TRANSCRIPT
EPISODE 36: Ella Hilal
Jim Freeze Hi! And welcome. I’m Jim Freeze, and this is The ConversAItion, a podcast airing viewpoints on the impact of artificial intelligence on business and society.
[UPBEAT MUSIC]
Jim Freeze Today on The ConversAItion, we’re joined by Ella Hilal, Vice President of Data Science at Shopify, an e-commerce platform with more than 1.7 million merchants across 175 countries.
Ella has more than 15 years experience in AI, machine learning, data analysis and the internet of things, spanning both academia and the tech industry. Today, she leads a team that’s building data-informed products, enabling merchants to start and grow their businesses on Shopify.
She’s here to tell us all about some of her favorite data-driven features, what “data-informed” means, and how Shopify empowers merchants to navigate the often-turbulent business of e-commerce.
Ella, welcome to the ConversAItion! We’re thrilled to have you.
Ella Hilal Thank you so much for having me. I’m super excited to be here and have this chat with you.
Jim Freeze Yeah, this it’s gonna be a great conversation. So you have a pretty impressive technical background including a bachelor’s degree, a Master’s degree and a PhD in Engineering, and have worked across both business and the academic world. And from what I’ve read, you’re also what I would characterize as a tech evangelist and, in particular, an advocate for women in technology. Can you talk a little bit about your passion for the industry and how that shaped your career path?
Ella Hilal That’s a great question and it’s actually honestly sometimes we don’t stop and do retrospection on how the journey came about except when we are in conversations like these and I’m always very appreciative when I get questions like these because it makes me pause and reflect. So I came from an academic family. My dad is a professor of Quantum physics.
Jim Freeze Wow runs in the family.
Ella Hilal And growing up, we were always doing some experiments in the house. I remember the first time like oh you can stir sugar in a cup and it dissolves, what does that mean? How does that work? And we were always having these discussions growing up and definitely it seeded something in me.
And I think I was like growing up and seeing how women are positioned, how we sometimes face different types of challenges. And not just women, just different types of underrepresented groups in general in tech industries. I was very appreciative of the environment I grew up in because my dad always created this environment of learning.
So going through academia, I enjoyed learning a lot but I also was always aching to apply this knowledge in reality. So I love academia, and I think the biggest thing that I took out of it is not only the technical stuff, it’s also the continuous learning mindset, and the fact that you learn how to learn. And this is what I say about the superpower of the PhD is that you need to step back and learn how to learn, and how to aim effectively into the things that need to be done. And taking that and applying it into industry actually did serve me very very well. And honestly you don’t need to have a PhD to apply these lessons, this will work very well for anybody. I think what triggered it for me was like my academic PhD degree, but it can be triggered for anybody by always going back to first principles, or always going back into thinking in first, second and third order thinking. What’s the impact of a decision that I’m making at this point of time? What’s the confounding variables of the decision that I’m making? And just thinking this way helped me anchor my thinking, and definitely served me really well as I leapt into the industry.
And then when I started working in the industry and started applying all of this, there is a massive satisfaction that you get from building things and seeing the impact of what you build immediately. Um, and it’s exciting honestly, there’s so much beautiful impact that we’re working towards in different industries. And I got the bug, I loved it. I got excited, I leaned into it.
And I am very keen to make sure that I create the opportunities for other women in different programs and in different initiatives. And not just women, I really want to stress that. Like at Shopify, we’re focused on all underrepresented groups in different regions and in different crafts too. Like the underrepresented groups in engineering and in data science look different, maybe, than underrepresented groups in a different craft like marketing.
And I think one thing that is really important for me as a mother of a twelve year old girl now, as well as somebody who comes from that background that I told you about, is making sure that we have the conversation, making sure that we share the learnings, making sure that we create um not just opportunities for mentorship but opportunities for sponsorship. And there’s a key difference between both. Mentorship is sharing what you learn and just walking away. Sponsorship is helping advocate for others and creating opportunities for others at the tables that you’re at. And this is something dear and near to my heart. So I’m excited about it and hopefully I’ll continue leaning into this for the years to come.
Jim Freeze I Think that’s Fantastic. You know you said something there in your answer that I found really intriguing and I very much relate to it which is the notion of learning how to learn. You know when you said that I just thought about my own experience where I have different degrees and different disciplines and when I went through school going through these different disciplines I found that I did have to learn to learn differently, from a science to when I ended up getting a law degree as well and and I had to learn how to learn differently and I think that’s a really fascinating concept.
Ella Hilal Yeah, totally and I think also like multidisciplinary people tend to go through the same thing: needing to step back. The way you study for law might look different than the way you study for something different, just because of the different way you reason about the different problems you’re seeing right? like. It broadens your perspective, and the fact that it also honestly is very humbling.
Jim Freeze It is.
Ella Hilal Getting into a new thing is like you, you think you know until you learn that you don’t know. And I think it’s humbling but it’s extremely empowering when you say, “you know what, I’m actually gonna dig into this. I really don’t know and I need to go learn it.” And the ability to learn it effectively and fast because you switch multiple domains.
And again I really want to stress that this is not the skill that you only get when you are multidisciplinary or like when you have a PhD. It’s a mindset. It’s a growth mindset. The ability to know, okay, when do I really not know, when do I need to check my assumptions here, when am I coming in with things that I need to push back on, when do I need to go back to first principles? And if you ask yourself these questions effectively and frequently enough you’re gonna find yourself always willing to go dig deeper and peel the layers of the onion of whatever is there and that will just make you understand it better and will help your brain pick that type of thinking faster and faster over time.
Jim Freeze That’s great counsel for all of us. So thank you for that. So let’s talk a little bit more about Shopify. Today, you’re the Vice President of Data Science at Shopify, and as I understand it, your team is focused on empowering merchants with data-driven products. For our listeners, can you share a little bit more about what Shopify does and what your team is specifically focused on?
Ella Hilal Definitely and I’m happy to. So let me tell you a little bit about Shopify, and then let me tell you about why this is so energizing for me personally.
So Shopify has the mission of empowering entrepreneurs to make commerce better for everybody. And that’s the mission of empowering entrepreneurs to come build their dreams and just live their financial independence. And this is very dear and near to my heart.
When I was at the university, I had a friend of mine who we were both in a graduate degree program. And she got pregnant. And she did have her baby but her baby was not so healthy. She needed to step back from her degree. And a single mom at the time, she needed to do a side hustle. But with her baby not feeling so well she couldn’t work regular hours because all mornings she was in the hospitals and like doctor checks and she started at a Shopify store at night selling accessories that she used to make at night when the baby slept. And that was that and that was actually how she survived the few months that she needed to prioritize the baby’s health and the doctor checks and stuff like that. And after some time she was able to go back to her degree and finish it and um and she’s doing awesome and the baby’s do good and the baby now is doing amazing too. Thank goodness.
ButI just want to stress how much Shopify at this moment of time empowered her to do what she needed to do to take care of her family, but she also was able to have some financial independence because of that. And Shopify is doing this for millions of merchants around the world. Whether it’s somebody who’s a small mom and pop shop or like big merchants around the world. They can come on Shopify, they start their business and they can grow over time. So the mission of Shopify itself is very dear and near to my heart for my very personal journey.
So that’s what Shopify does, and Shopify is honestly extremely merchant-focused because like we are believers in empowering the rebels, empowering the independent entrepreneurs to grow in an economy where it’s leaning into consolidation. There is a power to come from everybody to be able to live their dream and start their business and build cool stuff that they’re excited about whatever it is.
Jim Freeze That’s great. And I love the word empowerment because that actually gets right at the heart of my next question which is: so Shopify offers a pretty wide range of data-driven features from personalized onboarding to a business name generator, which I think is a really cool idea having gone through the process of naming products and companies in the past, it’s a lot harder than you think.
I recently read an article about another feature, Shopify Capital, which offers funding to eligible merchants. Can you talk a little bit about that, and share a few examples of your favorite features and how they support Shopify’s broader efforts to empower, as you’ve said, merchants.
Ella Hilal I definitely and I’m happy to. As I mentioned, we are merchant-obsessed. Our CEO always talks about this, how can we make sure that we remove barriers for more entrepreneurs to go build their dreams, right? And one of the biggest challenges that merchants face is getting loans to start their businesses. And definitely the traditional loan path is we know that there’s a lot of friction on it especially when you’re a small entrepreneur, you’re just trying to start, you’re doing this as a side hustle next to your job and you have kids and you have life, right?
So Shopify at this moment took a step back and thought about, how can we build this? And one of the key things that we thought about is, oh we have a lot of data, we can look into that. So the team went and built this amazing capital offering product that looks at a lot of different signals. This one is really interesting because it doesn’t look at demographic signals. It looks at shopper behavior signals and other things that can predict the shop’s performance. And from there, we started issuing different levels of capital, all the way to significant amounts to merchants to get like the inventory they need, or to get the initial parts of the product they need.
And it’s fully driven by machine learning models and data and stats and it’s a great example of a data product that powers entrepreneurship and leads to this financial independence that I was talking about.
Another one that is also dear and near to my heart is um is onboarding. When you are coming into Shopify, your first task is: start a business. Like for most people not for everybody. But for most people. And I don’t think that’s a small task by any means, it’s actually pretty hard.
Jim Freeze No, it is. You’re absolutely right.
Ella Hilal So having a really smart intelligent assistant that customizes for you what your next best action is at this moment of time based on what that we know about you, what type of store you want to start or what type of business you want to do. If you don’t have, for example, inventory, if you’re not a maker, you can source products from someone else, right? So we have a recommendation engine to become your assistant to drive for you the best next action. And that’s one of the things that is dear and near to my heart because that one is in a moment of like oh my goodness, what am I doing now? What’s my next right step? And having this intelligence embedded in the right place is really essential and to remove the barriers if we really want to empower merchants. I can go on and on, we have a lot.
Jim Freeze That’s great. Ah, no, you touched on some really good ones. So I want to get to this notion of data-informed. So you’ve kind of talked a little bit about how data is embedded across the company and so all departments feel empowered to tap into the value of data science. How does that work and what does data-informed really mean?
Ella Hilal Yeah, so this is something that is beautifully positioned by our CEO, Toby. So he always talks about the difference between data-driven and data-informed. So data-driven means that you follow the data and what the data says blindly. And if we are trying to build revolutionary products to empower entrepreneurs in a time when they have a lot of headwinds, we need to be data-informed. And that means that the data gives you information to give the merchant an unfair advantage based on the scale and the unique data set that Shopify has to set you up for success, or to set the product managers up for success.
And it’s a really interesting concept because, you don’t have to follow the data, but having all of this data available in a digestible way at your fingertips helps you definitely move things further and faster. And for us to build something like that, the data scientist needs to understand the business domain effectively to be able to help power data-informed decision making. And this is why we have embedded data science team across the different organizations and making sure that we work very closely with the different groups, and we are at the table. So data science is definitely at the table with PMs, UX, and we’re able to bring in the different insights and this is how we become data-informed. We pull data in, we have qualitative research, we have usability testing. And we take all of that and we make decisions based on all of this information, not just what a specific experiment says.
And definitely that’s very empowering for the data scientist, and that actually sets the team up for better success because of the deep understanding of the business domain and what does it look like, and what problem are we trying to solve. And hearing directly from merchants is a big thing for us. We all need to listen to merchant interviews or go meet merchants ourselves and this is very important for us to understand the depth of the domain. So the way I try to say it is, data scientists at Shopify are problem solvers. We find a problem and we try to solve it. We’re not just executing on a direction or something like that, we actually own the problem with the multifaceted teams and we work together on it.
Jim Freeze I love that. That’s great how it’s embedded within all the functions of the company. Last question for you: the ecommerce landscape is continuing to evolve obviously on a daily basis. It clearly sounds like Shopify is doing a pretty good job making sense of where it’s headed. With that in mind, how do you see the role of data science and AI continuing to grow or evolve in e-commerce over say the next five or ten years?
Ella Hilal Oh that’s a beautiful question. So definitely we are in a world where personalization is a key for all our experiences. We are in a world where the most limited resources are time and money for most entrepreneurs, and we want to make sure that we use automation and intelligence across the board to help them. And this is not just in ecommerce and data science, it’s actually across the board. You get to see that in every different system or platform you use, you’re expecting it to cater it for you. And if it’s not, your experience is not as good. So definitely personalization is a key, making sure that we have more versions of intelligent assistance or like customized um information is more and more in demand each day.
The next version of that which is really important, and this is where a lot of companies including Shopify are highly focused on is what we call responsible AI. How do you build AI where you’re not tied to the historical models of the past, or pulling in demographics information. or whatever. It’s more about how can we build AI in a responsible, reliable way that helps with better scalability, integrity and reliability across the board, and I think that’s definitely essential.
Jim Freeze Yeah, it’s interesting, that theme of responsible AI has surfaced in many of our podcast episodes. So it’s fantastic to hear that Shopify is very focused on that as well. Thank you so much for joining us today. It’s truly been fascinating to hear about how Shopify is using data to create better experiences for your merchants and ultimately all of their customers. So thank you once again for joining us today.
Ella Hilal
Well thanks for having me. It’s a pleasure.
[UPBEAT MUSIC]
That’s all for today’s episode of The ConversAItion. Join us next time as we sit down with Sidney Madison Prescott, Global Head of Intelligent Automation at Spotify. She’ll tell us all about the role AI plays behind the scenes of the popular music and podcasting platform, and how she empowers all of Spotify’s workers to identify new areas for automation.
This episode of The ConversAItion podcast was produced by Interactions, a conversational AI company. I’m Jim Freeze, signing off, and we’ll see you next time.
[UPBEAT MUSIC]
