4 Lessons learned trying and failing to build a SaaS startup

I tried and failed to build a SaaS startup.

Now that I got that out of the way, what went wrong?

I fell victim to the classic fatal flaw that plagues the 90% of startups that also fail: no customers.

It’s not for lack of trying. I’m sure I could have been better at cold emailing, networking, or advertising however even with what I was able to do, I didn’t have much to show for it.

To add insult to injury, I’m a product manager by trade and ship products quite effectively – I should know a thing or two about selling a SaaS product.

While I work as a product manager by day, on nights and weekends I had been trying to get an SEO SaaS company off the ground. Or at the very least, get it somewhere. I’ve been working on this for 3+ years and here are 4 lessons I wish I knew on day one.

1) SaaS is the big leagues, start smaller if this is your first venture

I really wish I knew this when I started out.

Going straight to SaaS with no prior startup experience is like going from beer league softball to a major league tryout.

It’s not that building a product is hard. Once you learn to code, building a product is quite easy. Turning that product into a business that can support you and a team and continues to grow year over year is an entirely different ball game.

Nowadays it seems that everyone is building a SaaS platform.

From IndieHackers to Product Hunt and everywhere in between you do not have to go far to see another simpler, cheaper, faster, better way to do x. I was tempted by watching other people be successful (or feign success) and thought “wow that looks easy, I can do that too.”

This is not a critique on SaaS. One of the greatest things about our current environment is that it is getting easier and cheaper to build a SaaS product, creating more competition and chances for people around the globe to build a successful business.

But SaaS is the big leagues.

Building a SaaS product is not just a matter of knowing Python or React.

The bar is being raised higher and higher. People expect a certain caliber of design, interactivity, and pizazz that, 10 years ago, just wasn’t the case. Not only that but building a SaaS company is so much more than building the product.

Your platform has to be impeccably designed and have a good onboarding flow with the right PQLs to upgrade users. You need a marketing and sales strategy (beyond posting to Product Hunt and Reddit) along with a go to market strategy. You need to ruthlessly prioritize features and feedback, pursue growth at all costs, and either do the nights and weekends thing to slowly bootstrap or get funding, which is a full time job in of itself.

Are there exceptions? Sure, but they are rare.

In short, to bootstrap a SaaS company (or grow one until you can get funding) you need to be a marketer, sales rep, product manager all wrapped in one – throw in developer and designer if you don’t have a cofounder.

This is not impossible. There are people who build SaaS startups but it’s hard, especially if SaaS is the first venture you try to start.

I tried to build an SEO SaaS product and I got immediately bogged down. Not just in the code, but in everything that goes into building a polished product that people happily pay for.

After 2-3 years of slogging through I ended up stuck at $150 MRR and all I could think was “Gee, if I had $100K-$200k to throw at this problem and hire people this would be a lot easier”.

An easier way to get that kind of money is to build a business that is easier to manage and sell it. About that long slog to $150MRR, that brings me to my next point.

2) You don’t have anything until people whip out their credit cards asking to buy

When you try to sell a SaaS product you will get a lot of feedback. Most of it will be positive, especially if that feedback comes from your friends and family.

How do I know that?

Because most people are nice (there are exceptions) and they want to please you (again, there are exceptions). Most people will happily get on the phone or a zoom, listen to your pitch, and say something to the effect of “wow that is neat, yeah sure keep me in the loop” which roughly translates to “I’d never actually pay for this but I want this guy to let me get off the call”.

I had several conversations that ended like this, and I’d happily run to my friends, family, and team exuberant that we had “another customer” only to find out that it’s not the right time, they need to review their budget, or some other reason they didn’t want to actually pay.

This lesson took quite a while to learn before it finally sunk in. You don’t have anything until people whip out their credit cards asking to buy. Until someone asks you when they can use your product and how they can buy it, you have a product but you do not have a business.

The best book you can read to speed up learning this lesson is The Mom Test which shows how to ask good questions and how to read between the lines of what people are telling you. It took me an embarrassingly long time to purchase and read this book, so my advice is this: if you’ve also been putting it off don’t. Buy it now – that link is not an affiliate, I honestly think you should read it.

Speaking of which – talk to far more people than you think you need to.

3) Talk to 50 people to make sure you understand the problem, and then talk to 50 more just to be sure

Talking to your friends or family does not count as customer research. Those people are going to try the hardest too please you. Unless you have brutally honest friends and family, in that case keep them around.

You need to talk to people in your target market.

Identify who you are building this product for as quickly as possible, and find those people – a boat load of them.
Aim to talk to 50 people to properly validate your product. Get a sense of what problems they have, what their workflow is, how they spend their day, and how much they (or their company) would pay to have their problems solved.

Do you absolutely need to talk to 50 people? No, but having a number that high helps keep you focused. If the results are inconclusive, talk to 50 more. Keep going until you hear nothing new, and the data points start to form together to tell a story.

That story is the basis of your product (or service). That story is the problem that people will pay to have solved. All you have going into these conversations is a set of assumptions. I made the mistake of thinking my assumptions were enough.

They weren’t.

Talk to as many people as you can to have a firm grasp of the problem, otherwise you risk making the mistake that led to the next lesson I learned.

4) SaaS is not always the answer

SaaS is absolutely the answer to a problem that is solved by SaaS.

SaaS is not the answer to every problem though.

Like I said above, SaaS is a very hard business to start with. But even if you have the resources and time, it is not always the answer to every problem.

Like they say about someone with a hammer, to a product manager every problem can be solved with SaaS. I set out to build a simple SEO platform. There are already dozens of those and more popping up all the time, and for good reason – all I hear from people is how damn hard SEO is.

I misunderstood that to mean that the tools people use to do SEO are complicated – and that complexity as well as the price of the tools is the problem.

Jumping off that misunderstanding was the idea that if I could just build a tool that was easier than Moz/SemRush/Ahrefs then I’d strike gold. My eyes were filled with dollar signs every time someone said “yeah if you could make an easier tool to use that’d be great!”

What I missed what what the actual problem was.

I built one of those platforms and tried to sell. it. A handful of people paid for it, but most dropped off after trying it for free. Almost all said they dropped off because they did not know what to do with it. My mistake was to double down. To try and design a better UX. Redo the onboarding over and over again. Come up with a different color scheme that was “friendlier”.

I dug my heels in rather than listen, and it wasn’t until I actually listened that I realized it’s not that the tools are overcomplicated (they are complicated, but that’s a separate issue), it’s that to be effective at SEO you need more than a slick tool, you need a strategy.

I had spent so much time trying to perfect a platform that no one knew how to use that I missed out on the why. Moz, SemRush, Ahrefs, and fill in the blank SaaS all have pretty much the same data. If you pull a list of 1000 keywords, or a list of backlinks, or run a website scan you’re going to get pretty much the same thing from each platform. Data alone won’t give you success in SEO. Good SEOs know how to connect the dots to build an SEO strategy, however I was not selling to good SEOs, I was selling to marketers and founders new to SEO.

That led to them having lots of data and no idea what to do with it. It’s not wonder the platform wasn’t the hit I thought it’d be.

Meanwhile I was driving value for clients I consulted for (which made 90% of the revenue) and it clicked. People can pull a list of keywords pretty easily no matter the tool. It’s that people are not confident making decisions on SEO and knowing what direction to go.

They struggle with strategy.

I could add value by guiding those people in my target market to build a strategy. Once I began selling that – an SEO strategy, not an SEO app – credit cards started coming out.

Realizing that SaaS isn’t (always) the answer opened up doors for other products I could sell.

Product like:

  • One:One SEO coaching
  • SEO strategy group coaching
  • SEO templates, ebooks, etc.

I thank my (very patient) fiancee for helping me see the light here, and the Startups for the rest of us podcast for showing the different flavors of bootstrapped businesses out there.

Once I got out of the SaaS mindset I was able to see other opportunities for my side project to make money, and it started making more money than I ever could have hoped the previous two years.

I’m now doing around $1.5k-$2k MRR. It's not much, but the work manageable and it's growing. Plus it's scalable. I can do it while still working my day job and build up testimonials, case studies, and expertise to attract more customers.

The best thing? I’m not chasing features or reading yet another article about UX and onboarding hoping it’s the silver bullet. I’m creating and selling value. That’s it.

The bottom line

If I could sum up all these lessons into a quick soundbite it’d be this:

  • Talk to people
  • Make sure you are solving a problem they are willing to spend money on
  • Be flexible in your solution

That really is it. The biggest mistake I made was going in solution first and trying to find a problem and customer to match it to. Had I taken the time to talk to people and actually listen, and be flexible in my approach, I may be further along than I am now. Then again, this is all part of the journey. These lessons we learn shape us and make us better founders, so while I wish I had learned these lessons earlier, I’m glad I actually took them to heart.

  1. 1

    Okay, as you got to the point where you said you were going up against AHREFS and SEMRush I could see the problem! I do a lot of SEO and I meet with clients all the time - in an hour with me and a PPC person we can bury a client in stats and confuse the hell out of them even when they know what all the buzzwords mean. there is a lot of room outside of SaaS in SEO to make a buck without a doubt (I am building something that is very SEO-ish right now myself). There are a lot of ways to provide services ON TOP of the raw numbers as you discovered.

    The "good news" is that you aren't left at a place with nowhere to go - the "failed" SaaS brought you to a better place from which to build a product/service/thing that is actually needed and that people will pay for in the SEO arena - anything that can address the "what the hell am I looking at and where do I go from here" aspect of SEO stats is probably open territory.

    If you have nursed something to 1.5k - 2k /month you can scale to a paycheck and from there maybe scale it to 2 or 3 paychecks worth of value. Good Luck.

    1. 1


      If anything I'm encouraged. It'd be disappointing if I got nowhere and did not learn much from the experience however having found some success in the services on top of software/data gives me a direction to run in.

      Looking forward to the next leg of the journey, best of luck t you as well!

  2. 1

    Everyone is building a SaaS platform because the term is misused. Basically a SaaS product is a service that resembles a functionality that would otherwise be done with a software installed on the users computer. In this meaning Indie Hackers is not a Saas product, just as well the most so-called SaaS products are not. This way things that are not SaaS are marketed not as what they truly are, but as subscription based software usage, which will and does fail.

  3. 1

    Very Inspiring Tyler and well written. As a technical founder , I struggle a lot with Marketing, SEO even after running my SAAS business for almost 7 years. Just recently, I started learning SEO concepts and got subscription to SemRush and man it has been so interesting to learn how you really can find customers organically. Thank you for sharing your story.

    1. 1

      Thanks and thank you for reading!

  4. 1

    Definitely makes sense what you said. But I wouldn't necessarily go as far as saying 'SaaS is the big leagues.' Launching an EV company, an AI chip design company, developing military drones, software-defined-radio solutions as a defense subcontractor or raising funds to build a supersonic jet or 3D printed rockets - that's the big leagues in my opinion.

    Of course, different areas in SaaS-world have different levels of competition - and the SaaS tool you built (sounds like) didn't offer users really any distinguishing advantage that other tools in this area already didn't offer. It may have been easier to use - but like you said - the issue was that the users didn't know what to do with the data it provided (i.e. it could have been any SEO tool and they have the same issue).

    In a hypothetical scenario - once you learned why users aren't paying for the product (i.e. they don't have a strategy / don't know what to do with the data), you could design some set of "algorithms" that adapted to changing search engine criteria and optimized the SEO strategy dynamically for the clients using say the data you collected + reinforcement learning / multi-armed bandit techniques. You couple that with empirical data that the self-adjusting strategy is driving users to the site in a statistically robust way (i.e. hypothesis's test) and good chance that clients would be happy to pay.

    Again, I don't think this requires high levels of funding (i.e. something that you can't do yourself or with a few people) but rather continuous development and adaptation of your SaaS platform as you learn more about your customer's issues.

  5. 1

    Great retro. Thanks for the transparency. I can see myself falling into some of these same traps, and this was a good reminder to watch out for them

    1. 1

      Thanks glad it was helpful! We all fall into traps, when you can recognize it and adapt is how you grow. Best of luck on your own project!

Trending on Indie Hackers
How long did it take to build your MVP? 32 comments Dealing with lack of motivation due to failed soft launch 27 comments Bootstrapped my productivity app to 700 paying customers! AMA. 17 comments Free e-mail countdown timer that doesn't suck 💩 6 comments Return of the external link 4 comments How To 10X Your Design Rates in 2022 (Free Script) 2 comments