Lesson: Product Leaders with Adam Nash

Step #2 Metrics: Businesses have a key metric that moves the entire business

It's amazing how many businesses you'll find where there's a key metric that moves the entire business. For example at LinkedIn, LinkedIn really depends it's entire strategy on growth. The world's largest professional network make sense because if want to find someone, you go there and you find them. Or if you want to be found, you list yourself on LinkedIn.

That doesn't work if there's no one on LinkedIn. It has to grow over time for that proposition, in fact the best business question with LinkedIn is not why it's so immensely successful today, it's more of like how did they get started? How did they get that first million? Why would you list yourself on LinkedIn if no one's looking for you? Why would you search on LinkedIn if no one's there? That's a better question I'll let Reed teach that class, he'll do a better job.

I would say that it's amazing, when I got to LinkedIn, there was actually no one responsible for growth. It was such a high priority for the company that everyone was responsible for growth and we had gone to a size in scale when you had 70 people at the company everyone cared about it, everyone followed the number, but no one thought it was their job to move that number.

When our growth slowed down in the beginning of 2008 we actually made the incredible decision, which most companies now do, of putting someone in charge of growth and actually having a team focused on growing the size. It sounds so obvious now growth hacking is incredibly popular, Facebook had a growth team, Twitter had a growth team, everyone has a growth team, but actually, at the time, it was very unusual to put someone on more of a metric than an actual feature. You own search. You own the homepage. It wasn't "You own growth." What does that even mean? We had forgotten that in the era of Web 1.0 there were actually marketing organizations who owned user acquisition. You would hire people responsible for growth, and that had given way to almost no one doing it.

This actually generalizes to a good point. If you have a key metric that's key to your strategy, don't be afraid to put people directly on it, don't be afraid – if it's that important to your business, you can't do it with everything. If there are two or three things that your business has to get right, to work, putting someone directly on it in terms of accountability is a really good idea. Advocacy works, there's reasons why legal systems work this way. It's good to have some voice in the form always thinking about, "How do we move that number forward?" Or letting people know that, "If we do this it will hurt that number."

I know it sounds simple, but as a management technique for a team it's incredibly important to have that advocacy. Otherwise what everyone does is they give lip service to it. They watch it, but they don't feel responsible and you get a little bit of a tragedy of the commons, where it gets nicked at here and there, over the years and gets worse instead of better – despite the fact that everyone agrees that it's the most important thing.

Virality. This may be pertinent for folks. I blog a lot about this. If you go to my site, you'll see I've written a whole user acquisition series about how to think through user acquisition. I really basically go through Web 1.0 understanding of acquisition, social and virality and then I extend it to mobile and native applications. So hopefully that's relevant to the stuff that you guys are working on.

This is actually a very simple insight and it's a higher level one, which is why I put here. I thought I understand virality in the '90s. I thought I understood it at eBay. I was wrong. It wasn't until we had to jump start LinkedIn's growth that I had to get the heart of with the team of what really causes virality to happen. And it turns out virality can be understood as a basic business problem.

Let's say you have a restaurant. Someone comes into your restaurant, what's the likelihood that they will come back in within a week and bring a friend? That's basically the heart of virality. There is two elements to it. One is what's the probability that they'll bring a friend, like that's a little bit of branching factor. Will they bring one friend, two friends. How often will they do that? The second part, which is the week, is a timeframe. It's a cycle time.

A lot of people spend time, when the Facebook platform launched, focused on only one side of this, which was branching factor. How do I spread to as many people as possible? Some of them worked and some of them didn't. A lot of the ones that didn't got the cycle time piece wrong, which is they spread to a lot of people but then the cycle didn't repeat. Or if it repeated, it repeated, very slowly.

It turns out that at the heart of virality, there is an exponential function, like "m" to the "n" Basic exponential math. If you have a choice between making "m" bigger or "n" bigger, which do you do?

No smart Alec saying "m" is less than one, but then "m" is bigger than one. You move "n". That's cycle time. If you reframe virality into, "How many cycles will I repeat in a given time period?" you quickly realize that if you push viral events to more frequently-used channels, you'll do better.

So at LinkedIn – a lot of you probably get emails from LinkedIn – there is a reason why. It turns out the frequency that people read email is much, much higher than the frequency that people consume the feed on LinkedIn. That wasn't true at Facebook. People check Facebook a hundred times a day. But at LinkedIn, people would check LinkedIn once or twice a week. So viral events, viral loops going through the LinkedIn feed they wouldn't materialize. It would take too long. Through email, they catalyzed quite well, which is why you see a lot of email in the LinkedIn product.

It's not because LinkedIn set out to send a lot of emails, it just turned out when they implemented the email features, they did really well by the numbers. And LinkedIn does prioritize its metrics movers.