Good Company Q&A: Joel Gratz, OpenSnow

Joel Gratz turned his love of skiing and snow into a career. His passion for powder catalyzed the 2011 founding of OpenSnow, an app that starts where traditional weather forecasts stop, providing users with a pinpoint snapshot of conditions at ski areas in Colorado and beyond.

As the company has grown, Gratz has brought on local meteorologists in ski meccas across the West and beyond to offer laser-focused insights on the weather and added a wide range of features that have won a loyal following of powder hounds.

• Founding Meteorologist, OpenSnow (www.opensnow.com), Boulder
• Age: 44
• Hometown: Doylestown, Pennsylvania
• What he’s currently reading: Avalanche Dreams: A Memoir of Skiing, Climbing, and Life, by Lou Dawson

ColoradoBiz: How did OpenSnow come to be? What were you doing before, and what led to the company getting started?

Joel Gratz: The origin story was me going skiing when I was four years old. I wanted to be a meteorologist since I was a little kid. I loved skiing, loved snow and went to school for meteorology at Penn State and grad school for environmental studies and business at the University of Colorado. The only quote-unquote real job I ever had was as a hurricane and earthquake analyst for an insurance company. It was a great mix of business and weather, except I was obsessed with skiing powder and understanding how to forecast powder. I became obsessed with it and started an email list among friends trying to chase it. I made a blog about it, and then one day I said, “Well, this, this could be a business, and if there’s ever time to start, it’s now.” So, I quit my job and worked really hard for the last 15 years to build this company.

CB: How has OpenSnow grown over the years? I’ve seen you expand it into different regions outside of Colorado with local forecasters.

JG: It’s my dream gig. I am, on one hand, surprised that it all could work. Because most businesses, most dreams, just don’t work for whatever reason. And there’s been a lot of luck involved with all this, but we’ve stayed the course. I think me and Evan [Thayer] and Brian [Allegretto], who were the first forecasters; me in Colorado, Evan in Utah and Brian in California, one of the reasons for our success was that we just wrote every day during the winter for years and years and years and years. If you just stick around long enough, people will notice that.

We are at 12 full-time people now, and over 20, including contract forecasters, during the winter. We have no desire to get big and hire everybody we can find, but as the company grows and our desires increase, we will look for a person or two occasionally that will help us get to the next stage.

About 80% of our revenue now comes from subscriptions and about 20% comes from advertising, and that was flip-flopped six or seven years ago. We still have a lot of advertisers that stick with us and find a lot of value in advertising their products or their locations on OpenSnow and through our email list, but the majority of our business is now subscription, which just adds to the stability of the business as well. We know what our renewal rate is roughly and it has a lot of predictability year over year. That allows us to make investments, which we’ve done over the last year with a bunch of AI-based forecasting. In short, we are a profitable, nicely growing, private, closely held business, and we can chart our own course with no pressure to go out and raise money or sell or anything. It is just a dream come true for me and the team to work on this stuff that we love.

CB: How many subscribers do you have? Or do you keep that number under the vest?

JG: Yeah, we do. Publicly, we say that we generally reach hundreds of thousands to the low millions of people that use our service. A small percentage of those actually pay us, but it’s a significant enough percentage that keeps the lights on and the growth coming.

CB: What’s the driver of that subscriber growth? What are your customers looking for the most?

JG: We’re still largely focused on the Western United States and chasing powder. We have global forecasts, we have coverage outside of the United States, and daily snow reports and local experts elsewhere, but still a lot of the mystique around powder and the desire to get powder and the proximity to powder is still largely driven by the West. That is where the Venn diagram of powder access, population and resorts comes together.

CB: How have OpenSnow’s features evolved and changed in terms of tech and AI?

JG: A lot of the growth over the last few years has not been driven by our local forecasters, but by other convenience-type features. That said, I do want to be very clear, we are not letting go of our local forecasts. I am still writing all the time, and so are other forecasters, so it’s really just a mix of local forecasters and the tech platform, versus just being driven by the local forecasters as it was 10 years ago.

There’s a lot of weather technology that’s becoming AI-based, and we started digging in about a year and a half ago. At this point, we have made our own AI-based models that specialize in weather prediction in complex mountain environments. Normal models just struggle figuring out the valleys and the peaks and how it snows so much on one mountain and not at another nearby mountain. So we are using an AI system, which is just a fancy word for pattern recognition, to go back and look through past storms, learn from past storms and then use that to hopefully better forecast future storms.

An AI-based system can do what I do, but just with orders of magnitude more data and way faster and anywhere in the world in seconds. So, we’ve invested in that system, called PEAKS, and that’s going to be rolling out and powering all of our forecasts this season.

We have also created an avalanche forecasting AI-based avalanche forecasting system. This is not going to compete with forecasts from the CAIC [Colorado Avalanche Information Center], but we said, ‘If we can do pattern-matching for past storms and weather, can we do pattern-matching for avalanche terrain?’ A lot of it’s based on weather, so we’re working on that, and we want to be highly responsible rolling that out. We’re still trying to figure out the best way to do that, to show that side by side with the human-based forecast from CAIC and others.

CB: What’s next for OpenSnow in 2026 and beyond?

JG: We want to develop an AI overview, kind of like a Google search result of the location. When you search for most of the popular ski areas around the world and all of the ski areas in Colorado, we now have an AI overview. Some people want the Cliff Notes version: ‘I don’t want to look at all the charts and graphs, I just kind of want to be led through what’s going to happen.’ We’ve taught our AI effectively to write like we would want to write, to use our data and it works really, really well.

Again, we’re not getting rid of our local forecasters, but it’s another way for people to interact with OpenSnow and I can see that expanding over time. Again, not getting rid of local forecasters, but we can’t cover every location update every hour and respond to people’s questions. It’s just impossible as a human.

We’re also pushing more into year-round aspects of mountain living and outdoor adventure. Almost half of the people who use OpenSnow in the winter will use us during non-winter months as well, either to track smoke or to track lightning for their hikes or mountain bikes or climbs. So, we have a bunch of ideas that we want to push forward with that, because all the same technology that we use in the winter can be used in the summer, and it’s largely a messaging and branding issue to get people to remember that we’re still here.

CB: Where do you like to ski? Any bucket list destinations?

JG: I am so lucky. Knock on wood, if I don’t do any more powder skiing, I feel like I have gotten a lifetime of powder skiing. I’ve worked hard for it, I’ve gotten lucky for it and I’ve skied powder all over the place.

Our son just turned eight years old. Since he’s been 18 months, a lot of our skiing has been driven by where we can ride as a family. Last year, we skied powder effectively where we wanted to and how fast we wanted to, along with him. He got the bug. He worked really hard over the years. We put in our time with candy on the lift and those one-run days with the kid, but that is so rewarding to go out and show him the mountains.

We have also gone to Japan a couple of times, even as a family, because, as a person who’s been snow-obsessed since age four, it is an amazing location where the weather is the opposite of Colorado. Here, you need the atmosphere to come together in just the right way to create a big snowstorm, and in parts of Japan in December, January and February, it’s basically the opposite. The default state of the atmosphere, with cold air flowing across the warmer ocean and hitting mountains, the default state is effectively for it to snow.

The interview was lightly edited for length and clarity.