Tag Archives: experimentation

The shortest path to innovation

there is no innovation without experimentation

In the past week I’ve had some interesting conversations with colleagues, friends and random people about culture and innovation capability. There are a couple of themes that have come up, one of which I’ll touch on here: feedback as it relates to innovation.

First, let’s put one thing on the table: there is no innovation without experimentation.

Why? Because…

Why should companies launch imperfect products?

Why should companies launch imperfect products?

Although we think there are exceptions to the rule (Apple, Square), no company ever launches a complete product.

The Lean Startup advocates that entrepreneurs can and should launch products and services that are not %100 percent complete. This idea, of constant experimentation, is not new. Most products that are launched by startups are an initial prototype that tests for market validation.

Big companies, by their nature, don’t do this. At least not all of them.

Starbucks, for example, is an outlier. If you’ve read about how Statbucks got started, then you won’t be surprised. As outlined on a Fast Company article, they’ve recently taken to experiment with new marketing channels, such as Groupon, and in doing so put their huge digital platform to the test:

No investment in innovation means no investment in the future

accenture why low cost innovation is risky

“It’s absolutely ingrained in human nature that we simply assign responsibility for what went wrong instead of chalking it up to bad luck. And bad luck is out there. You can’t totally insulate yourself from it because if you try to, you really impair your ability to achieve good luck. And unfortunately, the world is sufficiently complex that you have to make the conditional assumption that you’re going to get some breaks along the line. Otherwise, you’re dead.” – Bob Hamman

Investing in innovation isn’t paying off for most companies. 93% of companies feel innovation is important, but only 18% believe it’s effective. That is the conclusion from a recent report from Accenture titled “Why low risk innovation is costly”. This isn’t surprising. 96% of all innovation investments don’t pay off. Of course, we’d love to flip that number and see our innovation efforts succeed 96% of the time. But, that will never be the case. It is something we have to accept.

I’m not being pessimistic. Luck plays a huge role in innovation. No matter how much expertise you’ve accumulated along the way, the simple act of doing things differently will make that expertise irrelevant. Accept it. Make peace with it. You have to expect and plan for the possibility of failure. There is a reason why we don’t see game-changing innovations. They are rare. For example, Google is a rare company that is betting on moonshots. Even so, with that intent, pundits are already saying that Google Glass and Google’s driver-less car will fail. Yet, most of us have yet to have contact with either product. When breakthroughs are attempted, skepticism is common.

Before setting out to innovate, we must remember a few things:

What do good failures look like?

what do good failures look like?

Question-to-innovate Series: This the nineteenth of a series of weekly posts where I will answer a few common questions about innovation. Please feel free to add your own response. Also, if you have any questions you think we should discuss, let me know.

A few weeks ago I set the record straight about the relationship between innovation and failure: Failure isn’t the goal, but it is part of the process of innovation.

Don’t be afraid of Trial-and-error: How to gain evolutionary advantage

trial and error

Why wait for the need to become obvious? Yes, why wait for it? If you are an established company, you have a set of challenges that you need to wrestle with on a daily basis. Growth, is one. Either by strengthening the existing value proposition of your current business, or by creating something entirely new.

With that said, to see what isn’t there, ask yourself a pair of questions: