One idea that's worked well for my clients has been to create a Skunk Works, which is effectively an autonomous, outside-the-bureaucracy, kinda-sorta-secret R&D team.
I've been on these teams before, and if they're truly left alone with a pile of ideas and resources, they can accomplish incredible things. (This is how Lockheed Martin developed some of its most incredible tech, and how Google takes its "moon shots".)
But it will be a challenge to management, because they'll have to stay out of it after the initial specs are delivered. If you can get buy-in, though, you'll bring back startup-level development speeds to your organization.
I'd be happy to discuss some of the processes I've followed and the structure of the teams I've been part of — just drop me a line.
Here's the critical factor: Willingness to fail. And with that comes Lack of blame.
Corporate cultures often have a "play it safe" mentality and a "blame game" process. Why would anyone want to risk their career?
A culture in which risks are encouraged and failure does not mean blame and firing is necessary.
Establish the team. Get the right people on board--people who already have a risk-taking personality. Set up resources for this team that you are willing to lose entirely, and provided the project was not truly mismanaged, no one will be blamed for losing.
One of the top issues with corporate execs and success or failure is their willingness or lack thereof to give total commitment to an outcome--"We're going to do this no matter what it takes." They need to see that from you. If they see you trying to play it safe, they're not going to stick their necks out.
We deal with this issue at The Economist and in the past couple years, we've been able to implement some lean thinking to move things along faster.
At the core, what you're trying to change is the culture.
Culture has 4 components that a group of people (people in your org) share:
1. Value system (what they consider important)
2. Belief system (how the believe things work)
3. Behavior (habits, rituals)
4. Artifacts (manifestation of the above)
You need to start addressing the first 2, then you'll see the progress (or lack thereof) in the last 2.
Here's an example: a bad approval process that makes product development move really slow. We discovered this happen because the stakeholders want to show that they're contributing to the project. Because of this, they want to make sure they get to review the project and give feedback. This way, they can show their contribution for the product's success and this would be good for their performance review.
From this example, the value for these stakeholders is recognition. Their belief : giving feedback (even if it's superficial and immaterial) shows that they're contributing. The behavior: blocking process in order to get a chance to give feedback. The artifact: feedback for the product in email, request to do reviews, meeting invites.
How to start:
1. Find your allies. These are people who also want to move faster. Ideally people who work in your team
2. Show real value. The reason why you want to work fast like a startup is so you can deliver value to your customers faster and more often. What can you do to give other people around you (who you want to influence) a taste of what that's like? One thing that we did was to do rapid prototyping (without code) and do usability test, then shared the video highlights with the stakeholders.
I'll be happy to share with you the way we did this at The Economist mobile team and offer some actionable ideas to help you move forward with the transformation.