Recap
The status while writing the previous post was:
- Landing page + Create Account button
- Did a little bit of marketing, reddit worked and I had to do more marketing
- MVP being built, but aiming to a very primitive implementation
MVP is ready 🚀
The idea was to build a very basic MVP where the users would upload the data I need to deploy their services (Docker Compose file, URL and size of the deployment) but I quickly saw that users (the first two that created an account) are not willing to test the platformlike this. They (I would too) expect a minimum product that works automatically and that I can test live.
With that in mind, I developed an MVP that allows the users to easily create and deploy a Docker Compose file. Together with that, I created a public Github repo with 3 templates that the users can simply copy-paste to quickly test the platform. And that’s pretty much it. I cannot bill yet (I still need to register in Spain) and there is a lot of logic missing in the backend. I did nevertheless developed some checks on the Docker Compose file to ensure they don’t exceed the resource limits of the deployment, and some logic over the exposed ports (I probably shouldn’t have done this yet, as it is problem in my mind but still never happened to a real customer…).
Looks like more than enough for now until someone is willing to pay for the service.
Marketing time (stop developing!)
Devs nature and imaginary problems
By nature, the developer tries to solve problems, the harder the better, totally ignoring if those problems must be solved, and when they have to be solved. I’m pretty sure this is one of the biggest reasons why developers fail to sell their products, becase it’s against their (our) nature to stop developing and start selling. But if you don’t sell, you have no customers, and without customers, which problems are you fixing?.
Now that I have a working MVP I know that I have to start selling and stop fixing imaginary problems.
If I get some customers, then I can get feedback and see what they need next. If don’t get customers, well, time to think about alternatives.
Spamming internet and refining target customers
The first marketing action for Docker Deploy showed that between Reddit, X and dev.to, the best place to promote this product for free was Reddit: it drove the most traffic by far (+2 created accounts), with more or less the same effort.
With that info, I just googled what my possible customers would google, but showing results just in reddit. Queries like:
Cheap Docker Compose deploy site:www.reddit.com
Cheap Docker hosting site:www.reddit.com
But the results were meh or the same posts I used last time. Then I realized, a lot of people might be deploying specific things with docker deploy, and not just generically searching for cheap docker compose deploy. So what is people starting with Docker Compose? For example: Wordpress, Postgres (with admin ui), Redis (with admin UI) or Elastic (with Kibana). So the new queries where to spamm internet with my product were:
Cheap Postgres with pgAdmin hosting site:www.reddit.com
Cheap Redis hosting site:www.reddit.com
And so on…
I used some of the fitting results to promote the site and the next day I got two people creating and account AND a deloyment 😁. I take that as a win.
I will continue to promote the site and gather feedback from the users I get. From there I should get what is the next step.
Missing feature?
There is one missing feature that I strongly think I will have to add in the future and I think it will be a showstopper for some prospect customers: the ability of deploying private docker images. There are some off the shelf products like Wordpress that you can deploy and use with public docker images, but I suspect a lot of people want to deploy their app together with a DB/Cache/Proxy.
This seems like a very obvious feature to add, but the thing is nobody told me they need it yet. So I might be able get some customers without adding this feature, or maybe I will never add this feature. In any case, not adding this feature is a big excercise towards solving problems that I do have, like not having customers. Let’s see in the future if this is ever implemented.