This post was originally published July 4th 2024 on my newsletter, Noah's Ark.

For this weeks newsletter, I wanted to take some time to share my own personal journey founding a company and lay out in plain english what I’ve been working on and how I’ve been approaching it. Hopefully there’s something useful you can take away that will help you as you pursue your own goals.

I met Aman on the YC co-founder matching platform. As many of you know, using the YC platform can be an exhausting experience. It’s filled to the brim with college age Elon Musk stans, tarpit ideas, non-active accounts, con artists, and awkward engineers with a god complex building things no one will care about. In the rare instance you come across someone with potential, odds are they live far away, are already working on a project that isn’t of interest, aren’t actually looking to build a company, or aren’t aligned from a personality or ethics standpoint.

So why was I there?

For me the YC platform was a way to refine my own thinking and better understand what I was looking for in a potential project and co-founder. I knew that the 2 most important decisions if/when I decided to start a company were the problem I chose to tackle, and who I chose to tackle it with. The YC platform was an easy way to identify all the ideas and characteristics of people that I wasn’t interested in. Another sports app - pass. Building a job marketplace - pass. Another social network? Not interested. If I was going to build something, it needed to actually help people and bring some good into the world.

Then I came across Aman’s profile.

I scanned it, and then started over again reading very carefully this time. It seemed too perfect. There wasn’t a single red flag that I could see on his page. He had built the prototype of a deepfake audio detection model and was looking to help provide a solution to the growing problem of deepfakes.

Now this was interesting! This especially hit home for me because my grandfather had gotten a scam call earlier that year where someone had cloned my voice to try and get him to send money. At the time, I didn’t think there was anything that could be done - I certainly wasn’t a cracked engineer who could build a detection model, but here was someone who was.

I reached out to Aman and we began talking. No red flags. In fact, he seemed like exactly the type of person I would want on my team for anything, work or otherwise. High intelligence, high EQ, articulate, scrappy, high ownership, great decision making capabilities, high moral standards, and a good sense of humor. Over the next few weeks we identified that there was mutual interest, and decided to trial working together for a period of time. For the next 3 months I worked alongside him to help him begin building a business around the MVP of his detection model. The 3 months were great - we worked well together, and at the end I came on as an official co-founder.

Our mission was (is) simple: Protect human authenticity. We are starting by allowing people to trust what they hear again with tools to help them verify content authenticity and speaker identity.

Armed with our mission, an MVP, and some sales tools, we set out to save the world… and found out the world didn’t think it needed saving.

Or at least not in the way we thought.

It turned out that “deepfake detection for audio accessible via API” didn’t actually resonate with anyone, or solve anyone’s problems, and certainly wasn’t something that anyone was willing to pay for. Consumers wouldn’t pay for security products, and the root problem for businesses wasn’t ever “how do I detect deepfakes”.

Investors were equally indifferent. We were too early, too unfocused, with too little traction, and too little experience fundraising. We lacked the necessary clarity and focus of our vision, customer pipeline, and polished pitch required to raise capital.

But that was fine. In hindsight, it’s a good thing we didn’t get investment at that time. Money is fuel to accelerate and we didn’t know where we were going yet.

After a couple months of (failed) sales to tons of different prospective customers, we began to have a much clearer understanding of what the “real” customer problems were. In fact, we had mapped out over 30 unique customer problems that we could help address - none of which were the one we initially thought.

The new challenge became, which one do we choose?

We know we need to narrow our focus, but what is the best problem to take on, and who is the best customer to serve. In pursuit of this, we identified a couple different promising problems and customer categories, then tailored our positioning, messaging, and sales materials for each customer base. The results were super positive. The identification of the “real” customer problems and adjusted positioning and messaging started resonating. We quickly lined up design partnerships with high profile companies in several different verticals.

Then the hard part. We needed to choose 1. Not 5, not 2. 1.

1 problem that 1 type of customer would pay us to solve on a reoccurring basis. There is always room to expand in the future, but at the start it’s important to be extremely focused and build for 1 specific person at 1 specific type of company that has 1 specific problem.

We settled on helping security teams in highly regulated industries like insurance, banking, and financial services defend employees from social engineering, vishing, and deepfakes across voice and video communication channels.

This clarity of focus in who we are serving and the problem we are solving has dramatically improved our messaging, product, fundraising, and everything else related to our business, but getting there has been an iterative learning process.

In his newsletter, which I highly recommend, Rob Snyder calls this process “unfolding” (If you have been reading my prior newsletters I’m sure you are already getting tired of me quoting him, but he really has helped me improve my thinking).

By going out and selling our product, we were able to learn what people care about and identify what we needed to build and for whom.

Even now, I’m positive that we are still wrong.

In fact, I know that this is the only certainty when it comes to launching a business. We will continuously identify areas where we don’t fully understand our customer’s needs, and ways we need to improve our offering (Rob describes an “offering” as product+services) in order to better serve them.

The good news is that we are aware of, and comfortable with, this “unfolding” process. I don’t know where we will be tomorrow, but I can tell you where we are today, and I can clearly look back on where we were in the past and see that this unfolding process has dramatically helped our business and will continue to do so in the future.

There is a specific quote from Rob that guides a lot of my decision making and that I credit with keeping us laser focused on our customers and their needs (and working towards “building something people want” as a result):

“People don’t buy products, they buy solutions to their problems, your offering (products + services) is HOW you solve their problems”.

It is my firm belief that as long as we continue to have this perspective, we will avoid building something that nobody wants, and will avoid getting so enamored with our own tech that we lose sight of what’s most important. Don’t get me wrong, I think our tech is awesome, but it’s only valuable if it is solving real problems. Our job as founders and as a company is to make sure that this always remains the case.

Looking back on it, this was our root problem when we first started.

We were so enamored by what we had built that we went and sold the product (deepfake detection for audio), when we should’ve been selling a solution to a real problem. If we had done this sooner, then we would’ve realized we didn’t actually know what problem we were solving (and selling a solution to). It’s no wonder nobody cared.

The good news is that by going out and selling, we were able to identify this issue quickly and begin course correcting.

Now Noah, aren’t you worried that your competitors might see this?

To be honest with you, I don’t care. I’m focused on our customers, and better understanding their needs.

If someone want’s to try and copy us, let them. They won’t win because they are focused on the wrong thing. With time and people and funding they can try and duplicate what we have right now, and they might get there. But you know where we will be when that happens? Not here. Our unfolding process will continue and we will have adapted based on our learnings better understanding of our customer and their needs in the meantime.

I’ll close with why I decided to share all of this in the first place.

I think it’s important for everyone to get a realistic look at how startups come about, and what the process looks like as it’s happening. I think things are often glamorized and founders often don’t offer practical advice when they are actively still in the trenches (and believe me when I say we are still in the trenches). We are by no means wildly successful yet, but I am confident in our ability to figure it out, and I trust the unfolding process.

Finally, I think there is tremendous value in both writing and building in public.

Writing helps me refine my own thinking, identify areas where I may not have clarity, help me process things, and better articulate my thoughts and ideas in the future. One of my favorite thought leaders in the cybersecurity space, Ross Haleliuk discussed this concept in his newsletter last week and I couldn’t agree more.

The concept of “building in public” is also something that I have come to believe in over the last year. Outside of allowing you to learn and improve faster, showing others what you are doing increases your “luck surface area” by allowing people the opportunity to provide feedback on, or assistance with, whatever it is you are working on. This is especially true when (not if) you suck at something.

All of this to say, I hope you enjoyed learning a little about my journey with DeepTrust and found something useful and applicable in this blog that you can take away. On my end, I’m thrilled with our progress at DeepTrust so far, and I couldn’t be more excited about our future. Things are by no means easy, but I am confident in our ability to continuously keep learning and improving as we go.

The only thing I know for certain is that we will be better and wiser tomorrow than we are today.