What I’ve Learned Running Redfin’s Products Organization

This article originally appeared on Redfin's blog.

I still have the brown-paper sack on my desk, unopened, full of compliments. Melissa Blume, a Redfin director of product management, recently asked the team to write on a notecard what we appreciate about one another, and then she put the compliments for each person in a bag. I’m too mortified to read mine, but I know I’ll never throw away that bag. It’s a talisman on my desk.

It reminds me of all the bags I should fill for others at the company. But what it really says to me is that things can turn out better than you deserve, and that culture isn’t what comes from you as a leader, but what happens while you’re not there.

I’ve been running products for a few months now, since one of our execs left Redfin to build a Norwegian sauna in his backyard, and hopefully one day to build a new company too. I was despondent about his departure, but also a bit flummoxed by the new responsibility. I hadn’t been a product manager in 25 years, I hadn’t had a new job in 14 years, and I’d forgotten how to manage someone under 30. This post tallies up what I’ve learned from the experience.

Up ‘til recently, I’d known our products organization primarily through our exquisite, often unrivaled, software for finding, buying and selling homes. For anyone who has ever dreamed of a totally different life, nestled in a small town with more sunlight and less traffic, or just in a house with a regal bathroom, Redfin.com is a cult site, and the product team has had the swagger that comes with being touted on hit TV shows, by superstar athletes and celebrities, by people we meet on airplanes and at parties. 

Software to Answer Some of Life’s Deepest Questions

And up until recently, the main things that interested me about our products team were the problems we needed to solve. We’ve only just begun to use machine-learning to answer the most fundamental questions in commerce, about where our customers should live and how much they should pay for a home. We’re still learning to identify the customers who need immediate help and others who want the space to look on their own, how to determine from photos whether a home is in good shape or bad. 

We’re building mobile applications that talk to door locks so people can walk into a home any time day or night; our tools will soon ask Redfin agents to rate the road noise or to identify the repairs we need to make to a home we’ve been hired to sell. And we’re now guiding people beyond the initial listing search and through the whole 12-month maze of pre-approval letters and school ratings and title insurance; distilling so many choices and so much data about those choices into a personalized application is one of our generation’s great software design challenges.

What makes this especially hard is that we’re building tools for our agents to connect with consumers, but increasingly too for consumers to buy and sell homes on their own, which no one has ever done before. And we’re now working to predict the future so we can buy a billion dollars worth of real estate every quarter that can be sold by our home-flipping business at a profit.

And then we’re starting all over again to build the same tools and analytics for mortgage, home renovations and even title insurance. Wiring together every step of the move will let our customers close faster, with access to money no one else can give them, to win a home they never could have gotten working with any other broker. And the stakes are high: our customers are people who just put their parents in nursing homes, who just broke up with their girlfriend, or got pregnant, or lost their job, or got married, or drove across the country with their kids crying over the friends who were left behind. The transition they’re making isn’t from an Android to an iPhone, from Nike to Adidas running shoes; it’s to a whole new life.

Artsy-Fartsy, But Metrics-Driven

Building software to solve those problems takes art and science. I’ve always thought the weird thing about the team who designed this software was that they’re artsy-fartsy, but also metrics-driven. I used to forward around to all my friends the exec reports our product managers make, to show them that a PM can develop a model of how our business works inside a single product, as perfect as a Faberge egg. The team is so much more analytical than I ever was as a product manager, and more business-savvy. I feel so proud when the products team shows off our wares to the board, answering directors’ questions about how our products will make money.

How to Be Kind and Driven

And yet it’s the kindness of everyone on the team toward each other that surprised me as the interim head of products. When I was coming up, product managers were gladiators of ideas, barons of bluster. But that isn’t how you get the best out of a team, especially a diverse team. What’s so hard to do is to believe in an idea until it hurts, but then to be able to take seriously a contradictory idea. It’s hard to be driven and kind, to think and to feel; sometimes the way I get things done is by turning off my feelings; sometimes I think it’s why I like to get things done, because my feelings can otherwise be so loud.

But the team is much better than me at feeling and doing. It’s fashionable to talk about millennials as avocado-toast-eating patsies, but the way those people nod thoughtfully about disagreements when my feelings would’ve been hurt… it boggles my mind. We still have plenty to do to make design reviews and workshops constructive and supportive but the team works hard at it. We want to make sure everyone’s heard in a meeting. We spend a lot of time filling one another’s brown paper sacks. When I was the age of many of our product managers, I was the brown paper sack, existing purely to get praise.

Spiders at the Center of a Web

Here’s the other weird thing I’ve learned about our products organization. They’re like little spiders at the center of a web connecting our software, our service, our customers and our revenues. They’re the first to feel a fiber at the far edge of the web tremble, and the first to repair it. It was a product manager who told me our agents were meeting too many customers to follow up well with each one. It was a designer who told the executive team we had to find a way to get customers into homes instantly. It’s the product managers who party ‘til 2 a.m. with our real estate agents at our annual kickoff.

We All Sweep the Floors

And that’s why they know what’s wrong with Redfin, because they know all the people at Redfin. We aren’t like Uber, Instacart or Doordash. Our real estate agents are our colleagues. We have to make tools that fit perfectly in our agents’ hands, so our agents can sell homes for more money and still be three times as productive as our industry peers. And to do that, the products team and our agents have to work together as partners and friends.

Nearly all of tech’s problems are because our last contact with the real world is when the barista hands us our hot chocolate each day on the way to work. It’s what makes San Francisco and Seattle sometimes unbearable. A designer or a product manager can’t sit in a tub all night and ladle beautiful pixels over ourselves, and we can’t study our colleagues and customers like zoo animals.

We have to be a part of the world. It takes effort and humility, not the more ballyhooed qualities of vision or brilliance. That’s something that’s just part of the daily texture of working here, not an occasional leap of imagination or a trip to the field. And it’s what most people stumble on when applying to be a product manager or designer here.

Every Exec Should Take a Chance to Peek Behind the Scenes

But there’s one last thing I’ve learned trying to run products at Redfin, and it isn’t even about products or Redfin per se. And it’s that the farther you go in this life, the more you need to understand what’s happening behind the scenes. Who slaughters the chicken that has been gutted and grilled to top your salad; who dug up the gold you wear every day around your finger; who sorts the paper from the plastic in your recycling bin? There are so many things in this world we wouldn’t consume if we had to produce and dispose of them ourselves. That’s true at an ethical or ecological scale, but it’s also true in a company. 

Until running the factory that produces so many of our strategy decisions, I never realized how many people pitched in on a proposal before it comes to me, and never suspected how much was riding on my capricious reactions to that proposal. I’m still trying to figure out how to simplify the process and lower the stakes for making such proposals, which involves trying to tell people it’s ok to be wrong, but not too often.

But now that I’ve worked with the team that pulls together those beautiful, delicate ideas from across the company, I think about what a wildlife biologist once said to me about the lox on my bagel: “That salmon swam to Russia and back, then 250 miles upstream. It’s an incredible creature. Trees grow taller near the riverbed where it should have died. Go ahead and eat it by all means, but it should be a special occasion when you do.”

As a company grows, ideas are packaged up for execs to consume or discard like so many salmon steaks in a Costco freezer bag. But those ideas are precious, finite and friable, and the people who produce them are even moreso. That’s why it has been an honor, and an eye-opener, to be a part of this team.