How Tamir Bootstrapped a $3,500 MRR SaaS Product from Scratch Using No-Code Shortcuts
How Tamir Leveraged No-Code Tools to Fix What Zendesk Overlooked π
Hello π
Iβm so excited to bring you this inspiring story this week. I talked with Tamir who runs a Zendesk app agency.
Read about:
π Leveraging knowledge of Zendesk to identify untapped opportunities
π οΈ The tools that helped Tamir build a $3,500 MRR business
π How solving gaps in Zendeskβs offerings gave Tamir a significant edge
π£οΈ How offering free integrations sparked word-of-mouth marketing
Enjoy and happy building!
Spotlight β¨
π΅οΈββοΈ Customer research tool for Reddit
Reddit is a goldmine for startups, and GummySearch shows you where the gold is.
The first AI-powered research tool for Reddit helps you discover business ideas, understand your customers, and get public sentiment on any topic.
π 4 Cool Finds
No-Code Playbook is your guide to go from idea to your first product with AI and No-Code.
GummySearch helps you ideate & validate businesses with Reddit. Quickly find painful problems to solve and solutions people are asking to be built.
No-Code Bits - is a newsletter with daily bit size snippets of no-code tools, resources, and deals.
Reflio - Create an affiliate program for your SaaS in minutes, $0/month
π₯ No-Code Founder Interview
Learn every week from a real world no-code success story
π Share your own No-Code Success Story to be featured
Hello! What's your background?
My name is Tamir Bashkin and I fell into No-Code by way of a happy accident. Back in college, I dropped out of a CS degree which taught me enough to be able to read code.
But I knew I didnβt want to become a developer. This led me to start a Zendesk implementation agency, where my superpower was creating No-Code integrations.
After working on 100+ Zendesk implementations, I saw room for creating automated SaaS products. I began to do this within the platform for some of my clients and other clients on the platform.
This is what initially got me curious about No-Code tools that I could use. My clients needed quick solutions that would take Zendesk several years to create, test, and eventually launch.
Everyone around me discouraged me, but I knew this was a niche that hardly anyone was serving at the time.Β
Tell us about your product that you made?
Given my experience with Zendesk automation, I knew the Zendesk app marketplace well. I saw the marketplace as a fantastic way to learn more about customer pain points too.
One pain point that my clients were repeatedly asking help with was needing a quick and automated way to close tickets. Specifically, they struggled with tickets being reopened after an agent had already solved them. This occurred when a customer responded with a thank you message.
This led me to launch a new product called Thank You GPT.
Thank You GPT uses AI to identify thank you messages from customers and it automatically takes them out of the agentβs queue. Itβs a small task, sure, but thousands of them? That's a lot of time and costs companies want to automate.
I was blown away. After adding a paywall, I sent an email out to my customers and said: βPay if you like it or Iβll need to shut it down.β To my delight, I had 20 customers who paid on the spot.Β
Which No-Code tools did you use to build and run it?Β
I used the first version of Make.com, called Integromat - when it was still a niche product. This still serves as the backbone of the app, and all the logic runs through there.
Iβve also used akeyless.io as a vault to store customer API keys. This was required from the beginning since we have to access their Zendesk account.
Once the product started to pick up, I implemented segment.io for storing analytics data, metabase for dashboards, and Bubble.io as the UI of the app.
Iβm using Microsoft Azure and OpenAI to power the AI portion of the app, since itβs more secure. This allows me to cover the AI expenses via the Microsoft startups accelerator benefits, which remove a big operational cost for me. I believe that by the time the credits expire, AI usage will be a commodity so the credits help me get to that point in time.
π οΈ No-Code Tools
Make - for the logic and automations
Akeyless - for API keys
Bubble - for UI of the app
Microsoft Azure - for AI
OpenAI - for AI
What went into building the initial product?
The first version was created in a few weeks and $700 investment. I worked with a Zendesk app developer to create the app based on the requirements of one of my implementation customers.
He was responsible for creating the app and the front-end and I was responsible for the back-end side. This meant that I processed the messages and solved the Zendesk tickets if it was a thank you message.Β
We got published on the app marketplace within 3-4 weeks after we started developing the product. Overtime, it amassed ~500 customers that installed for free at that time.
That version used Regex to decide if a message is a thank you message or not. This meant that it was useful for some companies but it wasnβt solving a lot of tickets on the whole. Once we implemented AI, that changed and we were able to start charging for the usage.
What was your business model and how have you grown your revenue?
I have a lean MVP approach, and I am a huge proponent of bootstrapping. I started by charging $50 for the usage, and $250 for Enterprise companies that wanted me to sign an NDA.
We went from $0 to $1,000Β in revenue by launching AI processing and putting a paywall on the product.
The second increase in revenue was when we built a UI that allowed companies to configure the product themselves. This mean they didnβt have to reach out to us and it resulted in less churn. Customers were able to see the results of the plugin in real time and have more control over it.
How have you attracted users and grown your product?
Currently, the majority of our customers come from organic discovery in the Zendesk apps marketplace. Weβve been experimenting with paid ads on Google that had some success, but we were struggling to track the results properly.
Can you share some stats?
π° $3,500 MRR
~ 40 active customers.
~ 80 more customers in freemium
200 visits each month on the Zendesk marketplace page, out of which weβre getting 5-10 installs per month.
What's your advice for No-Code Makers who are just starting out or want to sell their business?Β
I believe the most important thing when you start out is finding some form of validation. If youβre a No-Code founder thatβs working on a CRM plugin, you can search the feature requests forums for ideas.
If youβre not working on top of an existing platform, looking at creating a βlow-costβ version of a VC backed product can be a validation point in itself.Β
The easiest mistake to make is to jump into building and creating too fast. First, you need to ensure youβve got a customer whoβs interested in your product.
Thanks for sharing your story Tamir!
You can find more about him on LinkedIn
π€ What are you working on?
Whether itβs pre-revenue, pre-build, or just an idea, weβd love to hear about it and re-post it on X and get you more π
π Tell us what youβre working on
π When youβre ready, hereβs how we can help
Kickstart your no-code journey with $100 off the No-Code x AI Bootcamp
Browse through the full library of No-Code Success Stories
Learn which No-Code Tool is the best for your idea
Find Online Business Ideas for customers on auto-pilot
π Partner with us
Sponsor No-Code Exits and get in front of 13,300+ subscribers
Love this story! It perfectly illustrates something I keep saying about digital entrepreneurship - sometimes the best opportunities come from spotting tiny gaps in big platforms.
I've seen this firsthand building various digital products (some worked, many failed, all taught me something). The real magic isn't in building something completely new, but in making existing tools work better together.
Think about it - Tamir didn't try to compete with Zendesk. Instead, he found a small but annoying problem (those pesky thank you messages) and fixed it. That's the kind of practical innovation that actually works in 2025.
This really resonates with my own journey experimenting with AI tools. Just recently, I built a simple meeting timer app (https://thoughts.jock.pl/p/magenda-simple-meeting-timer-that-keeps-you-on-track) - not because the world needed another timer, but because existing ones didn't quite solve the specific problem I had.
Three key lessons that really hit home:
- Start small but solve a real problem
- Use no-code tools to validate quickly
- Let customers guide your growth
Side note: His approach to AI costs using Azure credits? Brilliant! It's exactly the kind of practical thinking we need more of in tech D:
Hey, are you working on any side projects that solve a specific pain point? Would love to hear about them!