“How can you make a startup successful?”
That was my main question when I sat down with Yaron Samid.
He’s a multi-exit founder whose companies have been acquired by the likes of Microsoft and Cisco, to name but two.
Yaron is a regular lecturer at Stanford and Columbia business schools. He is the founder of TechAviv, one of the largest founder clubs in the world.
“My true passion,” he told me, “is helping the next generation of entrepreneurs.”
So he was more than happy to tell me the story of how he built and scaled his fintech startup BillGuard.
From the number of actionable tips and insights in this interview; it’s easy to see why BillGuard became one of the most popular fintech startups in the world.
But before we get into that, let’s go back to the beginning, starting with how Yaron started his entrepreneurial journey.
Q: Did you always want to be an Entrepreneur?
A: I always wanted to be a Founder and a CEO. I was very inspired by my parents who were both very entrepreneurial.
I was fortunate, I knew that I could have a non-standard career from an early age.
The idea of starting my own company was very exciting.
On top of that, I was just a really sucky employee – I was terrible.
I got fired four months into my first job out of university. They said I was: “too creative”.
This was back in ‘95 when the internet was getting started and I knew a little bit of HTML. So, instead of doing the job the way I was supposed to do it I came up with a web-enabled way of doing it.
They weren’t very impressed and told me: “go back to doing it the old way.”
I was very young and brash and I didn’t take that very well. They fired me the same day.
Which was the best thing that ever happened to me, because I started getting into the internet business. I joined a startup company and learnt from the CEO how to build companies and that’s how it all got started for me – being a bad employee.
“Getting fired was the best thing that ever happened to me”
Q: What was your “A-ha!” moment that made you realise you needed to start BillGuard?
A: So BillGuard was an accident.
My wife got our credit card bill and there was a transaction she didn’t recognise. She came to me and said: “Honey what is this charge?”
And when your wife comes to you and says: “What is this charge?” you’re like: “Oh God!”
But I didn’t recognise it either. So I googled it and saw a lot of people were complaining about the same charge.
That gave rise to what I call a lazy man’s idea: What if there was a way to harness all that collective human knowledge about bad credit card charges? Then use it to notify other people who had those same charges?
That would be a nice service that I could benefit from as a person who doesn’t check their credit card bill.
I had already built two companies before BillGuard so I had some experience with crowdsourcing technology and data science. So I set out to build a fintech startup to solve a problem that I had.
That was the genesis of BillGuard.
Q: What was the market vision that helped you shape BillGuard? Did it evolve over the years or has it remained the same since its inception?
A: When I started working on BillGuard the vision was to become a security company. Our tagline was literally “Antivirus for our bills.”
To advertise to consumers we said: “This system will check your bills for you and notify you if it finds mistakes or fraud.”
It was very much set up like a Fintech security company – that was the initial vision.
We were building a “Set & Forget” security product. This requires a lot of “FUD” Marketing – Fear, Uncertainty and Doubt.
“We had to convince consumers that, without our app to protect them, they would lose money to fraud.”
It’s a very expensive process. Big security companies spend hundreds of millions of dollars a year to scare people into buying their products.
For a fintech startup, it was too cost-prohibitive. Early on, we realised our Cost per Acquisition (CPA) was going to be too high; so we pivoted to a B2B model.
Banks approached us because they were interested in the technology we had built. Banks didn’t have a way to crowdsource knowledge to find bad charges.
So we tried to sell the technology to the banks. However, the sales cycle for banks is very very long. It’s painful. The kind of pain I wouldn’t wish on anybody.
“Simply, we were not set up to just sit there and wait for the banks to deploy.”
So, after two years of investing in a B2B marketing and sales team, with pilots in major banks, we made one of the hardest decisions we had to make. To pivot back to a consumer app.
This meant letting go of some amazing talent we had brought in to sell to banks. They just weren’t the right fit for a consumer business model. It was heartbreaking.
This time, instead of a security company, we built a personal finance company.
We completely redesigned the product. It was no longer a “set & forget” website. It was now a mobile app that would help you identify ways to save money and protect you from fraud.
That worked very very well.
It grew very quickly. Our product team did some strong product development, UX and design work led by my amazing co-founder and CTO, Raphael Ouzan. We were able to build one of the fastest-growing finance apps in history. Within a year and a half of launch, we grew to 2M active users. We were regularly ranked the top finance app in both the Google and Apple app stores.
At that point, we were acquired by a $2B unicorn company; while we were still growing rapidly.
“We completely redesigned the product. We were able to build one of the fastest-growing finance apps in history. Within a year and a half of launch, we grew to 2M active users. We were regularly ranked the top finance app in both the Google & Apple app stores.”
Q: Before pivoting, how did you validate the original idea for BillGuard?
A: We did a few things.
First of all, to validate the idea before we had a product or even a team, I went out and did surveys.
I asked people: “If we build a free product that would check your bills for you and notify you if there was a fraudulent charge that your bank missed; would you use it?”
Almost 100% of people said: “I would absolutely use it, I never check my bills, I know I’m losing money. I would love something to check for me – especially if it’s free.”
“That question was a big mistake.”
That kind of question is a “yes question.”
Everybody will say yes to that. You’re not validating customer demand. More specifically you’re not validating customer action.
People had the intent, but when we launched, suddenly nobody stopped what they were doing to sign-up and use it
Taking action is not free. People have to make the decision to go get your product. That requires a driving force.
Like if you have a headache, you will stop what you’re doing and go and buy some medicine. Unfortunately, we did not have that.
People didn’t know or have a fear that they were losing money. So they didn’t take action to get our product.
There was something else we did that was also, in retrospect, a misleading indicator. We joined a startup competition called TechCrunch Disrupt. It’s the biggest global startup competition.
“Taking action is not free. People have to make the decision to go get your product. That requires a driving force.”
We did very well, we came in second place as the Top Startup of the Year for 2011.
We told the story well and people were excited.
We got a lot of attention, we raised money from the best investors in the world.
We had Bessemer, Khosla Ventures, Peter Thiel from PayPal and Eric Schmidt from Google, etc. Tier-1 investment funds who all believed in this vision of outsourcing bad charges.
But customers still weren’t caring about the problem.
When we launched, the product wasn’t converting and the funnel had massive drop-offs; that was the real data. That was real customer validation.
The results made it very clear that there wasn’t enough demand for our security product. Hence the decision to pivot.
Q: So, before you pivoted who was your startup’s “primary stakeholder”?
A: We had two categories of consumer that we profiled. We called them:
- Spotters – Someone who is meticulous about checking their credit card bills. Line item by line item. They’re very nervous and aware that companies make mistakes and they check for them religiously.
- Slackers – People, like me, who never check their credit card bill. They know there might be some mistakes on their bill, but they don’t care. If they do check they skim for big charges and ignore the little ones.
We targeted the Slackers. We said: “Hey guys we’ll check your bills for you; we’ll do the work and notify you if there is anything that looks fishy.”
The problem with those Slackers is they don’t take action to solve the problem in the first place. They aren’t aware of how big a problem it is.
The Spotters are aware of the problem – but they take care of it themselves.
What we were able to do through our marketing was to find a sub-segment of slackers. They were aware of the problem, wanted a system to take care of it and were willing to take action. But, as I mentioned already, it was too expensive to market to them.
Building an MVP?
Looking to build fast and reduce your time to market? We can help you find the core set of features to gather user information and iteratively create the best possible product.
Q: Did the Slackers deal with the problem before your original solution or did they just not bother?
A: Most of them would only find out after the fact if they had some big charge on their bill they didn’t recognise. In that case, they would call the bank.
That leads me to another aspect. When you’re launching a product you have to assess:
- How big of a “pain-point” is the problem?
- How easy is the solution they currently have at addressing that pain-point and is that “good enough”?
In the case of bank fraud, the consumer has zero liability. They pick up the phone, call the bank and say: “Look I have this big charge here I don’t recognise, it’s not mine please remove it.”
The bank replies: “Ok no problem.” and they remove it. The bank does all the work.
So we had Low Pain + Easy Solution
On top of that here was one more deadly sin: low frequency.
99% of the transactions on your credit card are absolutely fine. Someone rarely has fraud on their credit card.
Even if there are hidden charges or mistakes, once you find them and get them fixed they don’t reoccur. The problem is solved.
We were left with:
- Low Pain
- Low Frequency
- Easy Solution
If you analyse it that way you would’ve never started that fintech company; you would’ve been dead on arrival.
That’s the kind of teaching I do now for other companies; it’s a very important lesson:
“You have to carefully, and quantifiably, address: how big is the pain, how frequently does it occur and what are the existing solutions for addressing that pain.”
Q: With that analysis in mind, how did you build a product that would still work?
A: We didn’t! We built a product that was good enough for the subset of Slackers.
The problem exists. At a macro level, it’s huge. We quantified it, it worked out to about $14B of unwanted charges per year in the US alone.
So the macro problem was big. But each individual wasn’t aware of how big a problem it was for them.
Even though we couldn’t afford the marketing; 50,000 people were using the first version of our product. They weren’t paying us anything but it was growing.
We had built a product that fulfilled our vision and worked well.
But the market didn’t have a demand for a personal finance security product; it just wasn’t there.
Q: Which KPIs did you set for the people who were using the first version of the product?
A: So obviously Cost per Acquisition (CPA) is the number one quantifiable way to know whether people will take action.
In our case, there was a high-friction registration process. You’d have to connect your credit cards to allow our system to monitor them; this process caused a huge drop-off in our onboarding funnel because we needed to ask our user’s to enter their login information for their online bank or card accounts. Even though we did not store this information, it was a very scary ask.
That was one of the key KPIs. We analysed, very granularly, our onboarding funnel. We knew we had to get our funnel to convert enough for us to support a freemium model.
Meaning, you got to use the system for free for your first two cards. To add a third you would have to upgrade to the premium model. In our case, to get that to work, we needed to achieve a CPA of $10 or less.
We initially had 50% dropoff in our onboarding funnel. It was almost always when we asked them to connect their credit card; so we could monitor their transactions.
Think about that for a moment:
People who care about fraud on their credit card are risk-averse.
The last thing they want to do is give the password for their bank statements to a fintech startup.
Looking back this seems obvious!
The problem is:
“As an entrepreneur, you’re so excited about everything that you get too emotionally tied to your product. You’re not intellectually honest enough to be true to the data.”
And the data showed that we had a $150 CPA – it was way off, the economics didn’t work!
We looked at the KPIs and our funnel’s conversion rate. We worked hard to build trust with our users and brought our conversion rate up to 75%. Once we had that, we knew we were on the right track.
The other KPIs for us were engagement and retention.
It is critical for any consumer app to get people engaged with the product and to retain them.
If people use your product regularly, its top of mind, and if it’s top of mind, they’ll tell a friend. If they don’t, forget about organic growth.
But let me give you an example. Nobody in the history of mankind has ever told their friend:
“I love my McAfee Antivirus!! Oh my god, it’s SO GOOD! You’ve got to get McAfee!”
“Nobody is excited about their security products.“
You don’t engage with it on purpose! You’re not supposed to. What you’re buying into is peace of mind. I set up McAfee and I know I’m safe – I don’t need to think about it anymore.
So we needed to change the discussion from security to saving money: “Do you want to pay full price for that coffee? Or do you want to get a notification on your phone that gives you a 25% promo code as you walk into the coffee shop?”. Everybody wants that product!
And it drives action.
If you see someone standing in front of you in line getting 25% off their coffee the first thing you do is say: “Hey, what’s that?” and they say: “It’s my BillGuard savings alert. It shows me that there is a coupon here I can use and automatically downloads it to my phone.”
Boom! Everybody standing in line is suddenly downloading the app for the coffee they are about to buy.
That drives massive organic adoption and growth. When we did this and made other UX tweaks to the mobile app, it brought down our CPA to less than $2 per newly registered user.
“We needed to change the discussion from security to saving money. When we did that it drove massive organic adoption and growth. Our CPA was less than $2 per user.”
Q: So you pivoted to B2B with banks, then back again to B2C with the personal finance management app.
How did you coordinate your team to deal with this tectonic shift?
A: First of all we had the painful decision to let go of the team we had when we were selling to banks. They didn’t fit with what we were trying to do on the consumer front.
Luckily, the core team were all B2C, personal finance, data science and UX experts who knew how to build great apps for consumers. Which was also my background as a product-centric founder and CEO.
That’s another big lesson for founders: you want to stay true to your DNA. Don’t try to be something you’re not.
I’m not the guy who puts on a suit and tie and tries to sell to banks. It’s not what I love and it’s not what I’m best at. It wasn’t my passion.
So, when we pivoted back to consumers, it was almost like: “Fantastic, now we are back home, doing what we love.”
In the end, we just needed to figure out how to do it right, because we were doing it wrong at the beginning.
Q: The technology for your startup’s MVP must’ve been complex. What were the most important features that you included in that MVP?
A: So the very first MVP was a very simple website.
You would go on and register the cards you wanted us to monitor. We would scan it for you and send you a weekly email saying:
- Transactions 1-20: Ok
- Transaction 21: We don’t recognise this charge in our system. Please check this charge and let us know whether it’s ok.
That was it. The way we got the analytics for knowing whether or not a charge looked ok to us was through:
- A database of merchants that we knew were legitimate.
- Scanning the web and message boards for scam charge reports.
- Asking the few “spotters” we had using our platform to click every transaction. We gave them a tool so they could go through checking yes or no on everything.
About 2% of our network were Spotters.
To put it in perspective that’s the same percentage of users as content contributors on Wikipedia. And in their case, 2% of users contributing equals the most accurate encyclopedia in the world.
So we leveraged the spotter reviews, this gave us early analytics. We then used that information to provide the other 98% of our users with reports once a week – “Set & Forget”.
Q: Looking back on your experience, do you have any advice for founders on product development?
A: The absolute most important thing when it comes to product development?
“Don’t infer that there will be a demand for the product you’re going to build. You have to quantify customer action – not the intent.”
Before you decide to build an MVP, write a line of code, onboard your team or reach out to investors you should:
- Create a landing page for your product that describes the value proposition.
- Run some ads, drive some traffic to that page.
- See how well it converts to whatever your Call to Action (CTA) is: Register, give us your email – whatever it might be.
Get your target users to sign up before you build the product. Then follow-up with: “Thank you for signing up, we’re working on the product now. You’re going to be among the very first to hear about it when we go live.”
That will give you a rough idea of how much it costs to acquire a customer based on the value proposition that you think is so wonderful.
What you will most likely find out is that your value proposition isn’t that wonderful and that it’s far more expensive to acquire a customer than you thought. But that data will enable you to iterate on your value proposition:
- The product
- The positioning
- Who you’re targeting.
- Your optimal marketing channels
You’ll be able to keep tweaking this until you get to a point where you will have a much better handle on your ability to acquire a customer that takes the action that you want them to take, and cost that works for your business.
You will have an informed, quantified answer to the question:
“Will customers want my product enough to take action at a cost that would enable me to run a sustainable business?”
You still won’t have metrics around engagement and retention, but you’ll have a sound starting point and a good indication of whether or not your product idea will be dead on arrival.
There are great tools for doing this. One of which is QuickMVP. You can create placeholder landing pages, run advertisements and get quantified results on your onboarding funnel. How much it costs, where people drop-off etc. It allows you to constantly iterate and A/B test.
If you go to angel investors, for example, with that kind of quantified data, they are going to know you are a serious entrepreneur. They’re going to see that you:
- Are responsible and credible;
- Have done your homework;
- Know the initial foundation of your growth economics.
Before building anything; you’ve quantified that people will take action on your value proposition.
If we had done that at BillGuard, we would’ve saved about four years worth of time.
Perhaps we would’ve jumped straight to the personal finance value proposition – rather than a security or bank product.
That’s a simple exercise to do and it’s my top recommendation for product development.
Q: Do you have any advice on raising startup capital based on your experiences with BillGuard?
A: BillGuard was my third company. So, when it came to raising money, you have to take into account I was a repeat entrepreneur and a well-known entity.
Investors were happy to back anything I was doing because I had a good reputation and track record.
“As a seasoned entrepreneur, it’s a lot easier to raise money.”
My main advice to first-time founders is to create a network with investors early. Reach out and build relationships with investors before you ask them for money.
Tell them what you are thinking of building. Ask them if, based on your KPI goals, they would be interested in investing in the future. Find out what KPIs they would like to see based on your vision.
And they will tell you. They’ve seen hundreds of companies and thousands of pitches. They can see the signs of a company that is going to have compelling growth.
Collect this information from a few investors. Then go out and execute those KPIs by scraping together some friends and family money.
If you succeed you can go back to the investor with facts and figures rather than just a story. Then you can start talking about a proper early-stage venture round.
Firstly, that puts them in a position where they have to take you seriously.
They’re going to look bad not investing in your startup if you did what they said they needed to give you some capital.
Second, it makes you look exactly like the kind of founder they want to back. It makes you look credible and your business predictable. They don’t know you; their biggest fear is you’re just somebody who’s selling them smoke & mirrors.
So, when you tell them you are going to do “A, B, C” they may doubt you. But, six months later when you have proof you did “A, B, C” you tick a box in their brain. You show them you’re a responsible, credible entrepreneur they can trust who has a good grasp on the growth levers of your business.
You mitigate the number one risk for an early-stage startup investor; the founder:
“Early-stage investors back great founders. They know that everything else is going to change. The product, the market, the competitive landscape will all change, but they’re betting on you.”
In short, developing the relationship and backing it up with credibility will get you a check.
Q: Last question, what advice do you have on hiring for a startup based on your experience?
A: The most important decision is your co-founder. I can’t stress this enough; it is literally the most important decision you will make as a founder.
Most startups fail because of some sort of HR dynamic. The vision might be good, the market might be good but if your team can’t execute against it then you’re already one foot in the grave.
You should find a co-founder who you feel absolutely confident to go to battle with for the next decade.
When it comes to looking for both your co-founder and your first hires, your ideal situation is to work with people you already have a relationship with.
Those first 2-4 startup employees are almost going to be co-founders. They are taking a huge risk betting on you. So, hopefully, you can pick from a group of people that you know.
“The most important decision is your co-founder. I can’t stress this enough; it is literally the most important decision you will make as a founder.”
The same advice goes with startup talent as it does with investors. You should be developing relationships with talent well before you launch your company.
You should be hanging out at universities, meetups, online communities and conferences.
Say you are passionate about sustainable food tech and you want to start a company in that sector.
Months before you should be a part of that community. You should be having conversations and contributing content to that community before asking anybody to join you.
“If you are in communities relevant to your startup, you’ll not only find talent, you’ll find customers – and maybe even investors.”
When you hire your team you must focus on startup people. People who can take volatility and work for lower salaries than the market.
They need to understand the value of the startup equity that they’re going to be getting. And you should be generous with your equity on your first few employees.
When hiring for a startup, I would optimise on intelligence and learning fast on the job over fancy degrees from fancy universities.
In the startup world, you need people who are street smart, creative, agile, adaptable.
If you’re building a consumer product and you suddenly decide to pivot to a B2B model you need to have software engineers who can pivot with you.
That requires a certain personality and skillset. Ideally, with developers, you want full-stack engineers.
Over time you’ll hire specialists but in the beginning, you want to hire very brilliant generalists.
Q: Is there any other advice you have for first-time founders that we haven’t covered yet?
A: Find a seasoned entrepreneur who can mentor you on the basics. Like how not to fall into term sheet traps from predatory investors. Add these mentors to your startup’s advisory board. Give them equity; at least 0.25%.
Thank You, Yaron
Thank you again, Yaron for taking the time out of your busy schedule to talk to me.
For me the main takeaway from Yaron’s story is this:
As usual, there’s no handbook, no top-secret formula to startup success.
In short, it comes down to a lot of hard work, learning from your mistakes, perseverance, good timing and taking the right advice.