×
Hey, I'm Seth
Each week I share the latest No Code MBA tutorials, interviews, and tool recommendations with 5000 subscribers.
I'd love for you to join as well.
  • 2 minute read
  • Practical lessons
  • Free access to content
Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.

New members: get your first 7 days of No Code MBA Unlimited for free! Click here to redeem

🎉 Holiday Sale - Get 15% off No Code MBA Unlimited ➞

00
D
00
H
00
M
00
S
Kyleigh Smith

Building the first SMS habit tracker, powered by GPT-3

Harold

Harold is the first SMS-based and GPT-3 powered personal habit building coach.

$80

Monthly Revenue

Visit Website

Built with

What's your personal background? What motivated you to start your own company?

  • I have 7+ years experience doing user research, product design, and product management for early stage startups.
  • I’ve always had my own ideas for products, but never had resources to build them. Once I developed my web development skills and no-code came around, I was suddenly in a position to start bringing them to life.

What no code tools did you use to build Harold and what purpose did each play in the final product?

  • Harold’s tech stack is basically a hodge podge of no-code tools stitched together, primarily Airtable, Twilio, Typeform, Unicorn Platform, and Stripe.
  • Autocode is the powerhouse connecting them all together (think of it as Zapier on steroids).
  • If it wasn’t for Autocode, Harold wouldn’t exist. It allows me to build all the custom logic necessary, build custom API endpoints and scripts, which allows me to get much more out of these no-code tools. If you haven’t tried it out yet, I highly recommend.
  • (Disclaimer: I’m a teeny tiny angel investor in Autocode, mostly because after using their product I realized how incredibly valuable it is)

What were the initial costs to get Harold off the ground?

  • Not really any initial costs! Mostly just the subscription cost for all these different tools.
  • Because I was bootstrapped, I also found ways to get discounts on these subscriptions. Mostly through Product Hunt’s Founders Club or discounts I got from being part of various other communities like OnDeck No Code.
  • For example I was able to get  $2k in Twilio credits and a year free of Airtable.

What was the process of building your product from idea to launching?

  • I knew I wanted to build the experience via SMS and I knew that Twilio was a good option for that. I also was curious in using Airtable as a database. So I started searching “SMS bot Twilio + Airtable” and I found this tutorial on HackerNoon that the Autocode team wrote (great acquisition strategy on their part). It walked through step by step how to build an SMS bot using Autocode to connect the tools and I was able to use that, tweak it slightly for my needs, and get the Harold prototype working.
  • Harold started texting me everyday to track my habits, and it worked great. I had never been able to stick with tracking my habits before, and with Harold I was able to track for 4 months straight. Once I realized it solved my problem and was a viable solution, I started figuring out how to allow others to use it too.

How do you attract customers?

  • Mostly by building in public on Twitter!
  • Product Hunt has also helped get some big initial spikes.
  • I’ve also been experimenting with SEO recently.

What are the biggest challenges you've overcome building Harold?

  • Driving traffic to the website
  • Figuring out the right monetization strategy
  • Not giving up

How much money is Harold making per month?

  • I’ve been running some monetization experiments which have affected the #’s recently, but its about $60-80 a month.

What's your advice for non-technical people who want to start a company?

  • Try to think of ways you can test or experiment with your idea without building something. Not being technical can actually be good because you won’t waste your time building something no one wants.
  • Tweet about your experience with the problem. Ask if other people have the problem too. (And not just once, talk about it consistently for months)
  • Can you manually create the experience that would eventually be automated (For example, to test Harold, I could have tried manually texting a few friends about their habits first before building anything)
  • Alternatively, you can try building relationships with technical people that have the same interests as you.

What are your future plans for Harold?

  • Run experiments to figure out the best acquisition and monetization strategy. Retention and Engagement is strong, so I just need to figure out these last two pieces.
  • Specifically I am starting to partner with Lewis Kallow, who is more on the marketing and content side, to help with the acquisition piece. We’re currently experimenting with SEO, such as this article we wrote about the best atomic habits habit tracker.

How do you think the rise of no code tools will impact entrepreneurship?

  • It will bring more entrepreneurs into the game for sure, which will inevitably change the game. Some things I see changing:
  • We no longer have to follow the traditional routes, like VC or only hiring people for full-time roles. I see a lot of makers working fractionally for each other to help each other grow

Is there anything else you'd like to add?

  • The best way to get into No-Code is to just start playing around! I believe you’ll start getting more and better quality ideas once you start to know what is possible.
  • You can also follow my Twitter List which has a lot of makers building with no-code: https://twitter.com/i/lists/1483508944572088323?s=20
  • If you want to learn more about my journey with Harold (and how hard it is to actually build and grow a indie SaaS, you can get a copy of my book: The Honest Guide to Indie Making. You can use this code for 20% off: NoCodeMBA

Learn to build your own apps and websites.

No code required.

Browse all courses

More Interviews

Get new interviews in your inbox, weekly.

Thanks! Please check your email :)
Oops! Something went wrong while submitting the form.