We recently caught up with Eric Ries, author of The Lean Startup, a book full of practical advice on how to apply metrics and scientific methods to startups. Ries maintains that many lean startup principles are also applicable to large enterprises, so we asked him some questions about how to integrate those lessons into large IT shops and the businesses IT serves.
As it turns out, even large enterprises might want to consider hypotheses tests instead of business plans, and the way to hire entrepreneurial employees might simply be to put the ones you have on an island of freedom where they're protected from existing business units. The stakes? Agility--the ability to innovate even while continuing to be responsible for maintenance and operations.
InformationWeek: Which lean startup tactics are most applicable to those who work in enterprise IT?
Ries: There is no such thing as a technique or a tactic that works all of the time, that works for everybody. These methodologies are specific to the context in which they're designed to operate. So when we talk about applying startup ideas in big companies, we don't mean to every part of the big company. We mean to the parts of the big company that are actually startups.
People find that hard to understand at first. It's a little bit of a strange idea, that a startup is not just two guys in a garage. It actually has nothing to do with what kind of garage you're in or what kind of office you're in. It doesn't matter if you eat Ramen noodles or not.
What makes you a startup is that you're trying to create something new under conditions of extreme uncertainty. So we're in a situation where we're not building something that's not fundamentally the same as what we've built before. It's something new. It's something that serves new customers that we've never served before. It's a new kind of product that's never been built before. It's taking a technology from one market and moving it to another.
So it's something where we face a lot of uncertainty. And the traditional management practices that run big companies today are not very well adapted to situations of high uncertainty. They work very well when we can make accurate plans and forecasts. And forecasting works when you have a long and stable operating history from which to make a forecast.
I often ask in my workshops, "Who feels like the world is getting more and more stable every day?" and nobody ever raises their hand because we're all facing increasing uncertainty in the work that we do. So more and more of our work, even in established companies, is becoming entrepreneurial, whether we intended it to be or not.
The specific practices that cut across company size and sector and industry are the ones that treat new products and new initiatives as experiments. And I mean experiment in the scientific sense, where we have a hypothesis, a belief about what we are going to do. And that is different from a business plan. A business plan says what we are definitely going to do in the future. A hypotheses says, here's something we think might work, if customers want the product, if they're willing to pay for it, if we can find a good marketing channel to reach them, if, if, if.
In school, if you have a 10-part test and you get nine of the 10 things right, you get an A. In entrepreneurship, if you get nine out of 10 things right, you fail. It is dead. Even if you have a great product at the right price in the right market at the time, but you can't market it to anybody, guess what? It doesn't matter--it still fails. So you have to get everything right for a startup to work. But on the other hand, the way to get everything right is not to be a perfectionist and make a detailed plan. That actually is too complicated, too difficult.
Instead, what we want to do is run experiments quickly and cheaply so that we can get the failures out of the way and discover how to actually find the right combination of product features, pricing, and marketing that will actually help us reach customers.
InformationWeek: Give us an example of a large enterprise that you've either worked with or interacted with that has had this type of startup mentality internally, where it, as you say, faced extreme uncertainly.
Ries: I'll tell you an example from Intuit, a company I've done quite a bit of work with. They have really embraced this as their management philosophy for dealing with new projects.
InformationWeek: So not just at the lower levels, but at the management levels?
Ries: The management level. And that's actually the key to creating conditions for people to be innovative. It requires a commitment from senior management to create that context. And at most companies--this is why their innovation efforts suffer--their senior leadership's not willing to do that.
One of the stories I tell in the book is a product that Intuit made called SnapTax, which is a great product. Basically, you get your W-2 at the end of the year, take a picture of it with your smartphone, and SnapTax lets you file your taxes right from your phone. It's a really cool, very disruptive technology, really different from the way tax filing was done before. It competes with one of Intuit's flashier products, TurboTax.