Yang Sui, CNN Digitals director of product management, shares his path to news product
This is Tony Elkins for the News Product Alliance and welcome back to Hi . . .
This week I’m happy to introduce you to Yang Sui, director of product management at CNN Digital focusing on personalization. He shares how product led him back to his love of news, how technology and analytics are different forms of storytelling and offers some valuable tips on how to make a transition to product management.
First though, we have some exciting news to share with you about the NPA Slack. On Thursday, July 8, we’re opening our Slack community and inviting anyone who identifies as a news product thinker to join us. If you or someone you know would like to join the NPA Slack, keep an eye on your inbox this Thursday for instructions on how to join.
Now here’s my interview with Yang.
What’s your day-to-day like? What are the jobs-to-be-done as director of product for personalization?
Personalization is relatively new for CNN, and for the traditional publishing and news industry in general. My day-to-day varies, that’s kind of the nature and beauty of being in product — you can be wherever you are needed. It involves setting long term vision and strategy, learning what personalization means for CNN Digital 12-to-18-months from now, what we are doing this quarter, what we are planning to do next quarter and who are the people we’re trying to serve?
In addition, it’s running day-to-day stand-ups, making sure teams are on track and decisions are being made. If we’re launching tests we make sure all the pixels look right and images and text align up. It really varies. I’m also thinking about how we grow our team in the next couple months and to align with a strategy and vision.
The day-to-day varies a lot.
Let’s dive into the personalization aspect a little bit. You came from Spotify and it seems like music platforms have really been at the forefront of personalization and user research. How do you introduce these ideas and types of thinking to legacy companies and publishers?
Personalization is more prevalent in other parts of the technology world, but media is trying to catch up. I was at Spotify for four years and spent a big chunk of time focused on podcasts and podcast recommendation. Part of that is really thinking about how users receive and consume information out of podcasts. What are some of the taste factors and taste buckets users have? Those things do translate, regardless which industry we’re discussing.
The overall mission is really to try to understand users better, understand content better and say, “For this user, these are the content buckets they are really interested in, and want to learn a little bit more about, to be more knowledgeable and well-informed citizens.”
So, how do we build a deeper connection between a user and our digital platform, our content and our talents? That’s what we’re trying to do. Obviously, content varies based on which industry we’re in. News is very different, right? We want to make sure users that are more engaged with our platform, but at the same time there’s an added responsibility for us as a news organization to make sure users are properly informed
How about data and privacy? What are the lines you use between collecting all this data and using it and making sure you’re using it for users’ benefit? Not just to manipulate them into an action.
It’s a key topic in the personalization-recommendation world. Obviously we want to do the right thing. A lot of this has really come from two-way communication. We let users know what we’re trying to do and why we are recommending things to them. I’m making sure it’s within the right context and making sure users understand, “Oh, I’m getting this because I’ve done X, Y, and Z.”
With a good amount of context setting and transparency, I think it can really help make a difference in terms of building trust. It’s all about trust. It’s all about deeper connections. Without trust, it’s really hard to get information from users. We want to be quite cautious about that.
How do you introduce these new products to consumers? What does the product cycle look like?
I would say a lot of news media companies have gone through a similar transformation as CNN has. Part of this is being more savvy about the instrumentation of data, being able to understand users better and being able to do AB testing as frequent, and scientific, as we can.
From the early-stage of product development we don’t have to have a perfect product. Start with user research and concepts, and even low-fidelity prototypes. It’s really key to get user feedback. As we’re going through different development processes we run tons of A/B testing to understand what really resonates with users.
Those usually run from weeks to months, and depend on the development cycle. The goal is to be able to learn faster and have data guide our decision-making so that when we’re ready to launch a product, we know what resonates with users.
Are you finding stakeholders are receptive to the idea of data guiding decision-making?
I’m wonderfully surprised since I joined the digital side of CNN how receptive people are with the idea of data-driven decision-making.
Obviously there is a gap between where media is compared to other tech industries. Part of this transformation is to get people comfortable understanding the work. This is a process. We’re doing Agile. Everything is about learning. There’s no success or failure with a certain task. It’s all about, “Do we gather these learnings fast enough.” I think that’s a mentality shift for a lot of people as well. Understanding it takes time and effort, and many, many rounds of iteration for us to get something right.
You talked about building roadmaps. How do you build a roadmap for such a large company, even if you’ve got a small piece of it?
I think a lot of this really starts with user research. Understand what the user’s pain points are. In the case of personalization, what we’re trying to tackle is users have their own interests and passions. Each user is unique. Everybody needs to know the latest breaking news at the same time, but there’s a desire for users to get a little bit deeper into the topics they care about, the issues they care about, the locations they care about and the community that they really care about. That’s where personalization recommendations come in. We’re really good at understanding what content would match up with users and interests.
In terms of building a roadmap, it’s about understanding what the user’s needs are and what are the steps to get to those user needs. When a user says, “I want to catch up with the latest topic,” what are the different steps to get there? So with those as guiding principles, combined with first party data analysis, we can set up the right metrics and targets to move to that direction.
A lot of the transformation we’re trying to do at CNN Digital is move away from the traditional set-and-done roadmap. It’s more about an agile, iterative, OKR-driven process.
There’s so many different ways to get there. The goal is to really rely on the metrics, with other people in the organization and give the space for engineers on my team to be able to experiment with different things.
It’s not just about, “Hey, I have an idea, let’s go implement it.” It could be the wrong idea. The beauty of having that OKR-driven process is everyone comes in and pitches their idea. “Have you tried this?” “Have we tried that?” Let’s go through the process and see which idea wins.
When you’re trying these new ideas and experiments that require content, how are you approaching editorial teams with these ideas and these tests you want to run?
Working with editorial is absolutely key. I came from a background of computer science and journalism. I went to Columbia and I was one of the first to do a master program they created for computer science and journalism. Most of my career, I’ve been focused on tech, data analytics and product management. This is kind of a homecoming to me — coming to media and news.
For me personally, there’s always a soft spot for the editorial team, for the program team and for the journalists on the ground. I want to make sure whatever we’re doing is aligned with their objectives. I want to make sure we’re doing the right thing to showcase the best piece of work by the talents we have in the newsroom.
Part of that is really about relationship building. Setting meetings with the programming team, with the editorial team to understand what’s working, what’s not working and what they are seeing from their editorial objective prompts. I compare notes and let them know what tasks we’re running and ask, “What are your thoughts and guidance? Does it fit in any sort of a news cycle and trends you are seeing?”
This is getting into the domain of what we call “human in the loop.” I personally don’t believe every story should be personalized. I think a really well-run personalized news feed should have a pretty heavy human-curation component.
You talked about homecoming. Tell us why news is important to you and how you’ve managed to merge these two things, news and technology, that you care about?
I was born and raised in China, and I came to the States for grad school almost more than a decade ago. While in China I was an engineering and computer science major. But, we did have a student run news website in college and I got involved in the community.
It’s the energy I felt in the student newsroom. Everybody got to hang out and formed a very tight community. We would run around and report on the campus, like this little concert happening by this student group. It was very lighthearted student news, but I got this feeling that it’s great to be the first one to know about something, to be able to tell it to people who can really benefit from the information and doing so with a group of people who share a similar mindset.
That’s when I started to realize I’m going to do something with news one day, but I don’t know exactly when and what it would look like. So when I came to Columbia and they launched this dual master program in journalism and computer science, I thought, “This is it.”
I got in and was really happy. I worked alongside people and professors, some of them were even on the judging panel for the Pulitzers. It was a very humble experience for someone who just came to the country. My English wasn’t that good. I just tried to write up and report as much as I could. It was just a great experience and great community.
Through that experience I realized I really enjoy storytelling. I knew I wanted to work in technology because that’s my foundation, and I wasn’t quite ready to work in news just yet, partly because of the language barrier. But, I did want to bring in my unique contribution, from a technology perspective, to the newsroom one day.
Early on, most of my career has been data analytics. The way I thought about it was storytelling with data — storytelling with insights that guide strategy. I finally managed to find my way back to news and meaningfully contribute to the newsroom from a technology lens.
What were some of the ways you built your skill sets up to get to this point?
A lot of it happened at Spotify. I joined initially as a data solutions lead. My role back then was being the insights and analytics lead for some of the highest prioritized initiatives. I was lucky enough to be part of the push into the audio space, meaning podcasts.
That was a glimpse into podcasts, news and information, and it was the start of that journey. Through that I discovered product management. I realized I really enjoy owning a space from start to end, rather than just constrained in the analytics.
Knowing that that’s what I want to do, I made a decision. I wanted to make that transition. This is something I really want to share with other people. Everyone comes to product management in so many different ways. There’s no set path or recipe for someone to get into PM. You ask 10 people, there’s 10 different journeys.
It’s all about taking initiatives, big and small, that you can start to move in that direction. That’s what I did. I asked around the company, letting people know, “This is what I do. This is what I want to do. If you see anything, let me know.” There are opportunities if teams need an interim product manager. There’s no guarantee it’s going to land a full-time job in the end, but if it’s two months, it’s two months, I’m going to take the two months. I’m going to train myself and be ready when there’s a role open up.
And that’s what I did. I took on a couple of those projects, proved myself, brushed up my skills and one day a product lead reached out to me saying, “We have the open role. Do you want to talk to the team? It’s interim to start with, the opportunity will be full-time in the end. What do you think?” I said, “Yeah, I’m all for it.”
Fast-forward, I really found PM is my calling. It’s my passion. Being able to work with a group of people who share the same goal, are really passionate about the space, bring the best ideas forward, work really well together to benefit users — all that comes together kind of lead to where I am today.
How do you see product changing news over the next few years? How do you see product driving news for the actual business and consumers?
Part of this is really about media as a business model and how it’s transformed in the past few years. We’ve seen a transformation from click or impression-based to direct to consumer. It’s about scale, and at the same time, do you have brand recognition? Do you have brand loyalty? Do you have people who love and care about your product and come to your platform every single day? Do they have the connection with the talents and the content that you have?
Being able to go through that mind shift, it’s huge. That’s where product management really comes in. If we put the user as the center of the question, and really understand who our users are, they’re more engaged and better informed. This is where this mindset will help to transform. How do we produce the stories? How do we package the stories? How do we program stories and where do we put them? How do we package them? What is the most appealing way from the user’s perspective? All that is going to have a big impact on the user.
And there are things that shouldn’t change, like journalism and integrity. It’s about knowing what’s important and reporting on those things. We shouldn’t change that.
For someone who wants to understand product more, or wants to make a transition into product, what’s your advice for them?
So one of the things I like to do, and I probably shouldn’t broadcast this, I like to LinkedIn stalk people. Just to understand their individual journey and how they got there. Again, there’s no set recipe because everybody’s journey is so different, but taking the hints of, Hey, that worked for Person A and something worked for Person B, maybe this is a next step I should take given the situation I’m in. That’s number one.
Number two is don’t be afraid to take risks. It doesn’t have to be the step you’re taking. It doesn’t have to be the final step. Understand this is a marathon, not a sprint. So every step is moving a little bit closer to where you want to be. Come home and celebrate and you did a good thing.
And the third thing is, just be curious about product in general. Open up random apps and websites. Put yourself in the shoes of a product manager. What are the things you want to understand? What are the things you might want to change? Really start training yourself to go to that mindset. It’s super important to have product curiosity.
If you found this via Twitter or someone forwarded it to you, be sure to sign up for the newsletter at newsproduct.org. If you’re interested in sharing your path to product I would love to talk to you. You can fill out this form or reach out to me on Twitter.
Talk to you soon,
Tony Elkins,
NPA Creative Director, Gannett Director of Innovation
