Crystal finds joy in using big data to develop business strategies and promote growth opportunities. Born and raised in the US, she moved to Indonesia to join Gojek in 2015 and helped drive the unicorn to its success today. She was in charge of data collection and analytics, but also picked up skills involving HR, marketing, and investor relations.
Earlier this year, she left the company for a new adventure. She is now scouting and building up notable startups through her role in Sequoia Capital (San Francisco), and contributing her knowledge at Reforge, which designs e-learning programs on data, engineering, marketing, and growth for experienced professionals. Miles away from Indonesia, Crystal continues to run Generation Girl, a non-profit organization that introduces young Indonesian girls to tech and science through fun and educational holiday clubs.
This interview has been edited for brevity and clarity.
KrASIA (Kr): What ignited your passion for data science and what keeps you going?
Crystal Widjaja (CW): Growing up in an Asian family, my parents had always wanted me to be obedient and to avoid asking too many questions. That left me a lot of time to think and observe my surroundings, especially in the formative years. So it really started out with me being an observant kid.
After high school, I joined the Carnegie Foundation for an internship program, where we focused on education-related research. The research process led to my first exposure to data and even Excel. I helped with basic tasks such as data entry and formatting the math problems that we were researching on.
One day I made a fatal mistake one day by dumping a large amount of data into a pie chart. Later on I was told that there were better ways to visualize such data. So I learned my first major lesson: when presenting data and findings, never use pie charts. And there are tons of visualization tools to use for a meaningful presentation.
I find it super fascinating that data can tell a story in its own way, objective and impartial. So my passion for data science really began with this simple story. And it continues to grow as I work more and learn more.
Kr: You’ve received so much recognition at a young age. People have noticed your contributions to Gojek’s immense growth through data science; you were awarded Forbes Indonesia‘s 30 under 30 and HerWorld’s Woman of the Year. What do these mean to you personally?
CW: It’s just insanely humbling. As the first data hire at Gojek, I did help to build the team and part of the company culture. But I would say those were very early days and there have been a lot more built upon my contributions.
In fact, during my final year in college, I did research on the Internet and got exposed to what the world offers. Then I laughed at myself, because I used to think I knew a lot when in reality I was barely scratching the surface.
Recently, I chatted with some people from Reddit. And after viewing my post history, they asked, “Wow, back then you were posting questions like what’s SQL and what’s a database, but you somehow became the senior vice president of data at Gojek! How is that possible?”
But really, I am glad people recognize this as a fact. For young people especially, it’s important to understand that you don’t have to know everything at the start. There’s nothing to be afraid of, for I honestly felt like I didn’t know what I was doing half the time at Gojek. But it’s more important to learn fast and apply fast, and you will become someone who creates a great impact later on.
Kr: During your time in Gojek (2015 – 2020), business grew from 20k orders per day to 5m, and the team grew from 0 to 100 people in business intelligence alone. In your opinion, what made it possible?
CW: Gojek was a small company back then. I only hired three people early on to help build the data warehouse and infrastructure that were needed for the company to move forward quickly. That meant I had to pick my candidates wisely, and the hiring process was crucial to our success.
Another thing is that we were working in a lean team, so there was just not enough time to evaluate and approve detailed plans. I had to place so much trust in the competence and work ethics of my colleagues. We all had to trust one another because we knew there was no time to double check anyone’s work.
The last point is we always had a plan. To me, a vague plan is better than no plan at all. We knew that external factors were unpredictable and even internal processes could change easily over time, but we always set a direction for the team. We needed to agree on a common goal, and what it takes to achieve the goal, so that we could put faith into the business and be assured that it would scale eventually.
Kr: What are the good and bad points of Indonesia’s data landscape at present?
CW: I think Indonesia has an advantage in entering the data scene relatively late. That means we are able to adopt the newest and most novel technologies. Many countries are stuck in legacy systems, with a fixed idea of their data landscape and roadmap, which could be outdated. Conversely, here we are able to implement the latest technologies right away because there’s so little to replace and rebuild.
The bad part is that we lack awareness of the latest technologies, and how to pick them out strategically. Even for a trained data expert like myself, I am constantly looking to improve my level of awareness. For example, if I were to redo things at Gojek, I would now incorporate a data build tool (DBT) that transforms data in Gojek’s warehouses simply by writing select statements — instead of working on a full Extract, Transform, Load (ETL) pipeline that would require more time and effort. So I think there needs to be strong awareness around the ever-changing, ever-improving data landscape.
For that to happen, we need to foster a strong data culture. There aren’t many data experts in Southeast Asia in general. While I was at Gojek, I would often reach out to data experts in the US and Silicon Valley — people who are deep within the heart of tech, and try to pass that knowledge back into Indonesia, because I do think we are missing a lot of those connections. So we really need such density of talent and expertise, so that people are encouraged to learn from one another and inspired to work towards success.
Indonesia is still behind many other countries in terms of access to quality education, and this is another impediment to our progress in the data scene. As mentioned, in the early stage of my career, I learned by asking questions on forums, reading Internet threads, and watching YouTube. But I am privileged because I studied English as a first language. Too many Indonesians cannot use English well, or even if they do, they may not have the capacity to read detailed technical documentation and understand exactly the nuances of how to implement a service. So I think education in English and increased access to educational materials would be beneficial.
Kr: Can you tell us more about Generation Girl?
Generation Girl was born out of a collective experience that several founders, including myself, felt strongly about being misunderstood and underestimated, and not having the right support for being women in STEM (science, tech, engineering, and math).
We want to promote the learning of STEM as an opportunity, for young girls who have not tried it before. For those who have tried it and are keen to continue, we also want to provide the right space and support community for them to achieve greater heights. Hence our summer club offers three different classes: beginner, intermediate, and advanced. Young girls can choose which class they want to join, based on existing ability.
Whether she is new to coding, has coded a little but lacks confidence, or is relatively well-versed in codes, we hope to make everyone feel comfortable to give it a shot in this environment. We make it safe and easy for people to fail and try again. Even if a girl joins the program and later decides that STEM is not for her, it is fine, because it is her own decision rather than a decision made by someone else or society in general. In the long term, our goal is to normalize STEM for women and empower them to make their own decisions. We’re proud that this program has expanded from an initial intake of 30 girls to 1000 girls today.
Kr: Last question, can you sum up 2020 for us? What‘s next from Crystal Widjaja?
CW: For the year 2020, I’m really trying to reflect and acknowledge my small wins. I’m happy to be working alongside incredible people at Sequoia, helping a group of pre-founders to ideate and create something beautiful. At Reforge, we’re working with five amazing companies, one of which has just raised 10 million in series A funding. I’ve been learning very much from people like Bangaly Kaba (head of growth at Instagram) and Brian Balfour (head of growth at HubSpot). All of these things make me feel proud.
As for what’s next, I honestly don’t know and I think that’s okay. What I do know, however, is that I would continue to pursue my goals in this direction. I’m constantly trying to find ways to be impactful on a larger scale. I still get messages from dozens of people a day and as much as I want to, it’s impossible to help each and every one of them grow their companies personally. So I am still thinking of ways to scale my impact and efforts, so that many people can benefit from my knowledge and materials on data. I’ll also figure out how to provide better access to tech and opportunities for the tech community to grow in Indonesia.