EndTests
End the hassle of e2e and UI tests
I just sent my first cold outreach message for EndTests.com to a QA Engineer on LinkedIn.
This is a big deal for me because I rarely do this, I'm not sure what to say, and I don't even know if I should text QA Engineers, QA Managers, or CTOs?
Let's see. It's all about experimentation.
I just sent my first cold out-reach message for EndTests.com to a QA Engineer on LinkedIn.
This is a big deal for me because I rarely do this, I'm not sure what to say, and I don't even know if I should text QA Engineers, QA Managers, or CTOs?
Let's see. It's all about experimentation.
I now have a working product. It's far from complete, has many rough edges and in all likelihood, some bugs.
This early stage version of the product is just a seedling. I have big aspirations for this for the next several years, but I need commercial validation. I want to work with people who actually want what I have built and will be building. My early customers also get to help me steer my product in a way that can benefit them and others like them.
I made a commitment, declared on my twitter, to start marketing and selling my SaaS on 31st of May 2023.
From that point, my daily tasks for EndTests.com will be mostly business tasks with only a little amount of technical items.
This way, I force myself as a software engineer to do what's necessary to build a business as opposed to keep building and building, something I just LOVE to do.
Now is the time to build more than just a software system, but an actual business.
A few weeks ago I changed the approach with EndTests in a major way.
EndTests started as a web page in which you put commands that control a browser. This showed that things are indeed possible.
I then started thinking of how I'll be using it and what will bring value in terms of ease of use and time saving. Afterall, EndTests is intended for professionals and businesses with web products that they want to ensure works continuously well.
So I started building an "Exploration" feature. The intention of this feature was to go through a system trying to explore what it does, and the end result would give the user lists of commands that achieve whatever things it finds possible to do. All of this was procedural.
Then comes a few weeks ago, after I actually finished building the exploration feature, which I was actually happy with, but required more work to get it to a usable level. For example, when I started considering more sophisticated steps for exploration, I started think about AI.
I decide to give it a shot. I started building and rebuilding projects, models, services, html processors, integrations with 3rd party tools, and all that comes with software development.
I took tons of notes, gathered information, and worked on the algorithms. Experimented with various approaches. I started designing algorithms and experimented with a few approaches.
I shared progress on Twitter, I shared moments of progress and moments of stagnation. Whichever way it was, letting people know I'm working, and knowing that behind my tweets is actual work, I felt motivated for more as I held myself accountable in front of people, to get things done.
It was tiring hard work but it was extremely fulfilling. Each step of the way, I was filled with doubt, excitement, determination... It's just an amazing combination of feelings that makes being an Indie Hacker so awesome.
All of this culminated, just today, with what I consider a major achievement for me personally and for EndTests. I've done something I've never done before and now push EndTests to new frontiers and myself professionally to new capabilities as a software developer.
I am still on the technical side of things, which is alarming for someone who wants to build a business, but I filled with determination and excitement that makes me trust that I will indeed transition my focus on business-building tasks soon.
That will be another world for me, and this time I'm going to focus on business results and just getting it done.
I wanted to push my product out to people as soon as possible. So I super quickly hacked login/registration to allow people to start using the app. Authentication on the site lacked basic features; it didn't even have the ability to do a password reset.
My rationale was that this isn't something core to my product. Yes, not even authentication is important to me. All I care about is showing a product feature that pertains to how the system that I'm building will serve you and help you achieve the task of automating UI tests.
My product is in the early stages, and I want user feedback as soon as possible so that the community, my audience, people who care at all to give feedback, guide the future development of the product with their suggestions and feature requests.
As I continued to post about EndTests.com, someone simply advised me that I will want to reach out to people who sign up to use my product. So it might be useful for me to verify the emails of the user's who sign up and not just have abc@example.com sign up.
So I decided to fix up auth. All I wanted to do is enable verification of emails, but this led to that and I ended up with a list of technical tasks to deal with. I was using #YesCode (manually coding everything with my hands), and with some tasks being tricky, I ended up spending quite some time on this but I finally got this done and working today:
Now my application has the ability to verify emails. It uses SendGrid to send a verification link that verifies the email of the user who signed up. At least now I know I can reach out to customers, ask questions, communicate with them about concerns, features they'd one and let them know about product updates.
This I realized now, is actually a business task. For a while, I thought I was spending too much time doing things that are just technical, but the purpose was purely business-building. I need to be able to talk to my customers! What was I thinking.
Now that that is out of the way, I'll shift my focus again on finishing up the "Exploration" feature, which allows EndTests.com to "explore" a site and make a map of possible actions for testing.
-
3
When I look at it from this perspective, authentication and email verification really play a crucial role because you need to be able to communicate and interact with your users. I'm really glad that you were able to overcome this challenge.
-
3
Thanks @dogukantezcan!
Entrepreneurs need to be wary not to spend too much time focusing on the "technology" side of things.
Sometimes technology looks like business when it's not! But that's not every time.
-
2
Was there a reason you didn't go with something like Cognito, Auth0, or Authress?
-
1
I like to remain in control of the important aspects of my product, while still using industry standards.
I use the tools provided to me by the framework I'm using, and perhaps very well known libraries, but not cloud services for this.
So basically, not cloud, and not from scratch either...
-
1
That's surprising to me. What would it take you to be convinced otherwise?
-
-
-
-
-
1
This is why my start-up, www.outseta.com, exists. We provide all the tablestakes features like this required to launch a SaaS app out of the box so you can focus on building core product.
-
1
Cool! Good luck!
-
Since people showed some interest in the automation of e2e and UI tests, I decided having people actually "see" what I have in mind by building up an example, would help direct discussions and help people imagine the possibilities.
I am going to build this focusing on the product features almost exclusively (which means that the UI will look really embarrassing), but these are the important things that will drive the value of the product, these are the things that I want to show and discuss with people.
So I started building the proof of concept, which is an app that allows a user to submit commands to a backend that controls a browser, effectively allowing the user to use the app to automate browsing, i.e. UI testing.
I've called it QAAS locally which is Quality Assurance As A Service. I would have immediately picked up that domain if it was available alas it was not. So I had to come up with some other idea, perhaps something more creative.
EndTests.com: ending the hassle of e2e tests.
It's about automating and managing the process of testing your app by interacting with its UI as if you were the user.
The result is that everything gets tested, from the UI on the frontend, all the way thru the web app and its backend services to the database.
I wonder now about 2 things:
- I use e2e tests in my current job, but do other people use it?
- What do people imagine a solution that helps with this to look like?
I posted some of my thoughts and asked questions on Reddit, IndieHackers, and Twitter.
Got some interactions and into really interesting conversations. Some people alerted me to things I didn't think of, as well as things that were on my mind but too early to start considering now. So much value in letting people know what you are up to.
I found interesting people being interested and that encouraged me to start this project.
11 Comments
Try to reach out positions that actually make the decisions, usually C-Level or Managers.
Even if your solution if for the engineering team, tailor your pitch around productivity or cost reduction. Most of the time those are a default scorecards they have to deliver to pocket their bonuses.
You are absolutely right. I realized this after a few messages and started to send to QA Managers.
Although a couple of conversations with two of the engineers I sent to were useful, learned a few things and about some products in the field.
I need to emphasize the time and cost savings indeed! Do I say this in my first message? In the messages I sent to QA Managers, I mentioned that I'm experimenting with new QA techniques, but didn't mention the product itself. I am trying to make it sound like a normal conversation
Try something like this "hi ____ I wonder if you have a few minutes in your calendar. I work as a QA Engineer for the past X amount of time and I'm trying to validate if the path I'm chasing to reduce costs and increase productivy is right. I would also love to show you what I'm working on, I think there is a opportunity here to help you and your team."
I would cold message something like this.
Also try to do consultive sales, that means that you will hear more about their problems and see how your software would fit their needs, instead of trying to teach them how to use what you wrote.
This kind of sales takes more time but usually is the type of sales that build trust around you and not your "brand".
Also those are the ones that will teach you most about the low hanging fruits on the niche you are trying to work on.
Engineering are clueless about top level decisions and what goals are. Most of the time they are discussing things about performance and overengerring stuff that should be discussed way down the line.
Once someone told me "if you are not a sales person, you are working for one." Have that in mind as well.
Also be clear on your first message what you are chasing. People can see through it most of the time. Try to respect the time someone is giving when they stop to read or hear what you have to say. You don't have to do the dance with everyone
I'm not really a QA Engineer, that's not my main work. I'm a full stack developer who worked on upgrading my company's UI Test framework and started to think I can build something that any company can use (intellectual property is not questioned/no legal issues).
I love the ideas that you present here and will be using the script that you suggest. Here is what I just sent a bunch of QA Managers. I would appreciate any feedback you might have on it:
Here's the image I attach: https://pasteboard.co/Yu16UiLlR71E.png
You could assume the pain of your target just don't mention it. I would remove this part "You're always looking for ways to make building tests easier and simpler to save time and effort"
This way will not sound like an ad and will sound more like a cry for help. And people do actually want to help others, if there is no strings attached.
You should add a script to do a demo for those leads. The demo could be a short clip of you interacting with the solution, there is no need to voice over it. Only if you have something that you need to explain. And if you do a 1-1 video clip with voice over, remember to say the leads name during it. The name of the person is the most important word you can use the grab it's attention.
Yeah, that part did indeed feel a bit odd. Will remove it.
The idea of creating a demo with voice over where the script is personalized to the user is a very good idea. Yes, before your message I would lean towards making it generic so I save time on recording, but I agree, it would be better for the connection!
I am still trying to get the idea validated though. People aren't even replying.
It's ok, sometimes that happens, and I'm ready to move on if things go south, but I really want to confirm whether my product is worth it or not...
About the idea validation step, just don't iterate over the product if there is no one paying. If you are going to build something to scratch that dev itchy. Build stuff that helps you to launch new things. For example, payment process, user auth, and document your sales journey. If your original ideia don't go through you could cut corners with all those stuff to launch a new thing.
Also, you could validate the idea yourself, get a QA freelance and try to use your own solution to speed things up. If that helps you, that is another way to make money indirectly with what you built.
My friend and I were looking at something similar but did a pivot since I'm not a QA and am passionate about something else but we wanted to use some AI to help us automate e2e testing.
But, I know that we at the company are using mostly Cypress and Playwright at this point. And on every project that I worked on, there were problems in maintaining and writing them at first.
A lot of service companies are selling end-to-end testing to clients, but if you can help them speed that up they would certainly pay for it since you are saving them money. You can probably try to reach out to service companies, and their QA managers or even some higher roles.
In smaller companies, higher roles are doing a lot of things. But your approach with cold outreach looks correct, and it's a great way to go!
This is great feedback that I find pretty encouraging. I appreciate it a lot.
Good to hear some validation. Positioning myself and my product is vital at this stage. I don't replace some testing "suites" that are out there, but I feel that I can indeed provide some value. I'm even willing to do some items manually or give concierge treatment to companies that adopt my product at this stage.
So the service companies that you mention, are these QA Testing companies, or Software Dev companies that provide e2e testing services to their clients? Should I be searching for QA Companies maybe? Or maybe some CTO/CEO of small-sized software development company? 🤔
Well, these service companies can be both, most of the software development service companies which are doing outsourcing have QA and Engineers. There are less companies that are doing only QA.
And some smaller startups might be interested since all of them want to save money now. Going for some big customers can be exhausting since sales cycle for them would be long one
Yeah, makes sense, time for LinkedIn again!
Thanks!