Hey API devs, how do you decide what API to create?

As a part of my weekly SaaS Newsletter, I cover a lot of ideas for developers too and covered about APIs, AI/ML products too. I would suggest you to give it a read.

But to answer your question, my pointers.

• See the trends on IH, PH launches etc.
• API of APIs is not bad. Many API products created wrappers for AWS, Google Sheet, Gmail APIs
• NoCode API tools is a big market too.
• Look at RapidAPI
• Many APIs are coming around OpenAI, ChatGPT, Image creations as well.

Good comments, but some of them smell like ChatGPT... might be wrong!

Hey @olschafer, when deciding what API to build, it's important to consider what problem you're trying to solve and whether there's a market for it. Start by identifying a pain point that you or others are experiencing, and think about how an API could help solve that problem. You should also research existing APIs to see if there are any gaps in the market or ways to improve upon existing solutions.

To determine whether an idea for an API is worth implementing, you can conduct market research, talk to potential customers, and create a MVP to test the idea. This will help you gauge interest and get feedback on your idea before investing significant time and resources into building a fully-fledged API.

As for my own API, I previously developed a service that blocks temporary and fake emails from signing up on websites. It helps prevent spam and improve user engagement by ensuring that only legitimate email addresses are used. If you're interested in this type of service, you can check it from the link below.

https://www.indiehackers.com/post/say-goodbye-to-fake-users-block-temporary-and-fake-email-addresses-aa42d4d879

Developers use APIs because it saves them time integrating and maintaining the underlying functionality, and businesses buy APIs because it helps them launch products to market faster, focus on their product's core value, and scale usage rapidly with low maintenance overhead.

A few questions to think through: How much time would it take a developer to build the functionality your API provides if they were to build it themselves? What complexity is your API reducing? How expensive would it be to run and maintain the API? If other APIs exist, what problems do they have, and what functionality is missing that you could add?

For developer focused products, it's less about building an "API" and more about helping developers feel like they have super powers. Help them build something faster and cheaper than if they were to do it on their own.

Just look at Vercel- they have a highly regarded DX and lots of developer love. They simplified common problems developers had deploying products and packaged it in a really easy to use product. Same goes for Supabase.

I would say it's less about the "API" aspect, and more about understanding your audience and their core problems. For you, that sounds like understanding problems developers face and seeing if you can build something to solve it.

When i need to add special infrastructure that is not part of my main value proposition and i wish the api existed so that i could concentrate on my main task. For example sending confirmation emails, getting product feedback from user, site/event analytics etc.

That is more of a business decision and should be evaluated based on existing competeitors, your value proposition, distribution channel to reach your customers etc.

So I first built an API for FInancial data and that was useless. There is simply no traffic

how do you decide what API to create

This is very broad. It's like asking "what business to make?". The API solves a specific problem, the same as a business. First think of a problem, and then see if it could be solved with an API.

Your question is quite vague. By worth implementing, do you mean, a sellable API service?