There’s gonna be a lot of #startups in the wake of all the recent #layoffs. The most valuable advice I can give is around really understanding that people want what you are trying to build. And that’s a long, complicated discussion. The tl;dr there is you need to validate with real users that there is actual demand for your product. Nothing else you do is more important than that, and everything you do is meaningless if you get that wrong.

BUT. That’s a long discussion. Here’s a shorter one, and it’s the 2nd-most valuable piece of advice I have.

Keep your product focused around one simple idea, and rip off everything else you can directly from other products.

Yeah, I know – this sounds like I’m telling you to just blatantly copy everything other than your core idea from other folks.

I am.

No, I’m not telling you to *ship* something that’s ripped off entirely from other sources.

Here’s the thing – you want to minimize the number of variables that you’re working with. Everything you don’t know explodes your risk exponentially. So there’s some part of your product that you can’t know – that’s the brand new thing you’re doing. Almost everything else you can likely crib from something that’s already established. And rather than innovate on it, I’m telling you to just ape it directly.

Why?

Because it minimizes risk. Let’s say you want to have a character in a side-scrolling platformer jump. Make them jump exactly like Mario. Same height, same timing, same forward distance, same air control, etc. etc. etc. Because that becomes 20 variables that you don’t need to worry about, and more importantly, you know it will feel right to your audience, because it is familiar.

Now, as you focus on your one really new idea, you’ll find that it naturally bleeds out into other things. Maybe your “new idea” is to have a game where a character can dash. That dash will affect the jump. When can it happen? How far does it travel? How does it change animations? Level layouts?

Because it interacts with jumping, it will change jumping. And as you tweak the new thing, you’ll find that you will likely also tweak the “stolen” bits. And you’ll do that enough that the starting point (Mario) will be totally unrecognizable by the end.

But if you had to tweak all the elements of dashing with all the elements of jumping and it was all unknown, you’d have a problem with hundreds of variables, none of which were set in stone when you started. This huge complexity with no constraints leads to chaos.

So start by eliminating all the variables you can by taking things that already exist and doing exactly that. Then as your new stuff comes online, experiment and evolve all the systems that you ripped off, and they’ll evolve into something new, but starting from a point where the results are known, and feel good to the end user.

By the end, you’ll have spent all your time focusing on the new thing, and by limiting complexity, you’ll actually be able to finish & ship.