I’ve stumbled upon the short video of Paul Graham, co-founder of Y Combinator. He is answering the biggest question of software founder has — when should you launch your startup?

I like his answer a lot because it reflects what I’m doing and what I’m advising my clients so much. I’ll talk about this in a moment, but first, let me tell you why this is an important thing to know form the very beginning.

Why thinking about launch early on?

Knowing when to launch sets the goal. The goal is your expectation and your criteria for success. It’s something measurable, and it makes you think about the outcome before doing anything.

The truth is that when you start thinking about the launch, you may realize that you don’t need an app at all. Do you know what is the biggest struggle almost any client I’ve worked with has? Customer acquisition. Have you thought about how you are going to get users to use your app? Because if you didn’t, you should, before even starting the development process.

I was approached by the potential client a while ago, John (not his real name). When I asked him how he is going to acquire customers, he told me that this is something a Chief Marketing Officer will figure out once his startup will get Series A investment round. He wasn’t expecting a development agency to ask questions like that.

But we do because we work with venture capitalists, and we know what they are looking for in startups. No one is going to give you money if you have a product and don’t have customer acquisition figured out. Seed round — yes, but not Series A. You need to start marketing before you start building a product, which is why you might not need an app at all at this point.

So, when to launch?

Paul gave an excellent answer. You should do it when your launch makes someone happy because they can do something they couldn’t do before. The more people you can satisfy while pissing off the least amount possible, the better.

But that doesn’t mean you need an app for that. Maybe you can make people happy by providing a service first. Or perhaps you can create an informational product.

I know, not something you expect to hear from the developer. And probably not something any startup school will tell you. But if you are bootstrapping it might be a fantastic way to build your audience, learn about your people and save a lot of money on development. There’s a multitude of ways to make people happy by solving their problems. Check Epic Guide to Bootstrapping a SaaS Startup from Scratch — By Clifford Oravec to learn more about bootstrapping.

And if you are sure that you need an app, consider using something like concierge MVP — fake most of your app and build only the customer-facing frontend. I’ll write more about this and other techniques that allow you to launch earlier later.

Risk of launching late

There’s a huge benefit of running a software development agency. Over the last five years, we’ve built twice as many apps for our clients that I’ve made myself as a developer. After such accelerated learning, I can tell from the very beginning when there will be a feature creep. Ever heard of it? It’s when the development process extends endlessly because new features are added to the scope.

You might think that it’s not a problem for an agency, but it is. Not only because the success of a client is a success for us, but because no one likes building things just for the sake of it. Feature creep exhausts the team and demotivates developers. It burns the client’s budget quickly and puts the whole business at risk. And guess what happens when the company runs out of money.

Founders often think about the risk of launching late because of competition. In my experience, this is rarely the case — in fact, it wasn’t ever a problem for an early-stage startup in my practice. Chances are your biggest competitor will be yourself, postponing the launch, and trying to perfect everything.

Risk of launching early

The opposite of feature creep is launching a subpar product that nobody wants. When you are in the lean startup mindset, you are starting to cut all the fat from your product spec to the extent when there’s nothing left but the most basic functionality.

Heck, we have this problem too. We want to serve our clients so much that we started building our own products. If you are an agency, you need to do this yourself — it will completely change your perspective.

Guess what, the first product we built internally (I’ll share it later) wasn’t perfect. It doesn’t have a unique idea, it’s interface wasn’t fancy, it got bugs and problems in UX. Not perfect at all. But the worst thing was that it wasn’t good enough to show people. And those who we showed it to haven’t started using it — the quality was ok for internal use, but not for the public.

The expectations today are high. We are using a multitude of apps that set the bar. Can you build an app that has smooth user experience as behemoths like Airbnb or Uber? Can you create a search experience that is as useful as Google’s? Probably not, at least not at the early stage. But that’s what you are going to be compared to.

P.S. Check out Startup School by Y Combinator. They have outstanding learning materials and will give you $15k in funding if you are eligible.

Originally published at The Startup: Build something awesome