Approaching D&I With Andrea Corey, VP Product Development at Nudge.ai
We recently spoke with Andrea Corey, VP of Product Development at Nudge.ai. Corey previously worked at Eloqua (later acquired by Oracle) where she performed various roles including VP of Systems Architecture, VP of Quality and Performance Engineering, and VP of Quality Assurance.
Corey is also an Activator with SheEO Inc. Corey has had immense experience in tech and has considered diversity and inclusion throughout her career.
Read on for Corey’s tips about:
Improving the recruitment process
Avoiding filtering out potential talent
The importance of reaching more candidate pools
Conducting interviews more effectively
Building better teams
Why soft skills are so important
How diverse and inclusive teams translate into better products
The potential for bias in AI
Implementing diversity and inclusion efforts at your organization
Interview with Andrea Corey
Stefan: Can you share a bit more about yourself and how you got to where you are now?
Andrea: After studying Engineering and Physics at Queen’s University I did custom software development projects at a couple of different companies. Then I was at Eloqua which was started around six months before I joined. I spent about 13 years there! I was lucky to work with great people, and it was a really different experience at a product company rather than custom projects. Near the end of my time there, I helped the team transition over to Oracle. After taking some time off, I started at Nudge.ai, where I've been for 4 and a half years now. It's been quite a journey for me!
Considering Importance of Diversity and Inclusion Early On
“The more I learned about diversity and inclusion, it became something that permeated in so many different dimensions beyond my initial goal - we ended up with a really diverse team in many ways, beyond gender.”
Stefan: What's the importance of diversity and inclusion in your role as VP of Product Development?
Andrea: I've been on teams where there are around 50 men and 3 women including myself so it was always something I was aware of, but I really started thinking about diversity and inclusion when I was at Oracle. At the time, we wanted to build our quality assurance team size and we needed people to do technical testing and non-technical testing. I didn't want the technical positions to be filled by men and the non-technical roles to be predominantly women.
I dug into learning as much as I could about interview and recruitment processes with the goal to get more women into technical roles. The more I learned about diversity and inclusion, it became something that permeated in so many different dimensions beyond my initial goal - we ended up with a really diverse team in many ways, beyond gender.
In my current role, I want diversity and inclusion to be part of the process from the very beginning, and not just something to think about when we are hiring. It's really important to look beyond what might be occurring at face value; if there are less women applying to certain roles, don't just assume that you can't hire women because of that. You have to scratch beneath the surface as to why less women might be applying.
Take that and extend it to thinking about how to bring in different types of people onto your team, and how to ensure individuals of particular groups aren't self-excluding themselves from applying. How do we make the process fair for all candidates? All of these complexities have to be thought about and explored.
Improving the Recruitment Process
“I really like to turn the interview process on a 180 and think about it from the perspective of the potential employee.”
Stefan: Do you have some pointers for someone going through that process?
Andrea: I have used and come up with some concepts with concrete steps. I think clear job requirements, a non-traditional interview approach, and thinking about this process from the viewpoint of the employee are all crucial. To explain further:
Crafting clear job requirements
When you are overly-specific in your job requirements, you end up filtering out a lot of potential talent. Many women in particular might self-exclude themselves from applying to jobs where they don't meet every single criterion. You're really hindering the process right at the beginning by unnecessarily limiting your applicant pool.
Tech is always changing and evolving and unless you have a very specific need to fill, there isn't much benefit to being so strict with your requirements in relation to years of experience or coding languages. If someone knows a coding language, it is not difficult to pick up another one because of their skillset.
So instead, think about what you really want to hone in on. I look at the soft skills that people can bring to the table. Their individual capabilities as well as eagerness and willingness to learn are more important to me than meeting strict requirements.
Have an open, flexible approach as an interviewer
Move away from the daunting, authoritative approach. Find ways to make the process more comfortable for candidates and think beyond traditional interview questions. You’re not there to stump them. I really like to turn the interview process on a 180 and think about it from the perspective of the potential employee.
Think about how an interviewee can best showcase their own skills. Be open to a flexible process and empower your candidates in the process. You might even consider asking them what they think is the best representation of their skills or talents.
Think about what a potential employee is looking for in your company
The entire focus is not only on interviewees putting their best foot forward - this is an opportunity for you to represent your organization. How will a candidate know that your company is the right fit for them? How will they know if it is a comfortable atmosphere? What is your workplace culture like? Think about what they want to get out of a position with your organization and how you can speak to that.
Building Collaborative Teams
“You’re going through a process in creating a product where everyone deserves to be heard and has something worthy to add.”
Stefan: In your view, why is it important to have empathy and soft skills in tech?
Andrea: Open dialogue and collaboration is so important. When creating a product, there's so many directions you could go in. However, how can you get through that process and come up with the best solutions if the team isn't comfortable working together? There needs to be ongoing communication within the team and with users to create the best possible user experience.
When that sort of discussion isn't valued, you run into a lot of problems with your product. You're investing a huge amount of time and resources into building something - you can't do that without being receptive to others.
One of the core aspects of being successful as an up-and-coming startup is understanding your users and being able to take the feedback and making ongoing improvements. The other part of that is that everyone on the team should feel comfortable speaking up, and their viewpoints should be respected. You’re going through a process in creating a product where everyone deserves to be heard and has something worthy to add.
Avoiding Bias In AI
Stefan: Nudge.ai is a big player in sales and the AI space. When it comes to the potential dangers of biased systems, how do you safeguard against that happening?
Andrea: Right now, we are doing a lot of Natural Language Processing (NLP) where we are taking in the body of written content on the web to accumulate data about which companies are in the news and which people are being mentioned. We then take this published content and match up the people mentioned to the companies. Based on our product needs and user needs, we make choices about what content sources to use, and there is the potential for bias in that regard. Ultimately, a team has to consider the issue of bias at all stages along the way and work to reduce bias in ways that negatively affect users.
A good way to start thinking about that is considering it as ‘bias in, bias out’ - that is to say, if the data you’re relying on is biased, the outcome of your model will be biased too. It depends on the type of data you are relying on, the algorithms, and the models that you are building. A typical example would be using biased crime data and then coming up with a prediction of crime levels based off of that. You can’t just accept the results without thinking critically about what you are relying upon at every step along the way.
Recommender systems is where it’s very relevant. If you're taking a set of data and recommending certain people certain things based on their interests or other facts, there is a lot of potential for bias there.
Another issue could be the public lists we create. To illustrate, consider an instance where one of the lists is about people in the startup community, and the makeup of the list itself is very homogenous. If we are then relying on that homogenous list and using it to recommend further lists, there are problematic implications there. Again, bias at all stages in the process is definitely something to stay aware of and something we will have to continue to wrestle with.
Getting Started on Diversity and Inclusion Efforts in Your Organization
Stefan: Do you have any last tips for organizations starting out with diversity and inclusion efforts?
Andrea: I would say think about it as early as possible! In more detail, I’d recommend:
Think about ways to engage your team around the topic
Don’t be afraid of open discussions. We talk about what diversity and inclusion means to us and always learn something new from each other. Your team can open your mind to new views and different angles you didn’t necessarily think about before.
Consider doing surveys
We do surveys to gauge our employees and understand them better. Right now, with a smaller team we do them about every 18 months. As our team continues to grow we will do surveys on an annual basis. It gives a chance to think about our company culture, whether people feel included, and how we can continue to improve our efforts.
Get to know your team beyond their credentials
It’s important to understand employees as individuals rather than just being aware of their professional capabilities. We get to know our team’s hobbies and interests to understand how unique we each are and also to diversify our teams. When you have 10 people that are all very similar to one another building a product, the result will not speak to a broad group of users or their user experience. Diverse teams build better products.
Thank you for taking the time to speak with us Andrea!
Thanks for reading! If you're interested in more resources like this you can sign up to our mailing list, or if you've got advice/experiences that you'd like to share - we'd love to hear from you!
Crescendo is the diversity education app for Slack and we’re on a mission to help companies create more inclusive workplaces. You can learn more here!