a concrete vision of the outcome you will achieve.
Your vision is a hypothesis that describes what success will look like in detail . This hypothesis must be explicitly clear and concrete. Humans do not do well with abstract ideas, and so a critical foundation to creating something new is making the assumptions and expected results of our project concrete.
3. Accept that when you start, you are starting with a hypothesis of how it will work.
Just because assumptions become concrete doesn't make them correct. The next step is to test them. Steve Blank counsels:
Start by asking yourself, “What insight do I need to move forward?” Then ask, “What’s the simplest test I can run to get it?” Finally, think about, “How do I design an experiment to run this simple test?”
There are limitless ways to run tests. For example, you are creating a marketing video about your shiny new product launch of healthy on-the-go snacks. As a first step, you hypothesize that the buyer is 25-35 year old male, and you need to make sure your messaging resonates. You speculate that it's going to be rugged people and wealthy. Before you start shooting, you put together a couple of images and tease them on Twitter and in AdWords to join a webinar for rugged snacks to see the type of response.
4. Formally reflect on the results of tests.
Once a test is complete, perform an "autopsy" to identify strengths. The language is morbid, but the point is that the test is done, and it can be inspected in detail without any further changes. This examination will produce the updates to assumptions and the design of your approach.
5. Repeat.
Working on something important probably requires another round of revision and testing before it is complete. Once the autopsy is complete, it is a time to again assemble the council of advisors, update assumptions, define the next hypothesis, and test it again.
Whether you are operating solo or as a part of a team, systematically following the steps of this cycle will lead you to embrace the necessarily messy process that produces results.