Andy Chen is Chief Product Officer at Glassdoor. After starting his career as an aerospace engineer, Andy got into tech by working at online travel companies such as Sabre Corporation and Travelocity. He later joined product leadership at Match Group, the parent company of popular online dating services like Tinder and Hinge, before coming to Glassdoor.
In our conversation, Andy talks about being drawn to the ambiguous, especially when it means getting behind an idea that only a few people believe in. He also discusses Glassdoor’s community building initiative, which posits that conversations generate richer insights than simply reading one-sided reviews and salary information. Andy also walks us through Glassdoor’s organizational structure and how they intentionally place one stakeholder from the “four-legged stool” — product, product design, engineering, and data — on every product team.
I look for industries where tech has enabled winds of change. I look for analogs or patterns in one industry where there’s success, and I see if it applies to others.
For me specifically, I’ve found that good ideas that everybody believes in, I’m not super interested in. But good ideas that very few people believe in, that’s where my brain goes. I immediately think “Are there going to be a lot of doubters?” That’s typically around industries that have dogmatic or institutionalized thinking — ones where there’s a natural resistance to change.
For example, I started my career in the consumer internet on online travel. I really liked the people I was working with and the work I was doing, but I got the opportunity to join an online dating company, True.com. This is a while back when online dating was still super taboo. All the reasons I heard why people didn’t want to use it didn’t make sense to me. I knew if people were spending time researching the internet for where they’re going to travel, they could also use it to find who they’re going to spend their life with.
That’s just an example, but I just knew that it was going to happen. I knew because my mom and dad were like, “Wait, you’re going to go work for an online dating company? Are you crazy?” And I took that as a good signal. If my parents think it’s crazy, and I really believe in the idea, then I’m even more interested.
I was at Match Group for almost 12 years, and in the last three years, I felt like I couldn’t drive innovation as strongly because I’d been there too long. When the Glassdoor folks reached out to me, I realized that the analogs between online dating and the job search are actually pretty insane. If you think about it, the resume was the original online profile.
I have a little bit of an activist streak in me, and the whole call to transparency, giving workers a voice, I just loved that. I knew after spending time helping people find love, which was a very noble cause, I needed to do something that was mission-driven. And I really liked Glassdoor’s vision around transparency in the workplace.
What made it a no-brainer was my first interview with Christian, our CEO. He talked about the idea of transforming Glassdoor into a community. We’re the leading product for reviews, company ratings, and salary information, but that’s all one-way Web 1.0 information. We have half a billion users that come to Glassdoor every year, but they only come once or twice a year.
We have all these people coming, but they’re just not sticking, hence the idea of using community to drive more engagement. We’re at 40–50 million MAU; how are we going to get to 100 million MAU? Community is, we think, the right way to do it. That’s what drew me to Glassdoor.
If you think about what Glassdoor is, we’re an insights company. Our insights right now are reviews and salary information. Community is just another way to deliver insights. So, instead of just reading a rating, review, or one of our machine learning-driven salary products, at the end of the day, it’s just you reading. We believe we can enrich our insights with conversation from our community. That’s the existing job to be done, providing insights about the workplace and salary.
In the modern workplace where we’re getting more remote, there is a need for connection. Community provides that. That’s a new use case for Glassdoor — this idea of connection. We have these things called Bowls, which are basically our version of subreddits or groups. You join a Bowl because you want to connect with a certain “professional tribe.”
For example, I belong to a Bowl for product managers. I have the opportunity to express myself, to feel heard, or to read something that somebody else believes. And I connect through that. Our most engaging bowls are Company Bowls where verified employees of a company can engage in discussion with each other and with leadership.
A key mechanic that we employ in our communities is semi-anonymous identity. I can post as Andy Chen, as chief product officer, or I could post as an anonymous employee at Glassdoor. Any Glassdoor employee can go into our Company Bowl and absolutely disagree with a comment our CEO made, or a comment that I made, and safely express their thoughts. And it’s so powerful. Our Company Bowl is an amazing tool for not just leadership, but for our employees to communicate with us and each other. We had a layoff in March and our company Bowl was our biggest insights tool as a leadership team to get a feel for how things are going.
To answer the question of how we differ from LinkedIn, LinkedIn is like an internet-enabled Rolodex. They have a good newsfeed, and I consume a lot of my professional news from LinkedIn. It’s a good way to keep up with mostly celebratory events that my colleagues want to share. It’s a great way to manage your professional reputation, but this idea of real talk doesn’t really happen there. And that’s fine, it doesn’t need to, but our thesis is that with the power of anonymity, we can actually have real talk as professionals. That’s what we endeavor for.
This is how I think about my growth model. I want to go from 40–50 million MAU to 100 million MAU to 200 million MAU. For that to happen, we need to get existing Glassdoor users that come to our site to register for community. To register for community, you have to do a little bit more than how you would register for Glassdoor. You have to go through verification before any content you post shows up on our site, so we need to make sure we can get people to do that. This idea of anonymity is tricky — we have to make sure people are comfortable.
So, registering for community is hurdle number one. Then, once they register and we start sending them community content, they start to engage. They’re coming back, clicking or liking a post, or they’re commenting on posts. The key is, once they’re engaged in our community, will their stickiness and retention increase? That’s the goal. Instead of coming back twice a year, they come back 200 times a year. We’re going to look a lot at retention and frequency of visits.
Generally, we’re doing the things any company should be doing — articulating what the purpose of the company is. I think that’s very important, especially for Glassdoor because we’re so mission-driven. The other part for us that’s unique is one of our values, what we call “good people.” It sounds so cliché, but it’s really true here. For a little bit of transparency, we even say we’re too nice, which is honestly a high-grade problem to have.
I think where we need to get better, and where I’m focusing at least with my team, is burnout. It’s chronic. We have very lofty ambitions and hire the right people who are going to burn themselves out if we don’t have the right process and tools. We just need to make sure when people are working, they don’t feel like they’re inefficient or like they’re having to rework things. Regarding autonomy, if they don’t feel like you’re calling the shots, especially strong A players, they’re not going to stay around.
Here at Glassdoor, we call it the four-legged stool. We literally have a Slack channel called One Team that every single product and engineering person is in. The One Team is made up of these four legs of the stool: product, product design, engineering, and that’s typically where we’d stop, but we’ve added a fourth leg, data.
Every single team that scrums together — what we call “the product team” — has a stakeholder or leader from the four-legged stool. We want all those four teams to have a sense of ownership across the entire product development life cycle. We want them in the room when we’re interviewing users, when we’re grooming the backlog, doing technical grooming, agile ceremonies, etc. Those are the four leaders.
When the product launches, we want every single person on that four-legged stool to be looking at the data. And that’s the ultimate form — we want them to be cheering each other on, so we think it’s one team.
Definitely marketing, our customer care team, our trust and safety team, our legal team, our security team, and our go-to-market team on the B2B side. They’re managing me, our CEO, and all executive stakeholder management. Product is working very closely with our UX research team, market research team, and our business operations team that helps drive our overall corporate strategy. Pretty much anything that you need to check the box before you launch the product, the product managers are handling that.
I don’t really know how important that is anymore. It used to be table stakes — you used to have to go pretty deep as a product manager because the product design role was not as mature. I used to have to wireframe stuff and that’s pretty much a product design function now. If you have a really good product design function, product design is probably inheriting more of the craft of building the product than product management now, especially if you have a user facing product. It’s different if it’s a technology or platform product.
So what does that mean for PMs? In my opinion, if product design is taking on more of the craft of the product and is an equal voice in all the agile ceremonies, I believe the strength of a product manager is product strategy — the ability to tie product to business outcomes.
What makes a really good product manager now is stakeholder management. They’re the person making sure everything we’re doing is the right, highest priority thing. They’re unblocking, driving clarity, connecting the dots, and they’re making sure every stakeholder within a modern tech-enabled company is taken care of. And that’s hard, it requires a deep understanding of so many functions, and I think that’s what a really good product manager is turning into.
One of the things I love is that with every new app that comes out, product managers download it, look at it, and understand what’s new, what’s not working, and what is working. Having that natural curiosity helps, and you can strengthen that by just being curious and intellectually breaking things down. That’s probably more of a trait I’d say I’d want a product manager to have — curiosity about how things work and how they’re built.
LogRocket identifies friction points in the user experience so you can make informed decisions about product and design changes that must happen to hit your goals.
With LogRocket, you can understand the scope of the issues affecting your product and prioritize the changes that need to be made. LogRocket simplifies workflows by allowing Engineering, Product, UX, and Design teams to work from the same data as you, eliminating any confusion about what needs to be done.
Get your teams on the same page — try LogRocket today.
Want to get sent new PM Leadership Spotlights when they come out?
The globalization of your product opens up opportunities for growth, however, every new market comes with its own challenges.
Hypergrowth happens when a company experiences an exceptionally rapid rate of expansion, typically more than 40 percent annual growth.
Detractors have long-term effects like negative brand perception, reduced customer loyalty, and a decrease in sales.
To proactively address liability concerns, you can create an internal product recall team for dealing with risks and ensuring quality.
Learn more →
