Tom Rochette

It is not enough to be agile and to measure, you must make sure your metrics guide you in the right direction. buff.ly/3zR5KBW

Sometimes it may be easier to precisely measure what you do not want, such as user dissatisfaction or unhappiness, than it is to measure what you want. buff.ly/3zR5KBW

The goal of an experiment platform is to make experimentation self-service and minimize the incremental costs of running a trustworthy experiment. buff.ly/3zR5KBW

There is a fundamental shift that happens when teams change from shipping a feature when it does not hurt key metrics, to NOT SHIPPING a feature unless it improves key metrics. buff.ly/3zR5KBW

Making controlled experiments easy to run accelerates innovation by decreasing the cost of trying new ideas. buff.ly/3zR5KBW

Twyman's law The more unusual or interesting the data, the more likely they are to have been the result of an error of one kind or another. buff.ly/3zR5KBW

In general, we recommend always having a portfolio of ideas: most should be investments in attempting to optimize "near" the current location, but a few radical ideas should be tried to see whether those jumps lead to a bigger hill. buff.ly/3zR5KBW

If you are on an experiment-driven team, get used to, at best, 70% of your work being thrown away. Build your processes accordingly. buff.ly/3zR5KBW

When running online experiments, getting numbers is easy; getting numbers you can trust is hard. buff.ly/3zR5KBW

Key themes in online experiments * It is hard to assess the value of an idea * Small changes can have a big impact * Experiments with big impact are rare * The overhead of running an experiment must be small * The overall evaluation criterion must be clear buff.ly/3zR5KBW

If humans and computers together could achieve dramatically better results than either could attain alone, what other valuable businesses could be built on this core principle? buff.ly/2CSTIy8

When you start something, the first and most crucial decision you make is whom to start it with. Founders should share a prehistory before they start a company together - otherwise they're just rolling dice. buff.ly/2CSTIy8

A startup messed up at its foundation cannot be fixed. Bad decisions made early on are very hard to correct after they are made. As a founder, your first job is to get the first things right, because you cannot build a great company on a flawed foundation. buff.ly/2CSTIy8

You should focus relentlessly on something you're good at doing, but before that you must think hard about whether it will be valuable in the future. buff.ly/2CSTIy8

Two rules for VCs * Only invest in companies that have the potential to return the value of the entire fund * Because rule number one is so restrictive, there can't be any other rules buff.ly/2CSTIy8

The biggest secret in venture capital is that the best investment in a successful fund equals or outperforms the entire rest of the fund combined. buff.ly/2CSTIy8

If you focus on diversification instead of single-minded pursuit of the very few companies that can become overwhelmingly valuable, you'll miss those rare companies in the first place. buff.ly/2CSTIy8

A company is the strangest place of all for an indefinite optimist: why should you expect your own business to succeed without a plan to make it happen? buff.ly/2CSTIy8

Indefinite optimism seems inherently unsustainable: how can the future get better if no one plans for it? buff.ly/2CSTIy8 Progress without planning is what we call "evolution".

The clearest way to make a 10x improvement is to invent something completely new. buff.ly/2CSTIy8