Lessons learned using “lean start-up” within a big corp

Posted by on July 1, 2012 in Blog, Uncategorized | Comments Off on Lessons learned using “lean start-up” within a big corp

Lessons learned using “lean start-up” within a big corp

Recognising When You’re a Start-up

If you work in a big company, especially a carrier, then you’re used to the treacle-flow pace of progress. If you’re working on a new product idea, especially with a degree of uncertainty, you might want to try the so-called Lean Start-up method EVEN THOUGH YOU ARE NOT A START-UP!

I’ve set-up, run or advised several “start-ups within an established corp,” including Lucent Technologies, O2, Motorola, McLaren Applied Technologies and Art.com.

Applying “Web and start-up” patterns/methodologies to bigger companies is kind of my thing. I wrote a book about it for carriers.

According to Eric Ries, a start-up is:

A startup is a human institution designed to deliver a new product or service under conditions of extreme uncertainty.

Embracing Failure

Carriers have been building all kinds of digital products with varying degrees of failure (from 0 to 100%).

You see what I did there?

I used the f-word.

It cropped up in a conversation I was having with the CFO of O2 UK whilst consulting there in 10/11.

Here’s the slide that got him excited:


What I was trying to explain was that for certain types of product development, the so-called “Lean startup” method might be more efficient in figuring out that a product will work, or not, before spending millions on traditional product life cycles.

Carriers have a long history of spending lots of cash to bring various digital services to market that quickly tank, even when based on sound customer insights (more on that later). They key idea to grasp is understanding when your project is a “start-up” in the sense that you DON’T KNOW whether the product idea is the right idea.

What you’re really doing is experimenting. Failure is highly likely, so you need to set the project up to allow rapid iterations of the core idea until something resonates with users, whatever criteria you use to measure that (see, for example, McClure’s AARRR metrics for Web)

The key point of the “Lean startup” method is not to build a product, but to discover a product.

Getting back to the O2 story, I led a project to try the method inside the huge mothership (O2 is a BIG company). Using a small team of software developers in London, we built a service called #Blue (“Hash blue”). It was used again to build another service, called connFu.

Lessons Learned

The #Blue story is a long and exciting one. Honestly, I’ve rarely had so much fun building a product inside of a big company. And O2 isn’t just any company – it’s the UK’s best carrier by far.

Innovators like me don’t get to experiment inside of big brands that often, not unless there are a few visionaries — and renegades — running the shop, plus someone to clear all the usual corporate crap that sticks to your chaps whilst trying to sprint to the finish before keeling over from all the crud. This is my first lesson.

Lesson Number One: “Hire” a “Hurdler”

Stephen Devereux is one of my heroes. He has a long and successful career inside of O2, including helping the launch of the iPhone after O2 won the exclusive with Apple. He was the “hurdler” (sometimes also called “shit umbrella”) whose job was to line up all the green lights for the project, like those small details called money, contracts, resources, etc. He had a foot in two camps: the “mothership” and the “renegades raft.” It is a rare mindset.

Lesson Number Two: Avoid “Prima Donna Syndrome”

What I noticed about Steve’s approach was a very simple observation: it pays to talk to people in advance and to give them a reason for why you want to do things differently. For example, we wanted to “break the rules” when it came to vendor selection, so that I could recruit a small software team in London who weren’t part of the O2 supplier family.

Despite having board-level blessing for the project, Steve didn’t just demand things, like he had a right to them – he went and spoke to the various departments (like purchasing) ahead of time and explained what we were trying to do and why (giving a reason always helps when breaking convention).

If you act like you’re the bees knees with your fancy innovation project, treating everyone else like old-school oafs, then you’ll soon realise how easily the mothership can swallow and suffocate your tiny insignificant attempts to be different. You don’t have to act different to be different – remain nice to people and tell them what you’re trying to do. It’s a lot more productive.

Lesson Number Three: “Stay on Point” (Or, “Ignore the F*ckers”)

Once you head off down the lean start-up pathway inside of a large company, you can expect resistance. There are all kinds of distractions. Of course, there are the usual suspects who don’t like that you’re doing something new and different, in case it proves that the “old way” is wrong. They want to shut you down, or – worse still – make sure that you fail so that the “new way” will forever be tarnished and then banished from the kingdom. Ignore them!

However, these to-be-ignored folks aren’t the real threat. Their objections are usually out in the open. The really dangerous folks are the supporters who default back to type by trying to add in traditional “customer insight” requirements, or any requirements at all. The problem with listening to them is that you start to pivot your project around their feedback rather than real customer feedback, which is the essence of the lean start-up method.

Don’t feel tempted to please your lukewarm supporters by bending to their requests. Stay on point!

Lesson Four: Move fast towards the unexpected

It’s always easy to find an intellectual reason why something shouldn’t work. In fact, ┬ácorporate projects suffer this kind of paralysis all the time. You know the routine. Three of you head off down a path that looks fruitful when a 4th guy pops up and says: “Yeah, but that won’t work because of X.” And then a 5th person chimes in.

Before you know it, everything looks doomed. The real problem here is that you end up trying to move forward in a way that caters for all opinions. Many well-intended innovation projects die this way.

In my experience, you need to run as fast as you can to get something in the hands of users as quick as possible and get real data and real experiences to guide your path.

Remember the adventurer’s maxim:

Unintended consequences dominate.

I learned that early on from my management hero Tom Peters.

In other words, you mustn’t only expect the unexpected, but you must run as quickly as possible towards it.

Something will happen that changes your perspective altogether and then the 4th, 5th and Nth person’s opinions become moot.

Lesson Five: “Hire a skeptic”

So I said that you should ignore the skeptics, but I was kind of lying to get you rapidly down the page here (following my own advice to move fast). Thinking ahead to the end-game, if you’re running a “start-up” gig inside a bigger entity, sooner or later you’ll have to go win the hearts and minds of the mothership skeptics who didn’t really want you to succeed.

If you simply turn up and say “da-dah,” presenting the results, it’s going to be difficult for them to say “OK, you were right and we were wrong.” The way to avoid that is to recruit one of the skeptics from the start, asking for them to come along for the ride, slowly positioning them to take the credit if and when you gig becomes a success.

Aim to recruit someone lower down in the organisation, like a product manager (rather than director), who is well respected by his or her superiors and peers. The objective is for your recruit to sell the success and methodology back into the mothership. That’s not your job. Your job is to make the process work. Stay focussed on that.

Lesson Six: “Keep a safe distance”

Other than hiring the skeptic from the mothership, and then managing the other mother-daughter relationships via your “hurdler,” you must keep your team far enough from the mothership that you can push on without hinderance.

The one thing about any large group of people, and companies are no exception, is that there are as many opinions as there are mouths to utter them. Opinions take too much time and energy.

If you’re running the gig, then you must have a strong mind and a good deal of confidence in your product ideas and following the pivoting process to discover the best configuration.

A “safe distance” means keeping the main team away from mothership distractions, but not so far that you’re out of sight. If you disappear from view, that can become a problem. Worse still, if the distance has a whiff of “we’re too good to sit/work with you luddites,” understand that they will shut you down!


Most of the lessons above are related to how to apply the “start-up” mindset inside of a large corp. The so-called “Lean start-up” approach is only one method and it only applies to certain types of digital products, mostly Web-related. Other techniques I’ve tried are internal ideas markets, innovation contests, incubators (e.g. I ran the O2 Incubator experiment), “labs” and various others. There are a lot more lessons to share.

The problems are never with the methods, but with “managing” the mothership. Therefore, you have to understand that your job as an “intrapreneur” is a lot different to the classical founder role. You have to do all the “founder” stuff whilst acting as a change agent inside the mothership. It’s a tough gig, but can be very satisfying, especially when you manage to align the stars of “small company innovation” with “big company power.” The project can be fun and the results awesome.



Subscribe to Paul Golding's posts and other provocative content by email