- This is where most startups fail. 80% of all startups never find a product market fit and die. Today we're gonna look at the five steps to a product market fit. You get this right, and you might just make it. (bright upbeat music)
- This is Raw Startup. (air whooshing) - I'm the founder of the Vivino. We found a product market fit so strong that we're up to 50 million users now. Every day, two to 3 million wines are looked up on the Vivino. That's what I call a product market fit, but it was not easy to find. We'll do a step-by-step guide to how to get a product market fit. We will looking detailed
to a perfect release cycle, that'll take you to a
solid product market fit. Finally, we look at how to
identify the product market fit, how do we actually know that
we have a product market fit? This is where most startups fail or win, so let's go and find that
magical product market fit. Finding the product market
fit for Vivino was not easy. It took 18 months of hard work and grind. 18 months of little progress and flattened numbers. Once you find it though, it's pretty awesome, good. Let's take a look at the five
steps to a product market fit. Number one, find problem
solution and audience. We start by asking ourselves
what is the problem and what is the solution? If you can state that in a
really, really simple way so that most people understand it, you've probably understood it pretty well. Here's an example, problem, "I don't know what wines are good and what wines are not so good". Solution, build an app that tells the user what wines are good and
what wines are not so good? If you have a hard time defining it, you probably wanna narrow it down and define it a little bit sharper. Next thing is to find an
audience you can talk to, an audience that has the problem and wants the solution. If you're a part of that audience, that's very helpful. If the problem is not relevant to you, you have to find some people you can talk to about
the problem and solution. People that have the problem and want the solution. There's no shortcut here. You have to understand
the problem and solution and find that audience and define that audience, so you can talk to them. and the audience can't be everyone, that never works. And please don't be
that founder that thinks he's such a genius that
he can invent problem and solution in his own head, without talking to anybody. That happens all the time. Products come out that nobody wants. Before we move on, we
wanna test the problem and the solution. Do that by looking at frequency and value. How often is user gonna use the product and how much value does it give the user? In a perfect world, the user meets the product all the time and it solves a big, big problem. In the real world, that's rarely the case. But you do need either
frequency or value to be high. If you have a product that's low frequency and low value, users just forget about it. They use it too infrequently, and it really doesn't add any value, so they forget about it. Let's take a few example. Let's start about Facebook. Frequency is high as
users use it every day or every other day. However value is actually low because it isn't like
it solves a big problem, it's just a bit of entertainment. Now, let's take a product like Uber. Most people don't use Uber every day, so you could say frequencies medium. But when it comes to value, when you really need a ride home, value is actually high. Finally, let's take a look
at a product like Zillow, where you buy and sell real estate. People don't buy houses very often, so frequency is definitely really low but value on the other
hand is very, very high. However, if we changed the audience to actually be real estate agents that sell homes all the time, you can change the frequency too high, and now you have high on both of them. The point is, you can change the audience and suddenly you have
high on both of them. So define the problem
solution fit and audience. Make sure either frequency
or value are high, maybe even both of them. Number two, find a way to start building. We all have limited resources and we need to find a
way to start building with the limited resources we have. And the answer to that is, "Please someone give me $20 million and I can start building" wrong answer. You have to start building
with the resources you have, or have a really good plan on how you're gonna start building. If you can't do that, you have to go back to number one and find a solution that you
can actually start building on. When you do this, ask
yourself questions like this, is the probably even solvable? What can we build with
the resources we have? Is the technology available? Can it even be built? How can you find the
customers for this product? So basically find a way to start building. You can't have a solution where you don't know
how to start building. Even if you want to build
a rocket to go to the moon, you probably don't start with that rocket. You start with a smaller rocket to show people what you can build. Number three, build an MVP. Built a minimum viable product, you probably saw this coming. It's in the world. Build the smallest possible
thing that's still viable. Something that has some value
to a few early adopters. You have to be really disciplined here. You must build the
smallest possible thing. No fancy features here. Here's some quick tips
when building an MVP. Define core features and stick to it. Don't increase scope, just build the smallest possible thing. Build your MVP fast. Make sure you build it fast, because if you don't do
that things will drift. You will start adding features. This is an MVP, just get it
done as spec and that's it. Get the MVP in front of
customers as quickly as possible. Make sure these are
real users or customers, not your friend. Find the people that are most desperate for the exact product, hopefully someone is
screaming for your product. Don't know any customers who it? Well, should you really
be building something for a user or customer you don't know? Probably not. Forget your friends, they're not really gonna use it. They will give you a poor feedback. They will love the product
because they love you. That just isn't there for
when you're building an MVP. Friends and family are what
keep dead products alive. You have to find real potential
users for the product. Get feedback from a small
group relevant group. There's no point and have a
big group as you test an MVP. It's all about having a
small, relevant group. Having too many people testing
will just creating noise and too much less usable feedback. This will be a very basic and not very pretty product. You need feedback from
people that really understand that this is an MVP. We don't need to know
that the icons are ugly, they're meant to be ugly.
(fingers snap) Now, that's how you build an MVP. Number four, find the metrics for success. These are the numbers
that show you progress on your product market fit journey. Let's talk about what
you wanna avoid here, vanity metrics. Vanity metrics are
numbers that you follow, but don't really get you
closer to a product market fit. Numbers that sound and look great, but don't really get you any closer. An example of a vanity
metric is downloads. Maybe it goes from a thousand to 10,000, that doesn't mean that
people use your product or that your product is any good. It could just be an
article in the newspaper or some celebrity mentioned your product. It doesn't mean the product is any good, it doesn't mean that anybody's using it. We need to find the real
numbers that truly tell you, are you getting closer
to a product market fit? I also wanna keep it simple, I don't wanna track 10 numbers, so I wanna boil it down to two numbers, two numbers that tell you you're getting closer
to a product market fit. The two numbers I wanna monitor are retention and usage. Retention is about finding out if people keep using the product, there's no point in making a product that people use once and then throw away. To get actual growth, you need retention. It doesn't have to be a fancy number, actually active users can
tell you the retention. At the Vivino we used active users, MAU was the big number, monthly active users. In the early days we actually also used weekly active users. So find a great number for retention, it could be active users. If you get a lot of new users, be a little bit careful
with using active users because that can take
your number up and down. Sometimes people exclude the
new users and active users, that can give you a better
picture of the actual retention. But the second number is usage. How much do people use the
product when they use it? For Netflix, it could be
minutes to watch per month. For our web shop, it
could be orders per year. For Vivino, it was scans per month. Basically, usage is how much
do our users use the product. So find a number for retention and find a number for usage, those are the two number to tell us if we're getting closer
to a product market fit. I have a full video about numbers you should track in a startup, so what's that sometime later. Some people might argue
that you wanna know how many new users come
in at the MVP stage. I do think that's less important. It's more important to get a few users in every week or every month. Number five, build a fast and insanely awesome release cycle. Yes, I said insanely awesome. And I mean it, because
this is the key to success. You have to build an insanely
awesome release cycle, nothing less than that. This is about improving the product fast. We'll talk about speed in a second, but let's first look at
what's in the release cycle. So here comes the release cycle. It's seven steps which sounds like a lot and they're all important, but most of them are really fast, so it's all pretty simple. Before you start, make
sure you enjoy this. This can be a lot of fun and you will miss it later. Building a product fast, releasing fast is just a lot of fun. It's a grind and it's tough, but very, very enjoyable. At some point later you'll probably miss how quickly you could
release in the early days. There are other things you won't miss, but you will miss the speed, so enjoy it. Okay, here (chuckles) we go. First step, brainstorm. Look at the feedback, throw all the ideas on a whiteboard. Listen to the team, listen to the users, and get all the ideas up there. This is not several days, this is one hour in front of a whiteboard. Make sure you engage the entire team, so they feel every idea is listened to. You can't build every idea, but you can listen to every idea. Keep them coming, keep them in there, and the team members that came in the idea will feel appreciated even
though it isn't built. Next is shirt size. Not all ideas are equal, some IBS take a lot of work and others are an easy fix. Shirts sizing means estimating how much work it is to build a feature. Keep it simple, use three levels, small, medium, large. And make sure you actually use the scale. Not everything is large, then you're not really using the scale. Next, you prioritize and decide. Make sure you know exactly what metric you're trying to improve, and keep asking yourself, will this feature improve that metric? Now, take the brainstorm
and the shirt sizes and decide what you wanna build. Next, is to spec it. No, you can't skip it, you have to do proper specification on what you're building, so the team on a sense
exactly what's being built. Again, no skipping this
step, it's important. Now, to the fun part, build it. Hopefully, this is where
you spend most of your time actually building the feature. Don't let feature creep in, stick to the specifications. Next, release it. Get it out there and tell all your users. Finally, we have the post-mortem. How was the feature received? Do the users use it? Did it actually prove the
numbers we wanted it to? In the early days of Vivino, we did a weekly release cycle, we released every Friday. Today, it's a little bit slower, we really try and keep it fast. Now, to the most important
thing with the cycle, the speed. do it as fast as you can. The faster you can learn from your users and build that and go again, the more likely you are to succeed. That's it with the cycle, keep going, and running through it as fast as you can. In a second, we'll talk about how to see the product market fit. But remember, the ones that can build a great product the
fastest will probably win, and the key is in that release cycle. Now, that's the full list, but the crucial next step is finding out if you have a product market fit or not, so let's look into that. There are hints of a product market fit, for instance, if users keep
using a flawed product, that can be a sign of product market fit. Let's say the product is
a little bit of broken, but the users keep using it. It means they like and want the product so much, they can live with the bugs. An example was early days Instagram, they were known to crash very often and the service went down, but the user stayed with Instagram because they loved Instagram. We had the same thing to the Vivino. The product really wasn't that good. It wasn't very good at its core feature, but people kept using it
believing that the bugs and flaws would be fixed over time, which obviously it did. Now, let's go to a pretty certain way to identify product market fit. More and more users use
the product more and more. This means going back to the
users we were tracking earlier, retention and usage. This basically means that
both numbers are going up, more and more users are using
the product more and more. For Vivino, that meant
more and more active users scanned more and more wines. We had a clear indication
of product market fit. One thing you have to be careful with is if you have a lot of new users come in, they can distort your active user numbers, and it looks like you have
a lot of active users. New users that use our
product for a short time, that's not a product market fit. Finally on identifying product market fit, I will tell story from earlier. So when my wife was pregnant
with our first baby, we went to the doctor and said, "Hey, how do we know when
she's going into labor?" The doctor was short and precise and said, "When it's there, you know it". It's the same thing
with product market fit. Once it's there, you know it. If you're in doubt if you
have a product market fit, you probably don't have it. I tell you when we saw at Vivino, we had no doubt at all, this was a true product market fit. If you want more content
just like this one, please consider pressing
Like and Subscribe, so you get all the future videos. Thank you very much for watching, now stop watching, and go build something. (bright upbeat music)
