On this episode of The Artists of Data Science, we get a chance to hear from Brenda Hali, a marketing guru turned data scientist who is passionate about using data to understand causation and to promote company growth. She gives insight into how she broke into the data science field, how marketing and data science are related in some ways, and the struggles she faced when breaking into tech.
Brenda shares with us her transition from marketing into data science, along with the importance of having the representation of women and other minorities in the tech industry. This episode really shows why diversity and inclusion in tech is so important, and how we can all play a role to help others break into the field.
Some notable segments from the show
[6:56] What marketers can learn from data scientists
[11:07] Steps to take when beginning a new project
[17:33] How to communicate effectively with your team in the post-COVID world
[20:56] Advice for women and minorities that want to enter into data science
Brenda’s journey into data science
Brenda heard of data science about six years ago, even though she did not have a background in tech. She had always been interested in learning more about coding, and so she decided to learn how to code on her own, by watching video tutorials and practicing at home.
Brenda took the leap into data science when she participated in a program that helped entrepreneurs from Latin America. She was a part of a small team, and they had gathered data from over 20,000 people! They needed to find trends in the data, but managing such a large data set was difficult to do.
Brenda was very passionate about this program, since she grew up in Mexico and lived across many latin american countries. At that moment, she decided that she needed to find better ways of analyzing data.
[2:25] “I think that I have heard data science a couple of years ago, maybe like six years ago. And I mean, even though my background is not exactly in tech, I didn't study anything related to software or taking undergrad. I've been learning by myself. I like--through YouTube, through tutorials. Like I learned how to do--how to program in front and development. I learned how to build apps. And I learned--I even created like a bot for social media because I wanted to follow some certain hashtags. But I never did that as a formal education. And in that curiosity, I was in my tech conference because I love conferences and I love the media people. And I was amazed about all the possibilities when I heard that the term big data, but that was back in 2013 and I didn't know exactly how I could transition into data science. After that, I think that the moment when I decided to make for real that transition to data science it was because I was working in this program for helping entrepreneurs from Latin America, and it was a White House initiative. It was Obama's White House initiative, and we have all of these data from 500 entrepreneurs that we needed to find. We have several datasheets like from survey, CV formation, application. Over twenty thousand people applied, and we needed to look for some trends. We had a really small team and they were mostly inclining to political science, public policy, but not really into tech. So I came to the team and I literally show them how to use people tables. So the problem that we had at that moment was that when we were transitioning from one White House to another White House, a way how programs are a way that are in a different sense sometimes before you go. You could find like a couple of stories more and just share those stories like this is changing the world, and this is changing the U.S. as well. But in this case, they want that number. So they wanted to measure that impact. And we launch a survey and then we have all these responses from entrepreneurs. We don't know how to analyze all of the data that we have. It was a lot of data, a lot of rows. We didn't know even how to properly manage, like to read it completely or to find trends. And I remember at that moment, for my team, my team, they were mostly from the US. I'm from from Mexico, so for me, I'm from Mexico and I live in a couple of places in South America, like across Latin America. For me, that was like really close to my heart because I was seeing the impact of this program. I didn't want the program to finish. So in that moment, I basically told myself, I need to do something about these. I need to know how to analyze data because there must be a better way than just reading or printing a bunch of things and analyzing it in that way. So in that moment is when I decided to really look for that transition into data science. And it took me a couple of other years to actually act on it.”
How do you see data science affecting marketing in 2-5 years?
Brenda predicts that marketing teams are going to focus on hiring marketing data scientists, instead of just hiring creative professionals. Teams that do this will have an advantage of having someone that can track the data and analyze it.
Furthermore, Brenda sees automation being on the rise within the next few years. She thinks that small businesses will also need data scientists to be more effective as well, since a majority of the data that is collected by these businesses is “dark data”, or data that is not being used.
Lastly, Brenda thinks that we are going to see faster GPUs, which in turn will lead to running algorithms much faster.
[9:01] “Well, I see how possibly every marketing team now is going to have a data, a marketing data scientists in their team more than a data analyst. Like someone that really understand how to track the data. So I see in the team part to hire more data scientists. That's one thing that I've seen because before these data, these marketing team, it was more like creative people. But now I'm seeing that they're hiring more people that are like with a major in math, with a major economic side. You can say like why someone that is majoring in math is in marketing if marketing is a creative field. No, like everything should be based on numbers. So I'm seeing that field into that. I'm definitely more automation, but way more automation. And right now, 93% of the data that a small business have go to dark data. That means that that data is not used. So I can see also how even the small businesses in their marketing teams will start acquiring or will start hiring people in data science to use their data possibly. And I see like in the five years, probably most of the business will be generating big data, so that's where I'm seeing. Besides that, everything is going to be faster because of the use of GPUs, GPUs are getting cheaper. And that means that, for example, if before you run an algorithm to predict something and it took eight hours, now it is going to take 50 minutes. So that process is getting faster. So probably we're going to reach the point in which everything that's going to be like close to real time with big data and that with marketing data and with all the softwares is just gonna get crazy. Well, I love it.”
Key takeaways from the episode
What data science and marketing can learn from each other
[6:56] Marketers might not be aware of the tools that exist that can help them understand the trends in data regarding impact. This is where data science can help marketing. Tools exist that can measure what music is catchy, and will therefore generate more money.
Steps taken during the beginning of a new project
1.) Understand the type of project you are working on. How does this project help the organization?
2.) Communicate clearly with the organization, to receive the right feedback.
[17:33] Be comfortable communicating with your manager and team members openly during this time. Even though you might not be in the office together anymore, make sure your team uses project management tools that optimize the communication between team members. Also, if you have some free time, take some classes to advance your career, and communicate this with your manager. Now is the time to learn and grow.
Breaking into tech is a woman
[20:56] It is difficult breaking into a field where you are not being represented. If you see yourself being represented, then it allows others like you to believe they can also achieve success. My advice to women is to find a community that is going through the same struggles as you, and find a mentor that can guide you.
The four things women need to make tech their next big success
Don’t be afraid to explore. You don’t want to have a fixed mindset into your career, since you will be doing it for the rest of your life.
Trust your plan.
If you need to, delegate tasks.
Be comfortable being the minority in the room.
# Memorable quotes
[15:02] “...you need to have communication with your team, and that communication needs to be in one place”
[15:47] “...experiment fast and let things go…”
[23:52] “Be careful with who you listen to, and be careful when those voices are close to you.”
The one thing that Brenda wants you to learn from her story
[29:43] Don’t be afraid to explore and find your passion. You have a long professional career ahead of you, so don’t be afraid to look for your calling. Also, never stop learning.
From the lightning round
Brenda's data science superpower
Brenda has a knack for growing, she's a data gardener! Her superpower is that she can start something and grow it.
The best advice Brenda has ever received
Brenda got a very insightful bit of advice, which I absolutely love - people care about themselves. They don’t care about you. We often think that people think about us way more than we think they do - but the truth is, nobody is thinking about you. They're too busy thinking about themselves.
What motivates Brenda
Brenda is the first one in her family to go to college and become a professional in tech.
Now that she has accomplished some amazing things academically and professionally she is motivated by the opportunity to be a role model to her siblings and cousins, and show them that there is a place for them in tech.
The advice that Brenda would give to her 20 year-old self
Change your career path.
Topic outside of data science we should study:
Nudge by Cass Sunstein and Richard Thaler
Data science bias blunder:
Instagram - block photos of overweight women in bikinis.
How to connect with Brenda