The idea came to me five years ago. I was at a friend’s backyard barbecue. I noticed a cool plant with amazing flowers, and so I asked the girl who owned the house what it was. She didn't know. I wanted the answer so I went to the Internet. I started thinking about how can you identify a plant? What do you do? What's the process? I got online and typed some words to describe the plant I’d seen, and that got me nowhere. That was my Eureka! moment. I thought there should be an app for this.

At the time I owned the science technology website redOrbit. I went to my developers and asked them if it would be possible to build an image recognition app that identifies plants. We dug into it and spent about $35,000 on research and development. We figured out that the accuracy just wasn't good enough to pursue the idea full-time and spend more money on it. So I gave up after a few months and decided to revisit it later.

That was back in 2012. Every year after that, I was revisiting the idea, seeing where the technology had gone, and talking to new developers. I think I went through three different companies trying to do it and didn't get anywhere. The success rate for identifications was about 65% at best. That's wasn't good enough to turn it into a company.

Then at the start of 2016, I was reading an article about the artificial intelligence behind self-driving cars and a new technology called machine learning. It seemed like a perfect fit for PlantSnap. I hired a couple of new developers and they dug into it a little more. It turns out we got really good results on a small dataset. That's when decided like the technology was at a point where we could move forward.

I formed PlantSnap Inc. in June 2016 and had a beta version of PlantSnap finished in August. From there we started building the algorithms. Building the app interface is easy. That's the front end that everyone sees. The hard part is the machine learning algorithm. That is our secret sauce. What you have to do is train the algorithm to recognize every plant on the planet, which is not an easy job.

Believe it or not, there is no global plant image database...at least not a published one. That became my job. I had to piece together a global plant species database, then find images for every species.

The algorithm analyzes a picture of a plant and uses images in our library of 250,000 species to match the new photo submitted with every image in the database. It all happens instantaneously. You use the app to take a picture of the plant, and the algorithm compares the picture to tens of millions of images. The app also gives you a percentage of how confident it feels it is correct, and lists three or four other species that it thinks might be the answer if it’s not 100% certain. From there, we hyperlink the results to a page on Earth.com that gives you more information about that particular species.

We are amassing the largest plant image database in the world. Halfway through this process, it dawned on me that I was building something that’s never been done before. Every image taken by a PlantSnap user will be added to our database and be used to train the algorithm. With every new image submitted, the algorithm gets smarter...it learns!

The images will be fed to another site I own called Earth.com. It’s the perfect repository for all of this information. At Earth.com we are building a page for every plant and animal species on the planet that includes information and images. Nothing has ever been done like this, and nothing online compares, so it’s a big undertaking. If an alien species came down and wanted to learn about earth, they would only have to go to one place. They’d go to Earth.com and get everything they ever wanted to know.

One of my early investors is a buddy who’s into birdwatching and out in the wilderness a lot. He's got a bunch of books, including plant identification books. He said it’s stupid for him to carry a lot of books around with him all the time. He said the app would be a great tool. That's one type of person I see using the app.

The other type of user are people like me, people who are out in nature all of the time and are curious about what’s around them. One part of our pitch is the concept of bringing people back to nature. Life is so fast paced. If we can get people to stop and smell the roses, look around them, and stop to think about all of these amazing trees, plants, and flowers, maybe people will start to realize how precious nature is and take small steps to protecting it. Maybe, in some small way, PlantSnap can help raise awareness about the climate change problem we've got right now. Right now it’s more important than ever that people start noticing and appreciating nature.

First and foremost, we will push it toward the audience at Earth.com. The site has about 6 million unique users at this point and a significant social media following. We can count on those people to download it as soon as it’s available. Then I have a team that does social media marketing for Earth.com who will transition over to doing social media marketing for PlantSnap. We have a mailing list of about 12,000 people and roughly 15,000 social media followers across all platforms.

We are developing the app for both iOS and Android.

We plan to make money by charging $1.99 for each download and then by selling ad revenue. Monetizing the app will not be a problem. That's my area of expertise because I've been doing it for the past 20 years with the sites redOrbit, GreenAtom, and now Earth.com.

In phase two, I plan to add a gaming component. It’s going to be like Pokemon Go but for plants. Users can see how many different plants they can identify. Every time someone takes a picture of a plant, the picture is added to our database. As our database grows, the algorithm gets smarter and more accurate. We don’t have the details of the game worked out, but for months I saw zombies walking around in nature looking for things that are fake, a Pokemon. My idea is to get people walking around in nature and looking at real things, taking pictures of real things, and learning about those real things as they go. Maybe we can incentivize people to use the app by turning it into a game, possibly with prizes or some other incentives.

It’s been stuck in my mind since I had the original idea. It just seems like such an easy thing that people should be able to do. People should be able to know what kind of flower or plant is growing in their backyard without having to go buy a bunch of books. It seems like such a basic idea that hasn't been done yet. I made it my mission to get this thing built. I believe there is no better way to get people interested in nature than to help explain what exactly it is they are looking at, how it got there, etc. I see this as a global game changer, and I had to get it done.

One thing that could go wrong is people may not care about what plants are around them as much as I do. I don't believe that to be the case. I think there are enough people in the world that do care, but there's a chance that they don't. I know PlantSnap works because I play with it every day, so that's not keeping me up at night. I guess what keeps me up is thinking that maybe people don't care as much about nature as I hope they do.

I studied Japanese and psychology at Vanderbilt. I then went to graduate school at Thunderbird and studied more Japanese. That's when the Japanese were ruling the world and had the biggest tech economy. I thought I’d be working in the business world dealing with Japanese companies. Then the Internet came along as soon as I got out of grad school. I was hooked and immediately figured the Internet was going to take over the world. I needed to be a part of it. I always had a passion for science and space, so the first thing I built was a space website called Cosmiverse. I did it because Yahoo was the only thing around, and I had just gotten out of grad school and wanted to learn more about space. I tried to find information about space on the Internet, and there wasn't any.

I built the site and then a year later space.com came along with Lou Dobbs and a tons of venture capital backing. I competed against them for a while. They were about to buy the company and had given me an offer, and just then the tech bubble burst. They couldn't get any more funding. Lou Dobbs quit. That whole thing went to hell. My company went under in 2002, and then I thought, OK maybe the thing to do is broaden the spectrum. I started another site called redOrbit that dealt with space, science, and technology. I had learned from my mistakes from the first site what to do and what not to do. redOrbit was successful from day one, and I ran that from 2002 to about 2014 when I sold it.

After that I moved to Telluride and started another website. The best thing about these science sites is that every day I learn about something cool. It might be about our planet or the universe or anything. I didn't want to get out of that field, so I started another site called GreenAtom. It is focused on science and the environment. Then one day I started wondering what was happening with the domain Earth.com. It was around the time that all of the attention on the Paris climate agreement came along, and I started getting into the environment. I go to Earth.com and see it's parked doing nothing. To me that seemed like the biggest waste. How can someone be parking on Earth.com and not using it for anything? I got hold of the people that own Earth.com and worked out a deal. Then I launched Earth.com in October 2016. Last year was a really big year. Everything came together. PlantSnap happened, Earth.com fell into my lap, and the environment became a hot topic. It all meshed together at the same time.

We will hyperlink every result from PlantSnap to the corresponding page on Earth.com. If people want to learn more about the plant, they can go to Earth.com to read about it. Earth.com is full of news and information about the environment. We publish between 10 and 15 articles a day about science and the environment. We have a section on endangered species that lists every endangered species of plants and animal on the planet. Right now we are building a page for every plant and animal species on the planet that will give information about the species along with an image. The site is meant to be everything you ever wanted to know about planet Earth.

I consider PlantSnap to be for the people. What better way to get people involved than to let them be a part of something that could make a global impact? PlantSnap is a simple concept but there's a reason it's never been done before. It's because it’s hard. I am building a database that doesn't exist on the planet, and I think that it would be cool for people to be a part of that.

I plan to spend the money to complete the database of 250,000 plant species and for operating capital needed before we launch. I expect the company to generate revenue immediately once we launch, but in case it doesn't we will have money to cover those low initial operating costs. The operating costs for PlantSnap are low after we launch. There's obviously the server cost and the bandwidth cost, but if it's functioning properly there's really not a lot involved other than people answering questions from customer service emails. Getting to profitability should be immediate and easy.

Nothing is going to stop them from trying to copy me, but can they make it better? And consider how long it took me to build this database. Another company is going to have to do the same thing. I'll have a yearlong head start on whoever decides to try to do the same thing, and as long as we have a good product I feel like that will be enough to come out as the winner. A year is going to be plenty of time to own the niche. We will have a year of data that people pump into the algorithm, which only gets smarter and more accurate as we go along. If our app is accurately identifying plants, someone else spending time and money over a year and making another app that does the same thing doesn’t make a lot of sense. I'm sure someone will try, but will it work as well as PlantSnap? I doubt it.

Cash in hand as of March 2017 is roughly $100,000.00, and our burn rate is $25,000/month.