I have been involved with and founded a number of startups. I founded Finland’s first VoIP operator in 2003, we got up to 500 enterprise customers before we sold the company five years later.
I was the lead Data Scientist at Pryte, a startup acquired by Facebook in 2014.
I was the head of Data Science at Siroop. The failed massive bet on E-commerce by Swisscom and Coop.
I continue to be involved with many startups, including my own windsor.ai and also some more Enterprise type ventures.
I have seen some spectacular successes and also failures.
I thought I would write down some of the most visible mistakes I see being made again and again. I will mostly focus on the Enterprise side of startups and Venturing because there are some repeating patterns of mistakes. By avoiding them the chances of success would be significantly increased. However they might not be so easily repairable due to the structures and dynamics in established corporations. Which is what gives real startups a chance 😉
I am super happy to see that so many of the large Enterprises in Switzerland are so active in the startup sector and get involved in hackathons and startups and new technologies.
However, when I try to think about real success stories from Corporate Ventures it is not so easy to come up with so many. Nestle’s Nespresso is one of the big successes, but its difficult to think of others. Let me know in comments if you know more and I am somehow just missing them!
The things I usually see go wrong are.
Idea and purpose
Corporate ventures are many times setup because the board or a CEO of a large enterprise likes an idea or says that the company needs to do something in a space.
Examples of this can be
- A large retailer sees its being threatened by E-commerce so they start a startup in E-commerce.
- A Telco is collecting large amounts of data so they want to monetise it and set up a startup to do it
Both of these examples are actually happening all over the world today.
The problem here is that the Enterprise starts a startup because the Enterprise wants to do something. Yes, from corporate strategy point it makes sense. But for the startup it is the wrong setup. The startup would need to make something people want. The startup would need to start from the users problems and work backwards from there. That way they would build something people want and people would be ready to pay for.
If the startup is built on the premise that the telco needs to collect more data it is very unlikely that this would lead to something people would want. They are not mutually exclusive but very unlikely to happen together. It is so hard to build something people want so combining two different purposes are unlikely to lead to success.
Same with the example of the retailer wanting to get involved in E-commerce. Yes from corporate strategy point of view it makes perfect sense. But unless they start from a customer and identify a pain that they could solve better than the existing providers it cannot lead to success. In addition to this they would have to be able to solve that pain. Really solve it, not just make slides about how it is solved. I will go more into this in the point about people.
In most cases of successful startups they do not do what they set out to do at the start. For example Slack was making games for five years until they noticed they had an internal chat tool they liked much more than the games. When a startup is started because a corporate board has ordered the startup into motion it is impossible to make these adjustments.
People
In Corporate Ventures in many cases the CEO and management positions in the startups go to persons who have been in the Enterprise for some time. Most likely he has also been a champion of the idea in one way or another. Sometimes the enterprise takes an outsider saying they want someone from the outside, but in most cases they will pick someone like them. This sounds good, sensible and plausible. This is why this is so counterintuitive and difficult to change.
The people who have been in enterprises for some time and succeeded there to some degree are very different from the persons required to start startups and make them succeed. In enterprises to succeed you have to be good at different things than in a startup. There is always politics involved and honesty can sometimes be frowned upon etc.
In a startup these things do not really matter. What matters is figuring out what people want and then building it. This usually requires being brutally honest. Especially before product market fit the sooner one admits failure the faster one can move on to the next attempt at solving the problem the startup set out to solve.
In enterprises if you admit failure you are usually stepped on by your peers who angle for the same promotions as you. So this particular point about being brutally honest about product market fit is a big issue. People who succeed in enterprises are not wired for these environments.
Even for founders without the pressure of VC’s its very easy to believe one has product market fit before it actually is there.
For Y-Combinator it is actually a red flag if a person has been in middle management in an enterprise. The only defining factor they found in successful startup founders was that they like to work on interesting problems, which does not happen in middle management in enterprises.
Y-combinator is probably the one who has done most research on this topic in the world.
Incentives
Continuing on Y-combinators research, they came to the conclusion that most successful startups are successful because the founders made them successful. Not here important but nonetheless its worth noting is that this has nothing to do with the idea the startup was about, as that usually change.
Peter Thiel compared starting a startup to climbing up a hill while the rest of the world is throwing rocks at you. The amount of problems and pressures one is facing is just immense. Even when you have success the problems just get bigger because then you actually have something and the stakes and pressure grows.
In Enterprise Ventures its many times better for the CEO’s and management teams career in the short run to convince the board the company is doing well even though they are not. They get a good salary and are usually not tied to the long-term success of the startup. In addition there is the tendency to look positively on what one is doing which easily sets one up for self-deceit regarding product market fit, go to market fit and many other things.
In enterprise startups the board is usually consisting of enterprise persons. If the board, does not know which hard questions to ask the dynamics between the management and the board become such that it becomes difficult to correct course before it is too late. The management might present the progress in a very different light than things actually are as that is their incentive and the board does not necessarily know which questions to ask.