When I think about what is likely to break future crowdfunding records, I think of products with mass-market consumer appeal: Smartwatches, cocktail coolers, revivals of popular 80’s TV shows.

So when Glowforge launched a next-generation laser cutter at MakerCon New York back in September 2015, I eagerly pledged to get one myself — but I certainly never expected it to make much of a splash beyond the niche of early adopting makers. If it raised a few million dollars, I thought, that would have been a huge success.

To my surprise, Glowforge set a new crowdfunding record, hitting $27.9 million in 30 days. It was a truly outstanding campaign.

To find out how they did it, I interviewed Glowforge’s co-founder Tony Wright, who previously cofounded Rescue Time (YC W08) and TomoGuides (YC S12.) He shared some great step-by-step details about how they created the most successful crowdfunding campaign in history.

Read an edited version of our conversation below:

---

Luke Iseman : You’d already raised some money, and both you and your cofounder are successful entrepreneurs. Why run a pre-sales campaign, rather than just start producing and selling units and gradually scaling based on demand?

Tony Wright : We raised a $9 million Series A round before the crowdfunding campaign, so we certainly could’ve gone that route for a while. There were a few reasons we went with a pre-order campaign:

• You build a strong relationship with lots of customers sooner rather than later. When someone pre-orders a thing, they’re going to respond to your emails, fill out your surveys, give you feedback on UI, etc. We have an amazing forum (that only our customers can post to, but anyone can read) where we can talk to our customers any time. We learn from our customers every day.

• Market, or lack of it, makes or breaks any startup that is aiming for a big impact. (See this old essay by @pmarca). We had some anecdotal evidence that people were excited about Glowforge, but it was hard to know how that would translate into customers. We didn’t (and still don’t) have any direct competitors that were as big as our ambitions were. The only real way to prove that people will buy a thing is to actually be able and willing to sell it to them.

• Lots of smart people we talked to were skeptical. We didn’t want to take years of our lives to learn if they were right. The biggest existential question for the company was: Is the thing we’re building as groundbreaking as we think it is? We were breaking a fairly cardinal startup rule: “Don’t try to create a new market.” There were certainly laser enthusiasts out there, but for us to be a meaningful company, the vast majority of our customers would quickly be comprised of laser neophytes. The good news is niche technologies go mainstream all the time, often with spectacular results. MP3 players, touch screens, PCs, tablets, 3D printers– all of these were technologies that had been around for years before the timing/technology was right to consumerize them.

• We were pretty confident we could do well. My cofounder Dan Shapiro had (as a side project) crowdfunded the bestselling game in Kickstarter history and understood the system really well. I had a pretty deep understanding of viral mechanics that we thought could help with a referral campaign. We’d both done PR with pretty wild success in the past, which gave us expertise and contacts that we could use to make the launch unusually noisy.

Luke Iseman : How long before launch did you start planning? What if any partners did you use to help?

Tony Wright : We started talking about the crowdfunding campaign in January of 2015, but probably started planning in earnest during the spring, about 6 months before we launched it. We ran the campaign on our own website, so we worked with Deltasys, a WordPress dev shop, to customize WordPress to take orders and do all of the referral system magic we wanted it to do. Paypal and Stripe(YC S09) handled the commerce side of things. We worked with a carefully-selected PR firm to handle some press stuff, despite the fact that Dan and I were both pretty strong on the PR front. But like most people who hire outside PR, we weren’t blown away by their results.

Luke Iseman : How did you prepare to launch?

Tony Wright : Ooof– this is a long list.

• We iterated like crazy on the video script and tested every edit of the video with a fresh batch of Mechanical Turkers, asking them if it made sense, if they thought the product was desirable, if the video was cheesy, et cetera. It was really educational.

• We did the embargo dance with press and it worked surprisingly well.

• We picked a launch event to give it a little extra punch, Maker Faire in NYC, which I think really helped.

• We iterated a few times on the referral system rules.

• We tested Facebook ads to our “coming soon” page to learn a bit about what headlines and imagery converted best.

• We had a “coming soon” page with email capture for months to build up a launch list. We had a few other email lists from past projects that were a good fit. These worked surprisingly well.

• We did some “sneak peak” interviews with folks like Tested.com to build interest.

Luke Iseman : What had highest ROI in your launch preps?

Tony Wright : I wish we could know for sure. I think the video investment, both in terms of our own effort and the money we spent, was huge. Building our own site and referral system was big. I think it was way more effective than giving over the whole experience to one of the hosted options out there.

Luke Iseman : What didn’t you do that you wish you had before your campaign launched?

Tony Wright : We weren’t staffed to do advertising in earnest, so we really only ramped this up mid-campaign. Most crowdfunding campaigns aren’t venture backed and are often cash-strapped… Which is a shame, because the ROI on advertising during crowdfunding can be great.

Luke Iseman : How did you engage with your email list subscribers to encourage them to purchase on day 1?

Tony Wright : We definitely believed that the first day or two were the most important days of the campaign. Usually campaigns are “U-shaped” — the first few days are big, the last few days are big, and the middle is pretty dead, so we had a ton of stuff queued up. I don’t think we did anything terribly clever with our email blasts. We worked REALLY hard to clearly pitch our referral system to folks. And we’d worked hard in the months before building up a launch list.

Luke Iseman : How much did you honestly think you’d sell, and how much did you end up selling?

Tony Wright : I’d love to say that we knew we were going to sell $27.9 million worth of these things, but we honestly had no idea. We’re so conditioned to believe this quote from Howard Aiken: “Don’t worry about people stealing your ideas. If your ideas are any good, you’ll have to ram them down people’s throats.” Like most realistic entrepreneurs, we were prepared for a decade long struggle to convince the world that they wanted this. To add to these concerns, no one had ever had very much success selling a high dollar item like this in a pre-order. With good reason: You have to really strike a nerve for someone to part with thousands of dollars on a pre-order item.

Internally, we thought we could hit $2M in sales and talked about calling it a “high five success” if we hit $5M.

Luke Iseman : What drove this? Referrals from x, remarketing on y, help from z…?

Tony Wright : I think crowdfunding success is a pretty complex multiplication formula, but the biggest multipliers are probably, 1. How desirable are the features and benefits you’re pitching, and 2. How many people want it. That said, I think we did a few things that were pretty smart.

• We had a world-class video made by Mikhail Productions. Dan and I spent hundreds of hours agonizing over the script. In the editing phase of the video, we tested half a dozen variants against videos from successful consumer hardware videos, with some pretty surprising results. Eventually, we landed on an edit that tested better than videos from other $10+ million campaigns.

• We had a 2-sided referral campaign. When folks bought a Glowforge, we gave them a special URL and said, “For each person who buys a Glowforge with this $100 coupon code, we’ll give YOU $100 off.” It turns out $100 is pretty damn motivating. We saw users buy AdWords ads linked to their referral codes. We had users post to Craigslist and Reddit with their codes. The result was that many of our customers saved hundreds or thousands of dollars and we didn’t have to spend much on advertising.

• Speaking of advertising, we had quickly experimented with ads to see if we could make them economical. Facebook and retargeting with Adroll both worked out really well. Adwords worked out okay. Twitter didn’t pencil out too well, but we’ve seem some promising results since then.

Luke Iseman : What percent of sales were from the referral system (meaning at least $100 discount)?

Tony Wright : I believe it was about 30% were directly attributable to the referral campaign, but we figure there’s some additional splash. Users who come in via a referral link from one of our customers convert at about 6X more than users who came in through any other channel. Every time I use social proof in any marketing effort, I’m reminded of how powerful it is.

Luke Iseman : How have sales gone since the end of the campaign?

Tony Wright : We’d heard other folks who’d done campaigns say that whatever you do in your 30-day campaign, you can expect to do 3% to 10% of that for each month thereafter with fairly minimal effort. That was our expectation. We haven’t released any numbers, but I’d say we’re really really really happy with what we’re seeing.

Luke Iseman : Are the same drivers still leading to sales?

Tony Wright : The whole month of the crowdfunding campaign, we threw pretty tremendous effort at making the campaign successful. If our biggest existential question before the campaign was “How many people are excited about what we’re building?”, the question we’re focused on since then has been “How great can we make the product and how soon can we ship it?” In other words, we’re working a lot less hard on marketing and PR. We’re still spending enough marketing dollars to continue to learn what works, and the referral system is still chugging along. So, it’s the same drivers with less energy behind them.

The new driver that we’re seeing is our customer base working hard to spread the word. I’ve honestly never seen a user base more invested in a pre-launch company’s success.

Luke Iseman : What’s the breakdown by product, and why did you choose to offer multiple models?

Tony Wright : We have two models, the basic, with an optional air filter, and the pro, which comes with an air filter.

The air filter was a no brainer. Lasers vaporize things, which makes a little smoke. A surprising number of lasers just say “vent it out the window”, which is fine for lots of people but impractical in a lot of settings.

Luke Iseman : How did you decide what features these models should have?

Tony Wright : The Pro model was a big internal discussion. We certainly saw the benefit of keeping the product line simple, and we’re passionate about democratizing this technology with a low price point. A $2,000 laser that doesn’t suck is pretty revolutionary. But we had a strong feeling that we had plenty of customers that wanted more power and cared less about price.

There’s also the complication that lasers are regulated by the FDA. To be a Class I Laser, you have to be fully enclosed with hardware interlocks. Our basic model clears that bar, but our vision system (complete with some pretty impressive software) allows for a really useful pass-through slot in front and back to work with bigger materials. Having two models was the only way to offer a Class I laser AND have the passthrough feature.

Once we knew we were going to have a Pro model, the question was, “What pro features can we offer that won’t require a massive re-engineering effort?” We upgraded some of the internals to boost the speed of the device and also offer a longer warranty for Pro models, so it’s great for Maker Spaces and other high production environments.

Luke Iseman : A bit off-topic from the crowdfunding campaign, but I have to ask: Does it drive you crazy to call a laser cutter a 3d laser printer?

Tony Wright : A tiny bit! That was another big discussion. Calling this a “Laser Cutter / Engraver” would’ve been a terrible decision if our goal is to sell this to consumers. Any time we demo’d the Glowforge to people who’d never used a laser cutter, they’d later say, “I really liked that laser printer of yours!” Pretty soon we stopped correcting them. We knew this would annoy some grumpy old school makers, but we’ve been surprised at how few of them are bothered by it. It was a calculated bet that costs us a few sales today, but gave us a category name when we’re talking to laser neophytes — which will be an increasing percentage of our customer base.

Luke Iseman : We’ve seen 3D printers, we’re slowly seeing CNCs, and now laser cutters. Is there another ‘maker pro’ tool you think is about to take off with innovations to make mainstream accessible?

Tony Wright : I think 3D printing has soured the mainstream world a bit on maker tools. They seem so magical but there’s still a big gulf between consumers and fabrication. I don’t know of anything on the horizon that’s aiming mainstream, but there are lots of cool technologies being explored for makers, like Carbon3d. The things we see today are grandparents to the Star Trek Replicator, which we’ll get someday!

Luke Iseman : When do you think someone will break your record?

Tony Wright : I’d wager it won’t be long. When a new marketplace opens up, the first people there are usually indie folks like we’ve seen mostly on Kickstarter. Remember the early days of the App Store? It was all indie developers making millions of dollars. Now you can’t find anyone in the top 20 charts of the App Store that aren’t 4+ year old companies with big teams and big bank accounts. We’re starting to see more and more companies come at crowdfunding well-funded, well-connected, and well-staffed.

Luke Iseman : When will I get mine? 🙂

Tony Wright : You can come up to Seattle and you can meet a Glowforge any time!

Seriously though– we’re committed most to building a great product and and shipping as soon as we can. We’re eagerly hiring amazing software developers, firmware folks, and more to help with that. We’re currently aiming at shipping the first crowdfunded units mid-year.

There’s plenty of hard work between here and there. But the good news is that we’re making beautiful projects with the beta Glowforges every day.

Luke Iseman is the Director of Hardware at Y Combinator. You can preorder a Glowforge here.