I Turned My Side Project into A $26K/Month SaaS Tool

David Bressler

•

Published: March 21st, 2023

saved!

David Bressler

Founder, Excelformulabot

$23K

revenue/mo

Founders

Employees

Excelformulabot

formulabot.com

from Orlando, FL, USA

started July 2022

$23,000

revenue/mo

Founders

Employees

Discover what tools David recommends to grow your business!

SendinBlue

social media

Facebook

Instagram

Twitter

Tiktok

productivity tool-icon

Google Suite

payments

Stripe

analytics

Google Analytics

design

Canva

freelance

Upwork

seo

Ahrefs

Discover what books David recommends to grow your business!

Want more updates on Excelformulabot? Check out these stories:

Hello! Who are you and what business did you start?

My name is David Bressler and I am the founder of Excelformulabot, which translates text instructions into Excel formulas within seconds with the help of AI. The website, as well as the Excel & Google Sheets add-ons, tailors towards a wide range of consumers of occupations ranging from finance to accounting to marketing. There are one billion Excel users in the world, so it’s a pretty broad audience!

While the website is free to use, users are limited to five formula requests per month. That’s enough for 97% of the users, but the remaining 3% are heavy Excel users that received unlimited formula requests, which costs $6.99/month. Since launching in September 2022, Excelformulabot monthly recurring revenue has exceeded $14K.

What's your backstory and how did you come up with the idea?

In May 2022, my wife and I had our second child - William. Paternity leave commenced! While on paternity leave away from my role as a Director of Analytics, I started dabbling in AI as I had been seeing it growing in popularity. Little did I know that the popularity of AI would grow exponentially to what it is today, driven by the popularity of ChatGPT.

My journey started when I asked the AI platform, OpenAI, a very basic Excel question - “What is the Excel formula to calculate the sum of column A.” In seconds, the formula correctly populated! I then started asking more complicated requests. As the questions got tougher, in general, I found that ‘out of the box’ accuracy was ~85%. In most cases, 85% is “good enough,” but not when it comes to data/analytics. This was my opportunity. Given my background in analytics and Excel specifically, I knew I’d be able to work on the model’s output to build a highly accurate AI model. Later that day, I Googled to see if there was an AI-based Excel formula generator and fortunately, didn’t find anything! At that moment, I knew I had something special!

With the AI industry growing rapidly, I knew I had to build quickly. There was no validation process to see if there would be a product market fit. I’ve been working in marketing analytics for 12+ years and use Excel daily. Everyone from data analysts to data scientists to CFOs and CEOs uses Excel. There are 1 billion people that use Excel, so I knew there would be a big enough market for it. At worst, I’d build a product for myself and work way faster and smarter than anyone else in spreadsheets!

Take us through the process of building the first version of your product.

I knew I wasn’t the only person in the world that would come to the realization that there was a need for an AI application that provided Excel formulas. I knew I had to build quickly, as I’m a firm believer in the first mover’s advantage - especially in the AI space.

I mentioned I’m a data guy, so I had/have minimal experience building a website, let alone a web application that has AI capabilities. I wanted to be first, so I built a barebone application that didn’t have a paywall, login, or much of anything. It was an input field where users would type their instructions, a button to generate the formula, an output field that would populate the formula, and a button to copy the formula.

I shared the website in the Excel subreddit, where users provided a tremendous amount of feedback and recommendations on the product, which was later used to develop the site as it stands today.

The post quickly went viral on the Excel subreddit, then was later shared in a few other Reddit communities (the one that gained the most traction was the post on InternetIsBeautiful), which acted as the catalyst of the social virality the site has garnered since development - going viral on TikTok, Instagram, Twitter, and Reddit.

Going viral was great, but what wasn’t great was the AI API costs that came with it. I mentioned that I didn’t have a paywall or even a login. I got hit with a hefty bill from OpenAI, so I knew I needed to hunker down and build a real website so I wouldn’t be tens of thousands of dollars in the hole.

The site received a sale within the first five minutes of launch. I get the chills just thinking about it because it was such an incredible feeling knowing that all of that hard work wasn’t for nothing.

I knew it would take several months to start making money on the site, so I set up a donation page to allow users to “tip” me for developing the service. I received a couple of thousand dollars in tips from a couple of hundred users, which was absolutely amazing!

Describe the process of launching the business.

It took several months of being heads down to grow from the basic MVP to a fully fledged subscription-based website that now has a paywall and log in!

The day the new site launched with login and checkout for users to receive unlimited formula requests, I thought - “Will people actually buy this?”

The site received a sale within the first five minutes of launch. I get the chills just thinking about it because it was such an incredible feeling knowing that all of that hard work wasn’t for nothing.

I mentioned I’m not a web developer. The site was built in a no-code web application called Bubble.io. There is not one line of code that I wrote on the entire website. With that, I didn’t have to outsource the development of the website. As a result, I was only paying for the API requests, web hosting, and Bubble.io.

If you’re not familiar with Bubble.io, I highly recommend checking it out. It’s a web application that allows non-technical folks like myself to build web applications. I was first introduced to Bubble in July 2022, where I was able to spin up the rough-looking, single-page MVP in just a couple of weeks.

No-code web applications like Bubble.io have been overshadowed by the AI boom. I’m always surprised that more people don’t know about it. The tool makes it incredibly easy to build your own SaaS and whenever I’d get stuck, I’d find the resolution on the active forums or via YouTube tutorial.

Managing costs was and still is, more important to me compared to other entrepreneurs building startups. I didn’t receive any funding. No loads. I’ve been bootstrapping the entire business, dipping into my accounts until the site became profitable. Because I didn’t have any other backing, I grew my startup while maintaining my full-time job to continue to support my wife and two kids, especially since my wife is staying home for the first year of our infant’s life.

Within the first week, I launched the site on ProductHunt.com, a popular website to bolster the awareness of your official product launch. With a heavy discount to celebrate the launch, the site generated several thousand dollars during the week of the launch. This was the reaffirmation I needed to keep going…

Since launch, what has worked to attract and retain customers?

After seven months, the website has eclipsed $14K in monthly recurring revenue and has increased in revenue each month, reaching $22K last month.

To date, I’ve spent only $5K on marketing, which sounds crazy considering I’ve had millions of people visit the website. But, it’s true. Having a product people want to share is a big part of how I’ve been able to attract customers, but an even bigger part - is influencer marketing.

As a data guy who has always been a performance marketer and result-driven, I always shied away from influencer marketing because it’s difficult to measure and quantify its impact. So much of the early traction came from TikTok influencers talking about the site as a “website that feels illegal to know”. I found that influencers are ‘monkey see, monkey do’. If a well-known influencer with large following talks about something, there’s a swarm of micro-influencers that will talk about the same thing a day later.

I tested out several influencer marketplaces but didn’t have any success since they mostly focus on retail. If you’re looking for an idea, build an influencer marketplace for SaaS. It doesn’t exist!

My strategy has been to directly reach out to influencers where my product would resonate well with their audience. Sometimes, they’re open to promoting it for free and sometimes it comes at a price (typically ranging from $200 - $3K).

I use TikTok’s Creator Marketplace to identify potential influencers. The marketplace is free and allows me to filter for keywords/audience affinities, followers, and engagement. One thing I’ve learned is that follow count does NOT equate to engagement. I specifically look for influencers whose audience is in the US, UK, or Canada, which are the top 3 origin markets of paying customers. Additionally, I try to spread out the audience demographics across influencers, ie: accounting, finance, data/analytics, etc.

Influencers just focus on video platforms like TikTok and Instagram. I wanted to ensure I was diversifying my social reach, so my strategy on Twitter was to engage in conversations with people talking about Excel. Additionally, I set up alerts to be notified when Excel influencers like ‘Excel Dictionary’ or the official Microsoft Excel accounts tweeted. I made sure I was always the first to comment, so other users would see my post and link to the website.

Because I more or less had other people promoting the site for me, I was able to focus on retention. I built out customer journeys via website promotions and automated emails with multiple touchpoints to deliver personalized messaging. Messaging varies based on the types of questions users ask in the generator, how many days out since they created their account, their level of engagement, if they are paid or free users if they registered with a business email versus a personal email, as well as several other factors.

Email remarketing gets a decent amount of traffic, but so much of the traffic is driven by word of mouth through colleagues, friends, and social media influencers. Organic search (largely made up of people Googling “excel formula bot”) and direct traffic make up the vast majority of site traffic, which speaks to the value of having a strong brand and a brand name that is easy to remember.

SEO has been a new focal point, as I’ve been building out content pages around people searching for specific formulas like “VLOOKUP” and “SUMIFS”. I’m hopeful that the site will rank well organically for these types of searches, as the site has garnered many high-quality backlinks from sites like pcworld.com, inc.com, wellandgood.com, and BGR.com.

While I never did any hands-on SEO at my full-time job, SEO has always been on my team, so I’m very familiar with the optimization levers. There are a lot of levers out there, but early on, my focus was simple - incorporating popular search terms in the SEO tags and copy throughout the site, while the referral links kept coming in (not needing a link-building strategy).

Something interesting happened from an SEO perspective. Because I was the first AI Excel site and the brand name is also the category (“Excel formula bot”), using Google Search Console, I started seeing a lot of searches for general searches like “Excel AI generator” and “excel formula creator”. These searches didn’t exist before my site, so I knew they were looking for my site - but didn’t know the exact name! These keywords, along with many other general searches, are now included in the SEO title and meta descriptions, in addition to other sections of the site!

How are you doing today and what does the future look like?

After several months, the website has eclipsed $14K in monthly recurring revenue and has increased in revenue each month, reaching $23K last month. Despite starting well in the hole, the site is now profitable and continues to grow. The gross profit has increased each month with increased subscribers, optimized costs, and diversified revenue streams by partnering with complementary affiliates - Excel courses website and Excel templates website.

Because SEO is not a channel that has overnight success, I started investing in PPC for people searching for Excel formulas or Excel courses. With a $10 cost per lead, I’m not breaking even when accounting for the lifetime value and lead-to-subscribe rate, but I continue to invest. Why? Because 1+1=3. When someone learns about the site, they tell someone else. And that person tells someone else. So, that $10 cost per lead indirectly turns into a $3 cost per lead.

In the coming months, I’ll be releasing additional generators and focusing on B2B outreach. The website skews young, due to the nature of social media, where most users hear about it. As a result, I plan to tap into an older audience - focusing on business decision-makers, to drive more team subscriptions, which only make up 2% of the subscriber base.

For the time being, I plan to continue to be a solo founder but may look to hire someone eventually when/if I start to struggle with bandwidth, especially someone that would complement my skills (ie: someone to help drive B2B sales).

Through starting the business, have you learned anything particularly helpful or advantageous?

It’s OK to be a copycat when it comes to web design. Don’t start from scratch. Building a SaaS? Visit a handful of successful SaaS websites to get inspiration on how they designed their site. They’ve probably done a lot of UX testing to get it to where it stands today, so you can capitalize on that and take learnings from a few different websites to build something special for yours.

This is a huge hack, but was a big reason for my site’s success: over time, gradually raise the “barrier to usage” of your product. What does that mean? When I built the website, it was available for free with unlimited usage and no login required. Then, I required a login. Then, I capped the usage. Then, I built the subscription site as it stands today. I gave people a taste of what it felt like to have unlimited formula requests and when I put up a paywall, they quickly purchased subscriptions because they had become so reliant on it. You can test this out with a free trial period.

I’m a firm believer in giving users the ability to fully experience everything your product has to offer, as opposed to the freemium model that limits certain features. How will your users ever know if they need those features without letting them try them first?

What platform/tools do you use for your business?

Web application: Bubble.io

Payment processor: Stripe

Language translation: LocalizeJS

Email management: Sendinblue

AI: OpenAI

Video editor: Screen Studio

YouTube, to learn how to use all the tools above

What have been the most influential books, podcasts, or other resources?

I’m a huge Scott Galloway fan. I listen to a couple of his podcasts weekly - Prof G Show and Pivot with Kara Swisher. Strangely, he’s incredibly relatable considering how successful his career has been. His enthusiasm for tech, innovation, and economics is unparalleled. He’s someone I look up to - a proud father, who has built wildly successful businesses and now educates and shapes young minds to build a better future. You can’t beat that.

Advice for other entrepreneurs who want to get started or are just starting out?

Building websites today isn’t what it used to be. You used to have to know HTML, CSS, and design. Not anymore. There are tools like Bubble.io and Webflow for anyone that has an idea but doesn’t know how to code. As I mentioned, I built my website without a single line of code. Those applications have been around now for years, so there are much more documentation and video tutorials readily available for people just getting started.

Another word of advice - don’t get hung up on trackability when it comes to media measurement. Of course, measure where you can (Meta & Google Ads), but don’t completely write off tactics just because they’re hard to measure. I can’t tell you which influencer drove the most signups to my site, but I can tell you that influencer marketing works.