How our $500k Kickstarter campaign crashed and burned
Naiveté, poor decisions and six expensive lessons we learned along the way
Already looking like you’re a year behind schedule isn’t a great place to be, but in our most recent Kickstarter update, we revealed that our software was finally complete. It was a major milestone.
Finally finishing the software was absolutely worth celebrating, but the celebration didn’t last long: Once we finally had the software sorted, we were confident that the hardware didn’t need any further updates. This meant that we were finally able to finalise the Bill of Materials (BOM), and to get the tooling (i.e. the injection moulding parts) ready to put Triggertrap Ada into mass production. This was the point where we received a final quote from our manufacturing partners. This was the point when it all ground to a halt.
How it all went wrong…
I’ve made no secret of the fact that getting Triggertrap Ada to market has been a rough ride. If you’ve been following our adventure, you will already be aware of the fact that we were estimating delivery in May 2015 — exactly a year later than we had originally predicted. What I haven’t discussed so far, is the financial impact this delay has had on this project.
So, let’s talk about that…
We completely blew our development budget
One of the first things you do when you’re planning to bring a new product to market, is to create a project budget. So that’s what we did. We used the information we had at hand — previous experience, estimates from our manufacturing partners, and quotes we received from our development agencies — to put together a comprehensive budget. We were more or less on budget by the time the Kickstarter project launched. In fact, we were feeling pretty good about ourselves at that point: While the pre-Kickstarter prototype was very late, we were able to launch our product on schedule by putting the Kickstarter page together way faster than we had planned, and the money we spent on the Kickstarter project itself came in around 2% under budget.
Sadly, that was the only part of this project that went to plan.
Pretty much as soon as we launched Triggertrap Ada on Kickstarter, things started to go wrong: On the same day (!) as we launched our campaign, we received a letter from a major camera manufacturer’s law firm. They took exception to the original name we’d chosen for Triggertrap Ada, and the expenses we incurred in our attempts to avoid a lawsuit, in addition to trying to register the new trademarks etc, meant that our legal budget was blown way out of proportion within weeks of the Kickstarter project going live. To say that this was a stressful period is understating things.
OK, so we averted getting sued by hiring a small army of lawyers, but we also reached the 200% mark for our campaign within a couple of days, so we were optimistic that these extra costs would be manageable.
Of course, the legal costs were only step one of the battle. The electronics and software design for Triggertrap Ada ended up costing vastly more than we had originally budgeted, in part because it turned out that we couldn’t use the microprocessor we wanted to (the electronics agency claimed that the original microprocessor didn’t have enough memory), and had to do several more design iterations than we had anticipated. Compared to our original project budget, we spent 9.4x more on this phase than we planned to.
In part because of the additional design iterations, we ended up having to spend two and a half times what we had budgeted on our prototyping costs — high-quality 3D printing and subsequent hand-finishing of prototype plastics is hideously expensive — and our industrial and plastics design went significantly over budget.
Today, we do have a working prototype of Triggertrap Ada with software we are happy with — which is a victory, of course — but in our frenzied focus on trying to get the damn thing ready for manufacturing, we lost sight of the overall picture, and by the time we totted up everything we had spent on getting to where we are today, we spent more than five times more than we had budgeted for.
So that was our first problem.
We screwed ourselves over on the production costs
After spending way, way more than we had planned for to get to the working production prototype stage, we discovered the next major problem: The tooling (i.e. the injection moulds needed to create the plastic casings etc) was significantly pricier than we had anticipated. That’s one thing — but that’s something that would be depreciated over time, spreading the cost of the tooling over the lifespan of the Triggertrap Ada product. In other words: It’s OK to spend £50,000 on tooling if you are planning to make 50,000 units — in that case, the tooling contributes to £1 of cost per unit. We were disappointed by this development, but we figured we’d find the money somewhere.
However, we also discovered that the Bill of Materials (BOM) cost was about three times more expensive than we had originally planned for. This is a much worse problem: whereas tooling costs are spread over the life span, the BOM cost hits us every single time we produce a single Triggertrap Ada unit. The way pricing works through the retail channels we use work means that every extra dollar of cost in the manufacturing process turns into a huge price increase. We discovered that we’d need to sell Triggertrap Ada at £200 — or around $350 — per unit. That was a problem: Our original goal was to be able to sell a base unit and a sensor for $99.
When we kicked off the Triggertrap Ada Kickstarter project, it was going to be the best product out there, in its class. Selling it at $99 would make high-speed photography accessible to a whole new generation of photographers, and we figured we would be able to sell tens, if not hundreds, of thousands of the product. Not only would it be an awesome product, it would also help lift Triggertrap to the next level in terms of being a photography accessory manufacturer.
We became very focused on delivering the products, and made some grave mistakes along the way — especially in terms of project management and partner selection, as outlined in my blog post from January. No shortage of monumental fuckups, then — but most importantly, we never revisited the price point.
The problem we’re now facing is the deal breaker: At the price point we’re now looking at ($350 instead of $99), the product doesn’t make as much sense anymore. Yes, it’s still a great high-speed trigger, but at $350, it competes in a completely different category than our original product idea. Put simply, it started to dawn on us that perhaps Triggertrap Ada didn’t make sense as a product any more.
The next question was what becomes of our 2,000 Kickstarter backers if we do need to cancel the project. The issue is that the $350 price is calculated based on a production run of 5,000 units. If we only make 2,000 units, things get significantly more expensive per unit. Most importantly, the tooling costs become prohibitive. To use the number from our previous example: Spending £50,000 to tool up for 50,000 units makes sense because you get the economies of scale, but the tools don’t actually become any cheaper if you want to manufacture just 2,000 units. That means that instead of contributing £1 per unit, if we are only ever creating 2,000 Triggertrap Ada kits, the tooling represents £25 per unit. In effect, that means that the cost per casing has just gone up 25x. Not pretty.
The core of the problem is doing a production run of 2,000: it’s a really dumb number. It falls between technologies: It’s not cost effective to use injection moulding due to the tooling costs, but 3D printing or other low-volume manufacturing processes are similarly expensive. If we were to 3D print the casings, the cost per unit is exponentially higher than injection moulding, because it involves more manual labour, it’s much slower, and the materials themselves are way more expensive.
And then, the pre-orders failed.
After our Triggertrap Ada campaign closed, we received a barrage of emails from people who wanted to buy the product, but who had missed the Kickstarter project. We set up a mailing list where people were able to throw their names into the hat. Nearly 4,000 people did just that, and signed up to be notified when Triggertrap Ada would open up for pre-orders.
As soon as we finally started getting a better picture of what the costs would be, we started discussing pre-orders. The reason for doing pre-orders was twofold: If a large number of people pre-ordered, it would tell us that the higher price point wasn’t a problem, and that photographers were willing to spend the additional money for this awesome product. Internally we expected that at worst 20% of people who had signed up to the newsletter would pre-order, and at best 50% of people on the list would convert into sales.
When we opened up for pre-orders in late December, we got a spectacularly nasty surprise. Instead of falling into the 20–50% range, or an expected £220–500k ($350–800k) worth of pre-order sales, less than 1.5% of the people on our mailing list converted into sales, and another few dozen people who weren’t on the mailing list placed an order.
To us, this meant two things.
- One: People didn’t want Ada at the higher price point, which means that we couldn’t sell them if we manufactured them.
- Two: We were relying on additional revenue to start our tooling and mass production process. Revenue that didn’t materialise.
The sum of all parts
Ultimately, it all conspired to be a perfect storm: Our R&D turned out to take six times longer than we had anticipated and cost five times more than we had budgeted for. The production cost per unit was three times higher, and the tooling costs were significantly higher than we had budgeted for. And the final kick in the groinal area: instead of getting a 20–50% uptake on our post-Kickstarter pre-orders, we got 1.5%.
Needless to day, things weren’t looking good.
As a last resort, we went and spoke to our bank to try to come up with a bridge loan. It wasn’t a particularly fruitful conversation: They had doubts in our ability to deliver the Ada sales we were aiming for, especially considering the pre-order failure. We turned to a series of investors to see whether we were able to convince them to have faith in what we were trying to do, but failed on that front, too. Or rather: The investors were very excited about Triggertrap as a company, but were less than convinced by Ada — especially when they’d taken a look at the numbers.
With all our backup options exhausted, we continued doing in-depth scenario planning, but to no avail. In our rose-tinted, raspberry-fragranced, truffle-flavoured best-case scenario, where everything on the Ada project goes 100% to plan; where people fall in love with Ada and started pre-ordering it like there’s no tomorrow; where our distribution sales went through the roof; and where we did six consecutive months of record-breaking sales through distribution, via Amazon, and in the Triggertrap shop— we were still screwed. Put differently: If we try to manufacture Ada, Triggertrap will be bankrupt come May.
And that was the best case scenario.
I spent days going through this every-which-way, and came to the conclusion we most wanted to avoid: We had to abort the Ada project. We got so goddamn close, but there was no way we’d be able to pull off the manufacturing phase.
It’s hard to describe the profoundly visceral reaction I went through as it dawned on me what was about to happen. To me, Kickstarter backers aren’t just customers. They’re our friends. They are mavens. They represent our most passionate, most ardent and most enthusiastic customers. The ones who promote us, and keep cheering us on when the going gets tough.
But… Triggertrap is a going concern; We have hundreds of thousands of customers around the world, and more than a million photographs are taken with Triggertrap’s Mobile products every month. If we commit to delivering Triggertrap Ada, there’s an extremely good chance that the company won’t survive. If that happens, we don’t just let down our Kickstarter backers; We also let down the six-figure number of customers we have around the world, the Triggertrap staff lose their jobs, and it all grinds to a halt. That simply cannot happen on my watch.
And so we only have one option left: Refund the remainder of the money we raised from Kickstarter to our Kickstarter backers, and double down on Triggertrap Mobile.
What have we learned: 6 expensive lessons
We’re in a very painful place at the moment — deciding to cancel Ada is one of the toughest decisions we’ve made in Triggertrap history. But we’re not going to hide from the fall-out, and as soon as we realised that things were looking ugly, we started digging into the reasons for why this happened, and what we’d need to change to avoid anything like this happening again.
Lesson 1: Fail faster.
I’m a big believer in ‘failing fast’, and if there’s one thing we failed to do above everything else in this project, it is just that. It might sound weird that ‘failing to fail’ is a problem, but in my mind, that’s exactly the case. There were quite a few points throughout the Triggertrap Ada project where we should have taken stock and pulled the plug earlier, or at least taken action when we realised things were moving in the wrong direction.
If we had been more critical of our software partners and inspected things more closely throughout the process, we would have switched software agency much earlier. If we had been across the BOM cost more carefully, we would have discovered that the product wasn’t as viable as a commercial product longer ago.
If we had failed earlier, we would have had enough of a buffer to re-plan and do things differently.
Or, if we had discovered we did have to cancel the project earlier, we would have been able to refund more of the money to our Kickstarter backers.
Lesson 2: Don’t let money mask the problem.
When we received the money from Kickstarter (minus their cut, the payment processor’s fees, and taxes), we had more money in our bank account than we ever had before. That sounds like a luxury, of course, but with that much money in the bank, we lost sight of some of the problems. We never said that explicitly, but I feel there was a notion that we would be able to throw more money at any problem that arose.
Of course, this wasn’t the case: It was always going to be very expensive to bring Triggertrap Ada to market. Losing a sense of frugality in the product development phase of the project is ultimately what caused us to lose our safety buffer.
Lesson 3: Tighter project management.
I mentioned this in my previous post as well, but one of the big shortcomings we had in this project was a really basic project management fail: We failed to manage the resources we had available to us (money, time) and if we’d known the full implications of having to change the microprocessor we had originally chosen, for example, we would have taken another path.
Lesson 4: Avoid scope creep.
This is the one I really punish myself for every day, not least because it’s such an obvious mistake. And yet, we fell into the trap. Scope creep is anathema to getting a product out the door, and yet, we fell into that trap really early on, when we decided to change the microprocessor.
The assumption was that changing the microprocessor would be straightforward, and that we would be able to get the UI we had designed. Realistically, the correct answer would have been “We would love a better UI, but our budget doesn’t stretch to re-doing the electronics and software from scratch, so we’ll have to live with a slightly worse UI than planned” rather than “Let’s write a blank cheque to the electronics guys to do major open-heart surgery on our electronics design.” Hindsight: it’s a bitch.
Lesson 5: Get the right skills.
We thought we had all the skills we needed to deliver this project. We were incredibly wrong. As soon as the Kickstarter money hit our account, we should have hired an experienced hardware product manager.
Lesson 6: Don’t be naïve.
Towards the end of the project, we engaged an extremely experienced hardware project manager, both to discuss how things were looking, and to see if we could salvage the project.
To kick it off, I figured I’d ask him how we should have run this project. The challenge we set him: “If you have £300k to develop a consumer electronics product, how would you go about it?” He looked me straight in the eye, blinked twice, and said “I wouldn’t. Not with a budget of under £1m.”
I recently spoke with another well-known photography hardware manufacturer, who estimate that to bring a single product to market, they’ll spend upwards of $30m — and that’s before they’ve mass manufactured a single item. I guess the takeaway from this is that we’ve been seriously burned. Sure, we made a lot of mistakes along the way, and there’s quite a few examples of Kickstarter projects who’ve successfully launched consumer electronics products… But the sheer amount of R&D that goes into a single product is staggering, and thinking that we’d be able to do this on our own was profoundly naive. A healthy dose of realism and scepticism — and taking professional advice earlier — would have gone a long way.
So… That’s it, then?
Yes, that’s it. I failed. We failed. It feels horrible, and it’s the end of Ada. But not the end of Triggertrap.
We were ambitious and hopeful enough to believe that we — a young, enthusiastic, and idealistic group of nerdy entrepreneur-photographers — could bring a brand new consumer product to market. We believed we could do it backed only by the support, enthusiasm, and funding of 2,000 other hopeful, excitable photographers who were willing to invest their hard-earned cash in seeing if it could be done.
We’re not the first Kickstarter project to get overfunded and then fail to deliver — nor will this, I suspect, be the last. We genuinely believed we could do something completely new, utterly exciting, and at a previously unattainable level of affordability. And we were proven spectacularly wrong.
Why did we believe all of this so fervently? Well, because, frankly, we’ve done it before. We’re selling the world’s best-selling iOS/Android remote control for SLR cameras. Our customers are creating awesome photographs with our products every single day — check out the Flickr pool, for example. Over the last three years, Triggertrap has become a touchstone for inspiration to hundreds of thousands of photographers, because of the products we’ve launched, the projects we’ve inspired, and the community we’ve helped build. We’re fortunate enough to have a team of truly talented people, each one positively bubbling with ideas for incredible new photography products they want to see brought to life.
In short, the failure of Triggertrap Ada is not the end of innovation at Triggertrap — not by a long chalk. We’ll be licking our wounds for a while, for sure, but we’re also taking this opportunity to intensify our focus on what we do better than anyone else. As sad as we are to see our high-speed trigger fail to take flight, we’re not about to stop aiming high.