4
5 Comments

What pricing model do you prefer on TinyFunction (Serverless functions)?

Hi,

I am having difficulty in finding best pricing model for TinyFunction.com . I felt its better ask you.

  • Unlimited in third option is based on your container limits
on November 3, 2021
  1. 1

    Your container-limit point is exactly where this pricing decision gets tricky. I’m testing a narrow $299 one-time service for founders freezing pricing or plan structure: an evidence-backed read on three competitor pricing pages, separating FACT/INFERENCE/UNKNOWN, with a specific recommendation. Sample: https://95gk4grdkw-byte.github.io/rival-brief/sample/ — is this a decision you’re dealing with now, or would you still rather do the competitor work yourself?

  2. 1

    @prakis are you deploying a lambda now for each function or are they containerized, or its code stored in a db and is executed by a cluster of workers? I worked on a similar project last year, but users where able to create workflows with ready-to-use functions. And I had a cluster of node workers running in containers doing the execution.

    1. 1

      Currently function code is stored in db and executed by lambda workers, that is the reason it is very fast.

      But I am implementing the first option now, where each function is independently deployed. I don't want to go with containers(on AWS) as they are taking alot of time to deploy.

      I am interested to know more about your project, do you mind to share some details? my twitter id @prakis

      Thank you.

      1. 2

        Long story short, I was working with a few fintech and ecomm companies a few years ago as a contractor. They kept writing random scripts to move and transform large amounts of data around. Like ETL but with additional interchangeable scripts and logic between the E and L part. Many people on the team where not tech savvy so I decided to build a drag-drop interface for their scripts, so non-tech users can put together automations on their own.

        But...

        I was not able to make a product out if it. For these reasons:

        • I could not decide on the part that executes the workflow. I tried AWS Step Functions, lambdas, ECS/docker, RudderStack, kubernetes argo workflows, all worked, but all required a server up and running, not cheap. Even tried Spark and Storm. Kinda went all over the place.
        • Did not know who the target user is, or how to market to them.
        • Did not know enough about the ETL / data industry.
        • This is a very complex industry.
        • Tried to pivot into simple cloud functions, like what TinyFunctions is doing, but saw a ton of other products doing the same exact thing, Glitch, Netlify, Vercel, Google, Firebase, etc...

        So I dropped the project.

        1. 1

          You tried alot of things, thats very interesting. Currently I think dedicated dockers is a overkill for my project, so I am just trying to use AWS in the background.

          I made TinyFunction in early 2020, I recently posted it on ProductHunt and it made to the 3rd product of the day. Thats how I thought maybe there is real interest in such tools. I am still not sure if there is commercial interest in creating functions alone (Daily visitor count is slowly fading now ). In creating functions alone I don't see huge market, I am betting on additional features which will add great value to devs.