Amr Shady: Data is King | Moghamra Podcast #016

My story starts with honestly my father. I think you always find the father or the mother in any entrepreneur story. My father is an entrepreneur. I'm an electrical engineer, I'm an entrepreneur. You can see how that worked. I was pretty good at math, and my father since I was probably... I remember when I was actually, back in the day, you go and the things that I used to play with were electric circuits, the models that you make electrical circuits when I was like six. Summers just sit down and solve math. Math and engineering, electrical engineering was baked in my day to day. It was a very natural thing. I didn't have to think twice about what I liked, and I just enjoyed it. I enjoyed solving these problems, and I could read numbers really well.

I'll jump many years. Electrical engineering, computer engineering back in the day. I think computer engineering is part of electrical engineering. I did a lot of work with TA Telecom, where we were doing a lot of work with mobile operators. We were processing five billion transactions a year, and we had so much data. But in 2000, I started investing a lot of time being super interested in data science back in 2012, around 2012 or so. And in 2015, I decided to go do a data science degree, but I wanted a data science degree that is not just technical only, it's a combination of technical and business. Scientists and business people have no idea what to tell them to bridge the gap.

Yeah, it happened. We'll get back to that. There's another personal story in 2013, why did I go to New York? But definitely on the MSB program, it was a mix. You go and attend, and then you can go back. You don't have to stay. So it's really underwriting the new economy. So we started looking at founders and how do we find the best founders as early as possible. And it was really inspired by being at Endeavor and being part of that. Back then, I was a board member at Endeavor, a selected entrepreneur. I had this network of amazing people. How can we find these people as early as possible to invest in?

We started by... the algorithm worked really well. The founder of the NYU Center for Data Science was the person who reviewed it. He was very excited about what we did and what we built. He joined as an advisor, and he was like, 'You guys have to commercialize this.' What happened was we left school, we formed a company, and we started selling the scores to VCs. But it was a very good way for me to get plugged into the Silicon Valley VC community and start building relationships outside of fundraising. For me, it was so much value in talking to VCs and understanding the data driven VCs. It was easy for me to build these relationships and get to know a lot of the VCs back in the day, 2016, 17, 18. It started with selling scores to VCs, and then it expanded into selling scores to banks like Silicon Valley Bank and others. Then we started to see the application of our data on financial services, and we started experimenting. We can move up the food chain. If we're able to identify the best, impact doesn't have to come from investing. Why are we narrowing down this whole thing to just investing? We can extend credit to these companies. We can extend payment rails that are problematic for them. Cards... if you look at founders in the US and outside of the US, getting access to corporate cards to pay for your advertising and your hosting was not there. We can have an impact on them outside of investing, outside of mentoring. There's another layer of value creation for them. I got super excited. It was very aligned with what we were doing.

Yeah, design sprints. A design sprint was first introduced by Jake Knapp, I believe, when he was at Google Ventures. It was a way for you to get to test or get to zero idea basically, knowing a problem right down to a user tested in just under four days. Because the typical process back then, startup with MVP and build something super scrappy, get it out there... how long is that going to take? One or two months? Or over a year? How many shots do you have? That MVP $180,000? What does four days look like? Four days, $12,000. Right. $12,000 versus $180,000. It was a no-brainer for me. Plus it's feedback right from the customers. And one of the worst things that I did was talk to advisers and talk to investors. I was like, 'Hey, we're thinking of taking this from here to here, what do you think?' But we were like... I think it was June 29, 2018, I believe, until November of 2018. Direction. So when I came to see design sprints, I was like, that's exactly what we need. We need to hear it directly from the customers. Stop asking people who you think are super smart. Just get it out there and test it with people, and do it super fast. Because we always knew you had to test with customers, but it would take too long. But I've been super passionate about the use of design sprints, and it made us be ahead of competition significantly, and lead the industry in many things in terms of propositions and features. Tribal pay, Tribal cashback back in the day, these are all things that came from design sprints we did. But we were one of the first to really adopt that ability to learn fast.

It's a four... yeah, on the book actually, there's a book called Sprint by Jake Knapp. It says it's a five-day program. If you do the masterclass and the actual training, you'll find there's a version that's four days. This was back in 2019, so I don't know if they've improved it, maybe it's three days now. I don't know. But let's say it's four days. It's a 4-day process where you start with problem definition, you start doing sketches. The best thing about a design sprint is the strong byproduct of it. You have five to seven people together working together. Back in the day for us, it was in the same room. You don't have your best salesperson pitching their ideas and the tech person who doesn't know how to sell ideas gets drowned out. It's not like this. Everybody gets an equal shot at showing how they think they would approach this problem, and it builds this connection between people in a different way. When you look at everyone at the company who's with us or even left, and you tell them, 'What do you remember about your experience at Tribal the most? What have you enjoyed?' They always say that time when we used to do these design sprints and stuff. People really like it. The main thing is discovery, some sort of initial sketching, and then you have a designer. The whole thing is four days of people just brainstorming around this is the problem, this is what we want to test, this is how we're going to test it, this is what it looks like, and then you actually test it. Condensed in a super tight deadline. Tight deadlines are usually good. You have to have clock management, and you get it done in four days. Design sprints, you're able to find that product market fit essentially. Initially, you start to see your hypothesis around what customers value. You can test willingness to pay, you can test many things. Market messaging, use cases. You see how people are using it and what are your economics. If you start charging, do people still use it? It gives you a good directional path of, yeah, this is a path where you can continue to go and flourish.

We decided at that point we started by building the product for real. People saw something that was not final. I think that creates a gap in people's minds, and they feel like they need to contribute to the idea, and they start throwing ideas at you. They want to add value to the process. When they see something final, it's a very different mindset for users. One of my favorite feedback was, 'Is this live or is this a test?' I love this feedback because that's exactly what you need to get out from a design sprint. People believing. Once they believe, they just say, 'I just need to press a button and it's live,' like it's beta or something.

And iterating on different acquisition channels and expanding your scope. You start with startups, then you start growing into SMEs. A lot of startups, friends of startups, investors in startups said, 'We invested in this company. We have the same problem. Can we get on board?' Yeah, sure. Get on board. A company that's doing $30 or $50 million a year. Come on board. We found that this problem started to become a problem for SMEs of different sizes. That started to become a problem for us. We saw that we had to focus. In growth mode, you're trying to service more and learn more. We were open to everything, but when we had to be efficient, we started to focus a lot more on who exactly has highest product market fit, looking at the data that says this is high product market fit, and then doubling down on them.

Saudi and the region actually was back in 2019 very different than what it is today. From a regulatory perspective, infrastructure, open banking. We saw LatAm from a regulatory perspective, infrastructure perspective, especially on open banking, market sizing, interest in general, speed of growth. We found strong tailwinds for us there, and we made a conscious decision to focus more on LatAm. The US, the currency, the amount of trade between both of them, the cross-border needs, are very different in Brazil vs Mexico. Operationally it was easier. There's a lot of factors that made sense for us to try and focus more on Mexico at that point. That's one of the reasons why we were like, okay, we're going to deliberately grow in LatAm during this phase, and out of all regions, especially in Mexico, we're going to double down on Mexico.

Both actually. We learned it the hard way and a good way. Doing it ourselves, but also Endeavor has this program with Harvard where they select every year a group of founders and execs from Endeavor companies to go to a special Endeavor Harvard program called the Scaling Entrepreneurial Adventures. I went to that one, maybe it was 2012 or 13? Sorry, 14. There was a whole session about selecting markets and how do you prioritize markets. But at the end, when you put the factors for me, we do it in a very special way when we need to go deep on selection and it's a tough decision. We have four or five markets that we really need to select between, and their numbers are close. We do Monte Carlo simulation on top of the market selection. Basically, you put all the factors you're considering, you try to put a weight on each of these factors, and you score markets based on that. You have to look at regulations: can you service this market globally? You look at the infrastructure. I can probably list at least 10 to 15 different factors that touch market, regulations, operations, etc. It's a very deliberate exercise that a lot of founders should be conscious about. Sometimes it's so clear you don't have to do the math for it. Sometimes it's clear our biggest market is, I don't know, Nigeria. This is where you just go.

We started doing a lot more direct. We hired an on-the-ground sales team, marketing, and started to try and grow a little bit more consciously inside the region. It involved a lot of marketing and partnerships.

I can't say we faced a huge cultural challenge. Just being exposed to potentially a new culture, meeting people. Sometimes you would have, every now and then, if you're talking to someone who doesn't speak English well - my Spanish is not that great. Maybe that part is a little bit harder, but from building a team it was good. We have a very diverse team overall. Honestly, I can't say we faced a huge challenge. I think some of the challenges in general were more on the macro side. Regarding cultural challenges, you just need to learn more and have people locally on the ground that you can learn from and iterate with.

We were trying to use that global platform to service so many countries. We started to realize we have to make a hard decision. This is where you have to say, 'I'm deliberately going to lose 20% of my business performance.' Overall, the team is stretched to solve small problems here and there in markets that are not generating much. You need to think about these tradeoffs. The number one role that I have as a CEO is to allocate resources: time and money, and obviously people's efforts, towards the things that matter the most. From a resource allocation perspective, our credit performance will go up, our revenue will go down. The tradeoff was worth it for us.

Overall, when you look at OKRs, we use OKRs almost religiously. There's so much talk about OKRs. People say it works, and people say it doesn't. What I know is for us, it's important and it works when it's taken seriously. In general, you need to try and shoot for the 80% target. If you achieve 80%, then you've really succeeded. Achieve 100%, then maybe you didn't shoot high enough. The way we use OKRs is as a change agent. OKRs for us are not target setting; it's a change agent. Something on the revenue side... I know it's extreme, but I know it's incremental change. Tectonic change means we need to change how we're operating in a function or across functions. I continue to think of OKRs mainly as a change agent, as a martial event for change. You set targets that you think we're going to get to 80%. If you get 80% there, then it's something. We try to set them to be a little bit more over-ambitious as opposed to being realistic that we can hit them. It makes it a little bit harder on the founder managing the board.

Really evaluating the CEO on three factors mainly. Number one: does the CEO or the leader of a function know what to do? Is the story clear, do we trust the decision making? How are we making decisions? That's number one. Number two: can the CEO or leader achieve what they set out to achieve? It's a function of having the right people in the team, the right processes. Number three: do you achieve the results? Most people focus on the last one, although I think the first two are very important. The third one is important, you really do try to show you're executing and achieving towards the goals you've set. It becomes hard when you say in my OKR, just setting expectations with the board, setting expectations with investors. This is what we're shooting for. Our base plan is we're shooting for 80%. Educating them on how you're setting these targets is super important. We try to celebrate the 80% as opposed to the 100% more.

Yeah, you can use Monte Carlo. You don't have to use Monte Carlo every single time. If a decision is almost clear, I wouldn't use it. We're looking at top economies. Our mission is to empower SMEs to grow and compete in a global economy. Emerging market SMEs... Saudi, Egypt, Pakistan, Turkey, Indonesia. You start looking at them. If you do a quick rank, a couple of contenders, then experience can kick in, knowledge can kick in. You can write it down. When you write, things become clear. We wrote down the top 10 factors that would make sense for us to consider as we look at this. Saudi became a very strong case for us to double down and explore further.

I think our biggest challenge is we did it too good. We were too good. When you think about scaling, it's all about people, having the right people. We went back to first principles. If you're going to scale and grow, it all comes down to people. But how do you create an engine that can identify, attract, and hire the best people as fast as possible, with the right volume and throughput? We invested in first principles thinking to create an acquisition engine that gets you the people. At that time, we were acquiring the smartest people in the market. If you have smart people in the company, you will be able to dominate the market as you hire and grow a group of smart people together. We did this too good. We got to a state where we built a talent attraction and acquisition process that could get us up to 40 new people every month. Can you imagine? You can go from 0 to 400 in 10 months. We did it too good. We hired amazing people, got them on board, and then the market turned on us. We had 400 people on payroll. It was an old-fashioned growth operationally. You didn't have to be in all these markets. There was a premise that you have to be in every market if a customer has operations there. Okay, I get it, but is this the segment that you really want to grow with in this market? You start making these tradeoffs. I don't have to be everywhere. It's operationally heavy and costly. The problem is we actually did it too fast. I don't think we had a problem scaling from that perspective, to be honest.

Definitely having the right people that match the problem. You hire the right team that matches the problem. If you have a good product, it's very hard to have a blanket answer for everybody. What I've seen in problems with scaling, they can come from acquisition being too high, conversion being too low. Is it a problem? Is it the wrong sales team, or does the product not have a fit for this specific segment? The market dynamics... the product itself was in demand. People needed the product. They wanted the payment facilities, the financing. It's about matching our acquisition and the cost of acquisition to the needs of the company.

Look at the book from the team at Social Capital, they call it the quantitative approach to product market fit. They give a whole bunch of analytics. There's a version two now. They give you stuff to understand your product market fit from a data driven way. That's step number one to understand if we have a product problem. In general, I think data and a data infrastructure is key. Just by nature of who I am, I think the founder's touch on a startup has to be there. People can shy away from it, but you should lean in. The founder's personality and presence in the company has to be there. Between myself and Dwayne and K, there is that touch. People can sense it through the values, through how we approach problems. Given my mindset of being analytical and data-driven. If you have an early stage founder, 'Do I have product market fit? How is my product interacting? If I continue to scale, would it be good or bad for me?' That's step number one. Then it's really experimenting on growth channels. I guess this answers how you were able to find that product market fit by building that infrastructure.

At Tribal, you have new products that come out, new life cycles. You're always keeping an eye on whether I have product market fit for this product set or with this segment. It might grow or fall as you evolve. It's foundational.

From a talent perspective, I think I've underrated how important having the right HR or people leader in the company is. When founders are scaling their companies, they have to have a good line of sight on how we're doing on our ability to find the right talent, get them excited, and get them to join. The talent function is made of three things: acquisition, performance management, and culture. If you focus too much on one, you lose the other. If you focus too much on performance management or trying to get low performers to be high performers, you're making a mistake. If you don't build the right systems to evaluate performance diligently, you're making a mistake. If you don't have the right tools and processes to attract and onboard people, you're making a mistake. It's a lot of work. And if you don't have a leader who covers these three areas, you can hurt the organization. I've been through it myself. You need to learn how to identify your areas of strength and where you need help. In interviews, I try to understand from a numbers perspective how analytical a person is. In performance reviews, one of the best tools I've used measures culture and engagement. We used Culture Amp for a long time, and it was great because you could see dashboards of how each team feels. When interviewing for HR, I look for someone comfortable with numbers, who has used these tools. The other thing I do that I found hurts when I don't use it is a product called Thomas International. They have a personality profile assessment with 20-something questions. There's no right or wrong answers, but it gives you a profile. It creates high self-awareness for the candidate and for you. You know the natural tendency of this person. There's also a job matching tool where you design the role not just about experience, but about skills and natural strengths. It tells you this person needs good people skills, ability to organize, influence ideas. Then it does a match. That's another angle I look at when hiring. It gives you potential interview questions very nuanced to this person in this role. It's been great for me. I also have my own personality profile there so people know what they're dealing with. It says I rely on data, I'm very shrewd with numbers, I might not come in with confidence but if I have the numbers, I will be super opinionated.

The founder's gut intuition sometimes tells you that the data is missing something, maybe this angle, or we don't have enough time to get all the data. So there are cases where you know it's the right thing to do despite the numbers.

I was back in 2010 when I went through the process. It was self-nomination. No one referred you. You nominated yourself. The chances of a self-nominated candidate going through the process and getting selected back then was like 1%. We said, 'Screw it, we'll do it anyway. 99% chance it's not going to go through.' But we believed in backing it. We've realized and got confirmation from other amazing people that it's worth it.

It's hard when your growth is a function of two different teams with different people, and it goes up to you as CEO to try to make it work. It's a daily thing. Some of the recent changes we've made: you want to grow revenue and sales, but it's susceptible to our credit reviews and credit management. The credit team pretty much has the right to say, 'This customer gets this credit line' or 'We're not going to take this customer.' We used to have these two decisions as separate, but now we've created a structure where sales and credit work together on a daily basis. Before, both were reporting to myself and my co-founder. It's hard to do that. Then a few weeks later I was reading about Jeff Bezos. Something super interesting: we've always been trying to improve communication between revenue and credit. Then we read that Jeff actually tries to eliminate communication altogether as opposed to streamline it. It made a lot of sense. Whenever you think these two teams are not communicating well, oh we need to improve communication, I think that's the wrong approach. You should avoid the heavy communication where things start going against each other. That's important, and I guess in some sense data does that. It's a centralization of everything really.

We did a lot of thinking about what's next for us. In Mexico, Alhamdulillah, we've got strong momentum, economics growing. Understanding how to grow from there. We have exposure to different markets, and we want to double down. For us, it was Saudi versus other markets. Saudi made a lot of sense. I presented to the board top 10 reasons why Saudi makes more sense than other markets. One of the biggest drivers is how much support SMEs are getting from the government. It's on the national agenda to make them better and grow. The market is very big. The need is there. Your ability to scale a product with all the nuances is better. It gives you optionality for other markets. From a currency perspective, you don't have to deal with devaluations because it's pegged to the dollar, making operational life easier. How Tribal is perceived in general is high. If we continue to build on that, there are many reasons. We've been excited to double down on Saudi in the coming period.

Thank you. Thank you so much. Thanks for having me.