Should innovation units be physically separate from the core business?

A Corporate Startup Perspective

Most senior executives can relate. It usually starts with some startup types within the company telling them scary stories about how startups are coming to eat their lunch. Look at Facebook, Uber, Twitter and Airbnb! Oh, look what happened to Kodak , don’t let that happen here. Oh, look what happened to Blockbuster, don’t let that happen here. Look boss, we need to innovate like startups.

Most senior executives agree. The trends are self-evident like a slap in the face. Only executives with their heads deep in the sand are ignoring such signals. But what should they do?

Inside or Outside

The answer to this question is where it gets interesting. You see, depending on who you talk to, you will get conflicting advice. Should innovation units be physically separated from the core business, or can innovation be managed internally. This is a sensible discussion to have. However, much of the advice you get is filled with polemics.

Each side takes an extreme view. One commonly cited problem is that traditional managers with MBAs are too stuck in their ways understand innovation. These managers are also incentivised to behave in a manner that stifles innovation. You can’t get anything done within a company that expects a 30 paged business case before it funds any idea.

True indeed. But I have argued elsewhere that it is equally true that setting up seperate innovation labs will not guarantee that any successful products emerge from there. For the most part it is innovation theatre. Look-a-here! We are doing lean startup, design thinking, customer development, business model canvas, minimum viable products… Sure. Whatever.

As usual polemical debates fail to get at the core of why innovation in large companies succeeds or fails. Our corporate startup definition of innovation describes it as the creation of new products and services, that deliver value to customers, in a manner that is supported by a sustainable and profitable business model. This is the job of innovators. If your Innovation Lab does not do this, calling it an Innovation Lab is a misnomer. Better to call it: The Lets Hangout and Work on Cool Stuff Lab.

Why Innovation In Large Company Fails

Internal based innovation fails when the large company decides to use the same processes it uses to manage it core products to manage its innovation projects. Business planning does not work for innovation. All estimates of ROI, NPV and ARR are fiction. Furthermore, this approach encourages managers to only ever invest in sure bets. This leads the company to keep working on the same things it has always worked on.

External based innovation fails when the large company decides to not build any management processes around its innovation labs allowing them to work on whatever they want. The lab teams work on cool ideas, but very few of these ideas have validated business models. And very few of those good ideas are aligned to the company’s strategic vision.

I have experienced successful innovations with great business models that wither on the vine because there are no managers in the company willing to pick up the products and take them to scale. These products become orphans that are eventually abandoned; creating a discouraging and uninspiring environment for future innovators. This adoption-for-scale challenge is often ignored even though it sits right at the center of the debate concerning internal versus external innovation.

The truth is that no matter what you do or where you start, innovative products will always need someone from the main business to make a decision about their future. How those people view the new innovations will ultimate determine those products’ mortality rates.

Whatever Shall We Do?!

The only indisputable answer to this debate is that innovation should be managed via different processes to those that are used to manage core products. This clarity is provided by Steve Blank’s distinction between searching and executing. How this distinction is instantiated depends on the company, how much management buy in you have and the innovators’ appetite for corporate politics:

Full Frontal Assault: With this approach, innovators tackle the hard questions upfront. They ensure that they get top level executive and middle manager buy-in. This air-cover will help in future situations when there is need for support and resources. This buy-in can be support for an external innovation lab or an internal innovation process. Regardless of the type of approach the executives endorse, strategic alignment is key. Innovators will also need to do the hard work of changing and adapting the company’s capabilities to ensure that they fully support the chosen innovation approach.

  • For internal based innovation, you will need to create an investment and product development framework that recognises the distinction between searching and execution. This framework should make clear how investments are made at each stage, how teams move from stage to stage and how product teams are expected to develop products during the search versus execute stages. The tools necessary for this work should also be provided to product teams. Steve Blank’s Investment Readiness Levels and Ash Maurya’s Running Lean framework provide good guides for this. Another great framework is Pearson’s Product Lifecycle which I played a small role in developing.
  • For external based innovation, you also need an investment and product development framework that ensures that the product teams in the labs are doing the hard work of validating their business models. The frameworks listed above work pretty well. In addition to this, you need to build a bridge between the innovation labs and the core business to ensure that the ideas coming out of your lab don’t end up being orphans. In describing ambidextrous organizations, O’Reilly and Tushman argue that innovation business units should get leadership representation at the executive board level. This allows ambidextrous CEO’s to be able to manage the trade-offs that often arise between innovation labs and the core business units.

Guerrilla Movement: Sometimes it is very clear that you will never get full executive endorsement for innovation. The executive are focused on cash cow products and the best you can hope for is support from a handful of visionary leaders within your business. In this case, innovators might consider leaving the company for greener pastures. Alternatively, you can start a guerrilla movement. Tristan Kromer, who is a great innovation ecosystem designer, has two main recommendations on how other ecosystem designers could manage such a movement.

  • First, Tristan suggests that innovation ecosystem designers should lower the costs of innovation. If they do this successfully, then they will hardly ever need high level budget approvals. Lean Startup methods provide a great tool box for lowering the cost of innovation. As do other tools like Optimizely which allows for cheap and fast AB testing, Heroku which allows fast deployment of applications with no server maintenance and low costs, or Usertesting.com which eliminates the cost of building a usability lab. These low cost tools can be used to navigate your innovative business model towards product-market fit.
  • But even in a guerrilla movement, every now and again innovators will need to surface within the company in order to get investment for their ideas to be taken to scale. They may also need some resources and support on their journeys towards product-market fit. For this, Tristan Kromer recommends that teams find diplomats who are individuals who will do the hard work of corporate politics and smoothing the path for innovation projects. A diplomat is usually someone who is well connected and respected in the business who can work outside of normal bureaucratic channels to call in favors and get things done.

Internal Versus External Is A Flawed Narrative

The above distinctions are meant to to help innovators to be conscious about the approaches they are taking within their company. Each approach has its own merits and flaws. I certainly have a preference for the full frontal assault because I like to know that I am working with full support from the executives. I find guerrilla movements too hard. Regardless of the approach you take, you have to acknowledge that someone important in the main business will at some point catch wind of your idea and have to make a decision. It is how, when and where this happens that is different.

Even within the Lean Enterprise approach proposed by Trevor Owens and Obie Fernandez, the innovation colony is setup by the executives in the main business. They make the decision to put money into the colony, and also to incentivise employees by giving them shares. The executives also decide later whether to scale any idea as a separate division, spin it in to the core business or sell it. The innovation colony concept is a version of managing innovation differently from core products. But without a full frontal assault and executive endorsement it is nigh impossible to setup such a colony.

So to me the answer is clear. Innovation has to be managed differently. But whether or not you do it within the company, in an innovation lab or an innovation colony depends on context, culture, resources and capabilities. I am quite averse to one-size fits all declarations. You can find examples of successful internal innovation systems as well as broken ones. Guerrilla tactics sometimes work, but they also have a high mortality rate for ideas. Pick your poison! You can also find examples of successful innovation labs, and badly managed ones. What really matters is that the company thinks through the implications of its choices with a clear view of what represents true innovation and how the work it is doing aligns with its strategic vision.