How Zano Raised Millions on Kickstarter and Left Most Backers with Nothing

One of only four Zanos ever to make it to a Kickstarter backer / Mark Harris

Kickstarter tasked me, a freelance reporter, to find out why a highly funded crowdfunding campaign for a palm-sized drone flamed out in order to give backers the full story, and provide lessons for itself and others. My report follows. Kickstarter had an advance look, but wasn’t allowed to make changes. (Read this for the background on my commission, or skip to the very end of the article for a brief summary of my findings, some additional details and disclosures.)

In spring 2013, the Secretary of State for Wales at the time, David Jones, visited a young company at the Pembrokeshire Science and Technology Park. On a windswept campus at the western tip of the country, Jones heard Torquing Group’s enthusiastic managing director, Ivan Reedman, share a vision of Welsh-made autonomous robots for industry, commerce, and the military.

“Torquing Group’s work is an excellent example of a successful small business with their sights firmly set on growth and expansion,” enthused Jones later. The key to Torquing’s growth, Reedman believed, would be a cutting-edge consumer quadcopter developed from his experience working on a surveillance drone for a local defence contractor.

Eighteen months later, flush with an investment from another company at the technology park, Torquing Group launched a campaign on the crowdfunding website Kickstarter for a palm-sized drone called the Zano. The Zano would dispense with fiddly on-screen controls in favour of mimicking gestures from a Wi-Fi linked smartphone, automatically tracking and following its operator, avoiding obstacles, and even shooting “selfie” photos and videos — for up to 15 minutes at a time.

Reedman’s goal was modest: £125,000 (about $190,000 at the time) to enable him to move his Zano prototype into full production. “Zano is up and flying, holding position, avoiding obstacles, streaming live video back to a smart device, capturing video and photos,” said the Kickstarter campaign page. “Our supply chain is 100% ready to go, from vital components that make Zano fly, to the very boxes that Zano is packaged in.”

Torquing’s promotional video showed some impressive footage of the drone in action. Drinkers at a Welsh pub smiled as a Zano flew up and hovered over them, displaying a countdown on built-in LEDs before snapping a photo that they immediately examined on a smartphone. Zano was then shown filming rock-solid footage of a mountain biker, following a motocross rider’s gestures, and automatically returning to land at his feet. Kickstarter itself selected Zano as a Staff Pick, which gave it prominently placed promotion on the site (the company now calls these Projects We Love).

The internet went wild. Torquing met its initial funding target in just 10 days, then blew right past it. As pledges smashed all Reedman’s “stretch goals”, they unlocked additional features: built-in storage, high-def (1080p) video recording, thermal imaging cameras, wireless charging, the ability to fly upside down, and more. Reedman touted extras like facial recognition and 360-degree panoramas soon — and all for a pledge as small as £139 (about $210). There were drones costing 10 times as much that couldn’t match Zano’s specs. The campaign promised delivery of all reward drones in an equally staggeringly short production window — in June 2015, just six months away.

At the International Consumer Electronics Show (CES) in Las Vegas a few weeks later, Engadget shortlisted Zano for its official Best of CES Award, one of just 49 nominees among an estimated 20,000 new products at the world’s biggest trade show. “Kickstarter made the prototype happen, and now it’s a very real proposition,” gushed Engadget in early January.

By the time the Kickstarter campaign ended on 8 January 2015, over 12,000 backers from around the world had pledged an astonishing £2.3m (nearly $3.5m), 20 times Reedman’s original goal. It was, and still is, Kickstarter’s most funded European campaign.

As 2015 rolled on, a flurry of updates from Zano — they would ultimately post around one a week — delivered encouraging news. The final chipsets for Zano arrived in Wales in mid-February, the firm said; a tiltable camera was developed; and Reedman’s engineers were building a high-performance computer to process hundreds of Zano videos online. Videos showed Torquing’s labs and assembly lines, one with staff using the Zano’s infrared obstacle avoidance system to play “drone tennis”.

To accommodate demand from enthusiasts who had missed the Kickstarter campaign, Torquing set up a website to take pre-orders, eventually racking up over 3,000 additional sales.

But cracks were beginning to show. Some key plastic parts were delayed, there were never-ending compliance and calibration tests to complete, and something was up with the propellers that arrived from China, according to campaign updates. Once-eager backers started to get restive, demanding to know exactly when their autonomous drones were likely to arrive.

Zano drones finally began to trickle out of Pembrokeshire in early September — no more than a handful at first, building slowly to total over 600 by the end of the month. The internet went wild again, although for very different reasons than before: Kickstarter backers were infuriated when they discovered that pre-order customers were receiving drones before loyal crowdfunders. Pre-order customers were infuriated for a different reason: Their Zanos were barely operational.

They reported that drones would repeatedly “bunny hop”’ a few centimetres in the air before landing again, or veer off wildly to crash into walls. Video quality was dreadful, and there was no sign of even basic obstacle avoidance or gesture control, let alone fully autonomous flight. In mid-October 2015, already months late, Torquing again pushed back delivery for the bulk of the Kickstarter rewards to as far off as February 2016.

On 18 November, the axe fell. Torquing announced via a Kickstarter update that it was entering a creditor’s voluntary liquidation, the UK equivalent roughly of an American “Chapter 7” bankruptcy filing. It appointed a liquidator who would bring its business operations to a close and attempt to sell the company’s remaining assets to pay its outstanding bills. Legal documents show that Torquing had not only burned through the £2.5m from its Kickstarter campaign, it had run up another £1m in debt. It was Kickstarter’s most spectacular flame-out to date.

No more Zanos would be made or sent out. Staff were sent home, and Torquing’s supercomputer was switched off and would be sold for parts. Because the Zano drone checks in over the internet with Torquing’s servers each time it powers up to retrieve calibration data and updates, the few drones in the wild were instantly and permanently grounded, like a dastardly robot army foiled in the last reel of a bad sci-fi film. After an abrupt final post on Kickstarter, Zano’s creators retreated offline and refused to engage with either backers or Kickstarter itself, contrary to the platform’s policies for failed campaigns.

By the end of November, it was clear that Zano was finished. But what had actually happened? Was the Zano project a scam from the word go, a money-making scheme to defraud backers? Were the Torquing leadership team inept, negligent, or incompetent, or some combination of all three? Or perhaps they were the victims, sunk by unscrupulous suppliers, malicious staff, or hackers?

Even more importantly, perhaps: Does the failure of a project this prominent and this well-funded call into question the future of crowdfunding itself? Can we ever trust the crowd again?

It was worrying about this final question, I guess, that prompted Kickstarter to approach me in early December to investigate Zano, Torquing, and (like it or not) Kickstarter itself. In the absence of any useful information from Zano’s creators, Kickstarter essentially asked me to research and write the story I would like to read about the collapse of Torquing. The firm wanted me to pay special attention to where the money had gone and what future project creators might learn from this collapse.

Over the last five weeks, I have followed the Kickstarter campaign from well before its inception to the bitter end, read reams of financial and legal documents, talked to as many people involved in the project as I could, and travelled to South Wales to see Torquing’s offices up close — as well as one of the only Zanos that ever made it into the hands of a Kickstarter backer.

What follows is as complete a picture of Zano and the Torquing Group as I could piece together in this short time, from both public and private sources. Some of the material here is not original. I am especially indebted to the many Kickstarter backers who contacted me, and to journalists who covered the Zano story while it was unfolding. Rory Cellan-Jones of the BBC and Cyrus Farivar of Ars Technica, in particular, have proven very helpful.

I have heard about enthusiasm, dedication and hard work, and endless late nights attempting to develop cutting-edge technologies and solve thorny problems. There were tales of teamwork, camaraderie, and community. But I found evidence of overconfidence, exaggeration, and obfuscation as well. Zano is also a story of arguments, personal threats, legal disputes and criminal investigations. A sum that totals £3.5m does not just disappear silently into thin air, nor should it.

Ivan Reedman (left) and Reece Crowther (right) / Torquing Group

Ivan Reedman was born in South Australia in 1975 to British parents. He says that he was programming computers at age 8, dabbling in assembly language at 12, and writing custom software for local businesses by the time he was 13. The next year, he registered his first company under the name Torquing — a play on words that his father came up with. After studying commercial law at college, he grew Torquing into a small IT consultancy fixing computers, networks, and servers, while he continued to work on his own operating system. Reedman has no formal technical qualifications, which isn’t out of the ordinary in startup circles.

In 2007, Reedman moved to London, where he met Anna Dietrich, who he later married. They together incorporated Torquing Technology Limited (TTL) as a limited liability company in 2008, again focusing on IT. “I used the money I made to fund my own research,” Reedman tells me during a long Skype interview in mid-December. “I’ve always been self-funded, working on artificial intelligence systems, computer control, and robotics.”

Reedman is calm, soft-spoken, and articulate. His passion for the Zano project is evident and he still genuinely believes that his drone is only days or weeks away from working as promised. Reedman answers most questions during our chat, and in many subsequent emails, without hesitation, although he does shy away from some queries focusing on financial, management, and legal matters.

In 2010, the couple moved to Dietrich’s native Pembrokeshire, with TTL taking advantage of a small capital grant from the Welsh government to secure office space at the Bridge Innovation Centre (BIC) in the Pembrokeshire Science and Technology Park in the port town of Pembroke Dock. Reedman then set about looking for business opportunities, including a half-hearted attempt to launch a golf-tour operation.

His breakthrough came in 2011 when a contact in the Welsh government put Reedman in touch with BCB International, a manufacturer of marine, law enforcement, and military equipment based in Cardiff. Barry Davies served as a sergeant major in the SAS, Britain’s special forces unit, and has written dozens of survival books. For the last few years, he has also been Robotics Project Manager at BCB. “If you mention BCB in the wrong way, I’ll come and kill you,” he jokes. At least, I’m pretty sure he’s joking.

“In 2011, we were looking to expand our current industry, and drones seemed to be way to go forward,” says Davies. “When we were introduced to Ivan by the Welsh government, I asked him to make an autopilot module.

“It was an easy opportunity to say, yeah, I haven’t done this before but I’ll certainly get into it,” Reedman remembers.

He went back to BCB with a proposal for an autopilot that would provide sense-and-avoid capabilities by integrating infrared and sonar sensors. The whole thing would be controlled over Wi-Fi, the wireless technology used in hotspots for laptops and smartphones. “We were quite taken aback, and thought, ‘Wow, that’s incredible!’” says Davies. “We said, ‘Are you sure?’ And he said, ‘No problem at all. We’ll have it done within a year.’”

After signing a contract with BCB, TTL moved into a larger “growth unit” at the Bridge Innovation Centre to develop the drone, code-named the AV Sparrow. In September, Reedman formed two new companies, Torquing Robotics Limited (TRL) and a holding company, Torquing Group Limited (TGL). (There was also a short-lived company called Torquing Environment that doesn’t figure into this story.)

With money coming in from BCB, Reedman took on a couple of employees and set about diversifying. Anna’s brother, Thomas, joined the company as a designer around this time, invested £60,000, and was eventually made a director of TGL in 2015.

Local news stories in early 2012 said that Torquing companies were working on everything from accounting software and monitoring technology for utilities to assembly line, mine detection, and surveillance robots, as well as robotics kits for schools. I could not find, nor could Reedman provide, evidence of any of these products reaching the market, although he insists his work for the Ministry of Defence went “exceptionally well”.

The same news stories had Reedman and Dietrich estimating their gross revenue at £300,000 to £400,000 in 2012. If they were doing that much business, they were not making much money. Documents filed at Companies House, the UK government agency that registers companies, show Torquing companies with profit and loss accounts of only a few tens of thousands of pounds that year.

In January 2012, Torquing Group won a Local Business Accelerators competition organised by the Western Mail newspaper. The prize included mentoring from Bill Mayne, chief executive of MSS Group, a business services organisation. “They talked about things for which I couldn’t quite get the commercial application,” says Mayne. “It was virtually impossible to tell what they did from their own material. And having spoken to them, it wasn’t much clearer, except that they seemed, on a positive note, very bright.” Sion Barry, then business editor of the Western Mail, notes, “They were a bit strange in that they didn’t want any help and didn’t even want the £25,000 advertising [prize].”

By late 2013, BCB was starting to lose patience with Reedman, Davies says. Torquing’s prototype AV Sparrow was proving extremely temperamental.

“It was very unreliable and in the end, I said switch it to use the normal autopilot, so we can at least fly it,” says Barry Davies. “They made an attempt to do that but it never flew properly. You had no control over it. It bounced, it did what it wanted to do.”

One of the problems with the Sparrow was its Wi-Fi control system. According to Davies, despite using very strong signals, interference caused the system to constantly lose its connection and fail. “He kept us on the hook all the time, saying, we’re there, we’re there, we’re there,” says Davies. “But it seemed to slip and slide and went on for three years.”

At the Innov8 innovation networking event at the Science Park in January 2014, Reedman held up an AV Sparrow to an audience of local businesses and said proudly, “This little guy is going into mass production now, which is excellent, and now we’re turning our hand to consumer technology.”

Nothing could be further from the truth, says Davies. “We tried to get the drone into production, to make 10 to see if they would work. None of them did. They just won’t go. [The project] was only recovered by having a search around the planet to find somebody who could make what we wanted in the first place. We’ve been back to our original goal and since then we’ve built a really good set of drones.”

Davies estimates that BCB spent hundreds of thousands of pounds with Torquing with little to show for it except a joint patent with Reedman for the drone’s unreliable obstacle-avoiding sonar. “We were really disappointed. It wasn’t so much the money, it was the time we spent,” he says. “We were way ahead of everybody else and we lost our slot.”

But Reedman wasn’t about to give up on three years’ work, although he now says that the Zano involves entirely different technology. At the Innov8 event, he revealed his plans to spin his Sparrow experience into something altogether more sexy. “In the next six to twelve months you’ll see an entire new product range…all assembled and made in the Science Park here,” said Reedman. “It’s going to be just the most incredible toy you’ve ever seen. We’re aiming to be selling hundreds of thousands of units a year.”

Zano was born.

The Bridge Innovation Centre on the Pembrokeshire Science & Technology Park / Mark Harris

Reedman says that Zano began with a larger, proof-of-concept drone called Bluepig: “It looked like a pig (it was ugly) and…blue…was the paint we had left over in the workshop when we built it.” Bluepig had a web-based ground control system that Reedman says could run on any modern smartphone, tablet or computer. A version of this was used to control early Zanos from smartphones before the apps were ready.

Zano’s Kickstarter campaign page claims that Reedman produced the first development boards for Zano in October 2013, then built a prototype miniature printed circuit board (PCB) by March the next year. But taking it further would require cash that Reedman lacked.

Luckily, Reedman had been busy networking with other companies at Pembroke Dock. Phil Busby is a local businessman who, in partnership with his father — confusingly, also named Phil Busby — has fingers in several pies. One of their companies, Mackenzie Corporation, has an interest in a scaffolding business. Another, Intelligent Trucking Solutions, was developing an automatic landing leg system for commercial trailers. The Velocity Drive aims to halve the time taken to hitch and unhitch trailers from lorries, and reduce driver injuries to boot.

Torquing Group designed and built an electronic control interface for the system, apparently successfully, as the Velocity Drive launched on the Australian market last summer. While they working on the Velocity Drive, Reedman pitched his idea for a high-tech, palm-sized drone to the Busbys.

They bit. Philip Andrew Busby and Philip Victor Busby jointly invested £75,000 and were appointed as directors of Torquing Group Limited on 21 March 2014, then as directors of Torquing Robotics and Torquing Technology on 31 July. Also investing £75,000 was another Australian, Reece Crowther, who joined the board of TGL on 30 January 2015.

Neither Crowther nor the Busbys responded to requests for interviews for this report. Crowther, a one-time professional goalkeeper with English Premier League club Crystal Palace, is thought by some Pembrokeshire County Council employees to have returned to Australia since the collapse of Torquing Group, but I was unable to confirm this.

With money in the bank, it was all systems go for Zano. Around May, Reedman spent £500 on an aging supercomputer from HPC Wales, the country’s national supercomputing service provider, which he intended to host a forum and process customers’ videos. Development boards for the drone itself were complete by June.

Torquing ordered 10 prototype Zanos for £25,000, and they arrived at Pembroke Dock in late September. These would be used to refine flight, avoidance, and video and photo capturing systems, but were obviously far too expensive as production units.

“To produce a consumer drone, we had to get a pretty low build cost,” says Reedman. “And the only way to reduce the build cost was to have some sort of commitment to reasonable volumes.”

The first step was to find a mass manufacturer. Reedman was keen to keep production as local as possible. He settled on Camtronics Vale, a contract electronics manufacturer with a 9,000 square metre facility in nearby Tredegar. “They were really, really excited about it, and had been right from day one. They invested heavily to support the project with special machines,” says Reedman. Camtronics declined my request for an interview.

With production in hand, the company now had to create demand. A booth at the CES show would cost a lot (over £50,000 by Reedman’s estimation), but would give Zano the chance to shine on a world stage. Then someone at Torquing, Reedman can’t remember who, suggested crowdfunding.

“They said Kickstarter was a great way to create a viral marketing campaign, to raise awareness of what we were trying to do,” he says. While Reedman continued to develop the Zano, he says others in the company, led by Crowther, now head of marketing, put together a Kickstarter webpage and promotional video. (Crowther appears prominently in the video.) Zano’s crowdfunding campaign officially launched on 24 November 2014, the jewel in the crown of Pembrokeshire Innovation Week.

A still from Zano’s promotional video on Kickstarter / Torquing Group

The Kickstarter video is a sore point with Zano backers, many of whom feel it gave an inaccurate impression of the drone’s capabilities and readiness. Doug Conran and Craig Holloway are two of only four Kickstarter backers who ultimately received Zanos in the autumn of 2015. (About 600 pre-order customers got theirs; more on this later.) As such, they are well placed to compare the performance of their Zanos with that seen in the video six months earlier.

“The video was a big part of my choice to support the project,” Holloway tells me over a pint in a quiet Swansea pub. The ex-Royal Air Force technician had planned to set up the drone to shoot footage of his wife as she rode her horse, like the mountain bike and dirt bike segments seen in the video. But when he first turned his Zano on, he got quite a surprise. “I tried it in a corridor and it didn’t work. It would just fly up a few centimetres and then land again. That’s all it ever did with the obstacle avoidance system on,” says Holloway.

Thinking that being indoors might be confusing the Zano, Holloway took it into his garden. “I tried it outside, turned the obstacle avoidance off and it flew off. I didn’t have a lot of control,” he remembers. “One time it flew off and I caught it and the propellers came off.”

Conran had a similar experience: “I unpacked my Zano and tried to fly it indoors, as I guess most people did. It really didn’t go very well at all, banging into the wall. So I took it down to a local park and tried to fly it there. After 10 or 15 seconds, it would just go off and do its own thing, zipping off sideways until it got out of range. Basically, it was just awful. The video was pretty poor quality too.”

How does Reedman reconcile the smooth, professional shots seen in the Kickstarter video with the erratic flight and grainy footage captured by Zanos months later? “It’s like any product promotional video,” he says. “You don’t just do one take. You might have to do 10 or 15 takes to get the best possible take. As [with] any commercial video, people are going to edit out the takes that aren’t as good.”

However, there are features shown working in the video that Reedman confirmed to me were not fully operational at the time it was shot. “It was very early stage prototype…and there were definitely things that didn’t work as planned. Initial tests [of autonomous hold position] were done in October 2014 as I recall. I then got around to finishing it off in June/July 2015,” he says.

The software that allowed a Zano to mimic gestures or automatically fly back to its owner was not even ready by the time the company folded. “I was planning to enable the return to base code…as soon as the app supported the various calibration and setup functions we needed,” says Reedman. “[Gesture control] was on my list to go live right after GPS functions were enabled.”

These features were included in the video because Reedman genuinely (but mistakenly) believed that he would be able to perfect them in time for the launch. “I don’t believe the video was misleading because Zano was very much about when, not if,” he tells me. “The platform was built to be upgraded.”

Reedman denies rumours among backers and other companies at the Science Park that Torquing used computer-generated imagery (CGI), other drones, or even selfie sticks to shoot the aerial videos. “To the best of my knowledge, wherever you saw a Zano flying, that was a Zano flying. There was absolutely no CGI in those videos,” he says.

At least one person appearing in the video was not a Torquing employee. David Thomas is the manager of the Bridge Innovation Centre, essentially Torquing’s landlord at the Technology Park. His employer, Pembrokeshire County Council, would not make him available for interview but did offer to forward my questions and his responses.

Thomas says that he was out for a walk with his wife when he was asked to pose for the scene shot in a pub garden. He says that there were a number of takes and that on each take, the Zano launched directly from his hand. However, he “doesn’t know whether the selfie was taken by the drone or by another camera” and “acknowledges that what was in the video may perhaps be misleading.” Interestingly, Thomas’s youngest son, Sam, was hired by Reedman to edit the Kickstarter video and film subsequent video updates. I haven’t been able to obtain the unedited footage, as none of the parties that would have access, if the raw material still exists, responded to contact attempts. I could not track down Sam Thomas.

But the video has red flags throughout. The shots that are clearly made by a videographer and include a Zano in frame have the same characteristics as video that ostensibly was shot, or is labelled as shot, from the drone. The cliff-diving scene seems to use a lens and tracking the device lacks. The Zano is never shown in a continuous shot from take-off through flight. In one case, footage appears to be reversed (a landing is used as a take-off); in another, a shot of the Zano is likely slowed down.

In the motocross scene, effects matting is visible in the gloved hand manipulating the smartphone, which has a screen that appears too bright in the natural lighting. Together, these would suggest that the screen’s display was replaced in post-production.

Without a forensics investigation, it’s impossible to know which parts of the video that purport to use Zano footage or show it in motion are accurate representations of what the prototype could do at the time. Neither in the video nor in the campaign is there any mention of simulations or rendering. Kickstarter’s Hardware and Product Design Project Guidelines, updated in 2012, forbid both tactics:

“Product simulations are prohibited. Projects cannot simulate events to demonstrate what a product might do in the future. Products can only be shown performing actions that they’re able to perform in their current state of development.
“Product renderings are prohibited. Product images must be photos of the prototype as it currently exists.”

Only one person clicked the Report This Project button at the very bottom of Zano’s page during its Kickstarter campaign. This button allows any user (not just backers) to sound the alarm on projects they feel are implausible or suspicious. Kickstarter says that while a single report is low for a project this size, that report did trigger a review by an analyst on its Integrity team. Kickstarter received a further 62 reports after the project funded, most since October, but would not share any of them with me due to its privacy policy.

The video did the trick, then, and as Torquing’s Kickstarter campaign roared into life, the mood at Torquing was celebratory. On 1 December, the creators awarded themselves an early Christmas present: Increasing the annual remuneration of directors in Torquing Group to £50,000 each.

After Christmas, Crowther and a couple of Torquing employees headed to the US for the CES trade show. Coincidentally, the booth opposite theirs in the Las Vegas Convention Center housed another crowdfunded British drone company, Extreme Fliers. In August 2015, just as Torquing was struggling with the difficulties of volume production, an Indiegogo campaign for the Micro Drone 3.0 would net Extreme Fliers nearly $3.2m from over 32,000 backers.

The Micro Drone 3.0 is a palm-sized quadcopter that offers 720p streaming video, social media integration, and inverted flying. Back in January 2015 at CES, though, Extreme Fliers had only just finished basic engineering and 3D modelling. Vernon Kerswell, its London-based managing director, was keen to see his rival Zano up close.

“When I saw it on Kickstarter, I was like, wow! This is amazing!” says Kerswell in a phone interview. “Then I started digging a little bit deeper, analysing the tech behind it, and I realised, ‘Oh my god, there’s no way they’re going to do this.’ This is completely bonkers. You can’t do obstacle avoidance with infrared, and the 15-minute flight time was completely ridiculous.”

Kerswell walked over to Zano’s stand to introduce himself and dig for more technical details. The Torquing team closed ranks. “They said, we know who you are and we’re not explaining or showing you anything,” he says.

Later at the show, Kerswell heard that the Torquing team were telling people that Zano’s technology was far more sophisticated than Micro Drone’s. “People were coming up to our stand and saying, ‘Yours can’t do obstacle avoidance and swarming.’ But theirs couldn’t either! Then they started saying really bad things about Micro Drone. At one point, I literally had to wade into a conversation to defend what he was saying against us,” says Kerswell. “At least we had a drone flying!”

Show goers were offered multiple explanations why all the Zanos at Torquing’s booth were grounded. Kerswell was told by Crowther that there was too much Wi-Fi interference in the crowded hall. “This is ridiculous because my drone was tethered with my IP address [over Wi-Fi] absolutely fine,” he notes. Other attendees reported being told that Zano wasn’t flying because Crowther’s laptop had been stolen, because Torquing was worried about industrial espionage, or even because its technology was military specification and might get confiscated by the US authorities.

Reedman now admits that Torquing did not show an operational Zano at CES because it simply wasn’t ready. “Consumers expect to see a fully functional, everything working consumer product,” he says. “Zano wasn’t at that stage at that point. With that sort of publicity, you make one mistake, and that’s the end of the product.”

Instead, Crowther’s presentational skills paid off. Without seeing a Zano so much as spin its propellers, Engadget chose it as one of the two best robots or drones at the entire CES. Editor in Chief Michael Gorman defended Engadget’s choice to me by noting that its criteria are innovation, design, market appeal, and functionality, “meaning what are the device’s capabilities, not that we have seen it performing all these functions in person.” That subtle distinction was lost on many Zano backers, who took Engadget’s selection as yet another seal of approval.

By the end of the campaign, Zano was the most-funded Kickstarter project in Europe ever. On 8 January, 12,075 backers had funded it to the tune of £2,335,119. Almost all had chosen a reward level that included a Zano drone with an estimated delivery date of June 2015. Now the hard work of delivering all those pledges would begin.

Zano with both flavours of propeller / Mark Harris

Back in the UK, the Kickstarter funds arrived in Torquing’s bank account on 29 January 2015: £2,094,833 after a 5% commission to Kickstarter and 3–5% to its payment processor. Kickstarter offers no advice on how to cope with such an avalanche of cash, but neither did Torquing request it. The directors decided to route the funds through Torquing Robotics Limited, and sweetened the pot with over £177,000 from Torquing Technology Limited, comprising Crowther and the Busbys’ investments and a little more besides.

Web sales of Zano accessories would bring in another £48,281 over the year ahead, and pre-orders promised another £561,794. But there was a catch with this money. PayPal, which processed the vast majority of these web sales (but none of the original Kickstarter pledges), would only release funds to TRL once its pre-orders had been satisfactorily fulfilled. PayPal’s public rules on what it calls “pre-sales” indicate it casts a jaundiced eye on such orders in general, will hold funds for items that can’t be shipped within 20 days, and may require documentation to release money even when products have shipped. It was another incentive for Torquing to hit the ambitious June deadline.

“Nobody expected [the campaign] to go that well,” says Reedman. “With the benefit of hindsight, it would have been a lot better to have capped it and to have had people wanting the product but not being able to get hold of it. A large undertaking was having 12,000 people wanting to communicate with us. This ended up taking a small team of people, just handling the emails and correspondence.”

The overwhelming success of the Kickstarter drive also altered Reedman’s production plans: “If we’d doubled our goal, we would have had 1,500 drones [to build]. We could have tested every single drone, literally have had somebody flying each drone before they got sent out. Going from potentially building 1,500 to 10 times that number was a monstrous headache and created a lot of issues in itself.”

Reedman was confident that his design was sound and the technology on track, but he realised that TRL’s artisanal production plans would have to scale up massively. One option the company explored was building its own factory in the Technology Park. If Torquing could arrange the financing for a dedicated manufacturing facility, it would be more cost-effective in the long run than relying on Camtronics.

“Torquing had been to see the Welsh government about a grant and a loan, which I think the government was minded to support,” says Steven Jones, Head of Development for Pembrokeshire County Council (PCC). We were talking over a nice cup of tea in his expansive office at County Hall in Haverfordwest. In April, Reedman had formally approached Jones for the remaining funds to build the factory, about £1.6m. “At the time, this was exactly what we were looking for,” says Jones. “Here was a company that had started off with a couple of people, grown, had this extraordinary success with the Kickstarter campaign, and was now looking to create 150 to 200 jobs.”

Jones asked Reedman to submit complete business and marketing plans to the full council. “We don’t as a general rule provide significant grants, or loans. But we were prepared to do that for Torquing. The money would have come out of our capital program supporting building roads, schools and houses,” he says.

As it happens, PCC never had to choose between education, transportation infrastructure, social housing, and selfie drones. The first phase of Torquing’s assembly facility alone — gutting and refitting a block of growth units at the Technology Park — would have taken 12 to 18 months to complete. This was far too long to help with Torquing’s summer deliveries. In May, Reedman returned to Jones and said that TRL would revert to its original plan of using Camtronics.

“We said, fair enough, if you want to come back at any time, we’re here,” remembers Jones. “They never appealed to us for any support or help after that meeting in April. We probably did dodge a bullet.”

One of Torquing’s leased vehicles outside its units at the Technology Park, and packaging for Apple products shot through the window / Mark Harris

Meanwhile, TRL was growing fast, snapping up lab equipment, a £29,000 3D printer, high-end Apple Macs, and perhaps a few shiny toys. Rhys Norbury is office administrator at Jellagen, a startup at the Technology Park that produces medical collagen from jellyfish. “We knew that Torquing were a big company, were very fancy, and had a lot of money, mostly based off the fancy BMWs,” he tells me during my visit.

David Grant is Managing Director of Kestrel International Circuits, an English company that supplied Torquing with several PCBs for the Zano. “If you get a lump of money, maybe you do get a bit silly,” he tells me during a phone interview. “They all bought kit. There were some odd things, like Busby Junior driving around in a brand new BMW M4 and Busby Senior having a brand new M6.”

Torquing also went on a hiring spree for software, firmware, and hardware engineers, and for administrative staff, marketers, and web developers. Exactly how many employees the company had is difficult to say. Reedman claims that there were up to 28 employees at its height. Documents filed with the liquidators record only 16 employees in mid-November, although some (a handful, according to Reedman) had been laid off earlier in the month.

While I could not find any former employees willing to be identified for this report, several agreed to share their stories on condition of anonymity. One, an engineer who joined after CES, confirmed Crowther’s tale of a lost laptop and recalls a fun, busy office. “It was hard work and everybody believed in Ivan and the product until the last weeks,” he says. “We all gave up weekends and did [a] stupid amount of overtime without seeking remuneration. It was like being part of an adventure with a tough start but [then] to be a lot better off after a year…of hardship.”

Another technical employee echoes this stereotypical startup experience. “I met a bunch of very passionate, skilled, and dedicated persons,” he says. “Too many times we worked until 9 or 10pm, some weekends included, to accomplish tight deadlines planned by the directors. But we were happy to do that. 12,000 customers were waiting for the product, and international success was just around the corner.”

But even in such a close-knit office, some things were off limits to new employees. The second developer again: “During the first weeks I was there, Ivan was used to closing his office door when he was testing the [drone’s] flight. This is the reason why I can’t say if Zano was able to ‘fly’, ‘fly properly’, or something between the two, at that time.”

The truth is that Reedman’s technical team was struggling to deliver on his promises. After Reedman added the microSD card reader, he says that he had to change to larger motors, which had a knock-on effect on the Zano’s weight. Reedman spent valuable time reverse-engineering communications from the camera module, while building a user interface for the Android app took much longer than expected. Getting the much-vaunted “follow me” feature to work relied on smartphones’ GPS systems, which proved frustratingly imprecise.

Updates posted to Kickstarter posted a generally rosy picture, although there were some warning signs. On 2 February 2015, Crowther wrote, “The newest (and final) iteration of ZANO’s PCB flight frame arrived at the offices today.” According to an update three weeks later, they soon arrived again: “The final loaded pre-production PCB iteration’s [sic] arrived at our facility…20th February… The good news, so far everything is as operational and no unexpected hiccups have occurred! So we’re on track to commission a pilot build of 500 units at the end of March.”

That deadline came and went without notice. Then, on 14 April, the printed circuit boards arrived at the Technology Park once more. “The final iteration of ZANOs PCB flight frame arrived yesterday! It was delivered to Camtronics this morning to be loaded,” wrote Crowther the next day. The pilot build was now scheduled for the end of April.

A pilot build is traditional in mass manufacturing, as it allows a company to work out production and assembly bugs at a larger scale than one-off prototypes. However, at some point in the early summer, TRL’s directors made the fateful decision to skip a genuine pilot build and head straight into full production. The company would still make a few hundred units initially, but it would also commit to tens of thousands of components, enough to fulfil every Kickstarter reward and pre-order.

“I was very strongly opposed to going down the avenue of committing so much to stock, so early,” says Reedman. “I made my opinion known, but ultimately, what happened, happened. I understand the reasons why the operations team felt they needed to do that, because that would genuinely keep to the June shipping date.”

Although the other directors did not want to be interviewed for this report, some at Torquing remember things differently. One of the engineers, who may not have been privy to high-level discussions, says: “Specifications and timescales were set by Ivan, as well as sign-off on bill of material, prototype, production. At no point I felt the other directors put pressure on Ivan. Quite the opposite, they would have liked to hear about any struggles to rearrange schedule and cash-flow accordingly.”

The other developer blames a pervasive culture of overconfidence: “The word ‘risk’ seemed to not exist at Torquing. During the first months after the Kickstarter campaign, we were already the ‘winners’, the way ahead was a straight walk down from the top of a hill.”

But bumps in the road kept popping up. In late May, Crowther posted that some of Zano’s plastic parts had been delayed due to a tooling issue. The decision not pursue a pilot build was coming back to bite Torquing. Additions that Reedman made to his initial design, and the fact that some of the plastics supplied were heavier than expected, had ballooned Zano’s weight from 55g as a prototype to 70g in pre-production. With the original propellers, the Zano could now fly for only a couple of minutes between charges — a far cry from the quarter hour that Reedman had promised.

A bigger battery could increase flight time, and Reedman told me he was trying to boost the battery size from 750 mAh (milliampere hours, a measurement of discharge capacity over time) to 1,000 or 1,100 mAh before he left Zano. A review of comparable batteries designed for drones (from makers and third-party replacements) finds even custom-fit modules would weigh at least 30g for 1,000 mAh, seemingly impractical without further design changes.

His solution at the time was to send back the original propellers for larger ones. However, says Reedman, “As far as [the Chinese supplier] was concerned, the propellers did work so therefore are not faulty and would not accept returns.” Torquing was left having paid for tens of thousands of propellers it could not use.

TRL quickly ordered new, oversized propellers. But these had their own issues. The quadcopter blades were now so large that they were within a hair’s-breadth of touching each other. More seriously, they also flexed slightly when in motion. Any tiny variation in production could cause problems. Craig Holloway, who owned a set of each kind of propeller, says that the larger blades would occasionally clip the drone’s plastic frame and each other.

Reedman decided to retool the Zano’s plastics to slim the weight down by a few grams, and so remove the need for the larger propellers altogether. But even then the Zano stubbornly refused to fly for more than five minutes at a time. “When we finally got the correct weight plastics, I discovered a performance issue with the smaller propellers,” says Reedman. The original [production] propellers were generating about 15% less lift than the ones in his prototypes. He admits, “Propellers that met the original specification were never purchased before I left Torquing.”

A still from Zano’s promotional video on Kickstarter / Torquing Group

Although the June deadline was now long past, some progress had been made. On 19 June, Ivan posted that the HPC supercomputer was 95% built and configured, with a fast 1 gigabit-per-second optical fibre connection. “A little more work and testing [and] this will be ready before the end of June,” he wrote. (Torquing’s “My Zano” online service to process, store, and share Zano owners’ videos never went live).

By 22 July, Reedman says that the drone had passed basic regulatory compliance tests, and Crowther felt confident enough to post that the first 2,000 Zanos would ship in early August, with all the remaining Kickstarter rewards following by the first week in September. Late — but not that late, especially by the standards of Kickstarter hardware campaigns in general, and large ones in particular.

Throughout the summer, Torquing had faced a constant barrage of questions, and, increasingly, criticism from backers eager to take delivery of their drones. “I didn’t think people would be that aggressive and nasty,” says Reedman. “I sort of learned to toughen my skin up a bit and not pay as much attention to it.” But by the start of August, Reedman says that he was suffering chest pain and having trouble sleeping. “[The doctor] said, you are just very, very stressed. You need to take time off,” he says.

There were signs that the company’s financial health was failing, too. On 21 August, the Torquing directors slashed their annual remuneration to £12,500 each. In September, Stuart Reedman, Ivan’s brother, who had been conducting some of Zano’s compliance tests, was told that TRL had cash-flow issues and that his outstanding invoices would be paid the following month. (They never were.)

In its dash to the finish line, Torquing needed every penny it could find. Kestrel’s David Grant says, “At first, for just a few bits and pieces, we were cash up front. When they moved to production, 20,000 units, we moved them to 30-day terms. They were unable to pay for the goods. They paid maybe £10,000 [and] kept telling us how they were trying to release funds for the products.”

The simplest way for Torquing to get more money would be to deliver the first Zano drones to pre-order customers. This would release hundreds of thousands of pounds that PayPal had been sitting on for months.

In an interview with BBC reporter Rory Cellan-Jones in August, Crowther admitted that the company was planning to send out Zanos before they were quite ready, saying, “We thought, the hardware’s done, the hardware is future-proofed, why delay?” Reedman agrees: “We started shipping units before they were ready.”

The first bulk deliveries of Zano, about 600 in all, began leaving Pembroke Dock on 24 September. Almost all were destined for pre-order customers, a move that angered many Kickstarter backers. A couple of backers, including Craig Holloway and Doug Conran, decided to do something about it. They contacted individuals within the company and pled for their Zanos.

In early October, Holloway and his family were invited out to the Technology Park. “Ivan gave us a bit of a tour,” says Holloway. “We were down there for probably three quarters of an hour, seeing all the products, all the various builds, and the software development guys.” Before Holloway left, he was given a Zano — although he couldn’t try it until the drone’s Android app was finally released the following week.

Footage from a backer who describes the Zano as “a continual nightmare of mega-underperformance.”

Within days of the app’s arrival, Torquing was drowning under a flood of complaints about drones bunny-hopping, flying off, or simply refusing to launch. “Things like that…caught us very much by surprise because we simply had not experienced that at all,” says Reedman. “When we started testing, about one in three drones, we found a few drones that suffered from that problem. Then it was a matter of going through, finding what was causing that problem, and fixing it.”

On 13 October, Ivan posted a Kickstarter update that blamed the problems variously on over-sensitive infrared sensors, manufacturing issues, propeller vibrations, and even drone owners’ smartphones. He announced that Torquing would release a software calibration tool to let owners address some of these at home. The very next day, Reedman says the company suffered a problem with its internet connection, bringing down the supercomputer, grounding all Zanos, and taking Torquing’s official forums offline.

Reedman soon got the high-performance server working again but decided to shelve the forums, which had become a hotbed of criticism, until they could “rework the forums into a proper customer support tool.” Bizarrely, Popular Science took this moment to proclaim Zano one of the 100 greatest innovations of 2015, raving, “It films you as you go about your vacation or your daily run. Presets can make it stay in one place, or track your every move.”

If only that were true. Holloway returned twice to the Technology Park in an effort to get his Zano up and flying properly. On the second occasion, on 30 October, Reedman personally downloaded the latest firmware onto Holloway’s drone and ran his new calibration tool.

“We got it flying [and] hovering around,” says Holloway. “Turn the obstacle avoidance on and put your hand close to it, and it would fly away really well…It was getting better.”

But the success was short-lived. The drone did not function as well when controlled by a different smartphone, the calibration software would not work on Holloway’s computer, and when he switched to a glow-in-the-dark Zano frame that he had paid extra for, the drone began bunny-hopping once more. None of the more advanced features — selfie videos, follow me, swarming, social media integration, or facial recognition — were ever operational.

Holloway’s drone, hand-delivered to him at launch, updated to the latest software, and fine-tuned personally by the company’s founder, was probably the best Zano in the world. And yet it still lacked the features and reliability of many cheap toy quadcopters from China.

Seeing it sitting on a pub table in Swansea, I caught a glimpse of the drone Reedman was trying to build. Its plastic case was well-engineered, the infrared sensors gave a nice futuristic touch, and the spare batteries and accessories felt solid. I could see the care and attention that had gone into its design: this was not an out-and-out scam. So as it pathetically blinked its red LEDs in an attempt to connect to an iPhone app that never existed and servers that were turned off, it wasn’t difficult to imagine just how disappointed those first owners would have felt.

While shipping to pre-order customers was bringing money into TRL, much of it was going straight back out the door again in refunds. Days later, Phil Busby called a meeting of directors and announced that there would need to be some redundancies. “We knew things were tight, but it was a shock how it happened,” says Reedman. “I think three or four, only technical people from my team, were laid off.”

Then, on 10 November, Reedman announced to his fellow directors that he was resigning from Torquing Robotics Limited, Torquing Group Limited, and Torquing Technology Limited with immediate effect. “It was largely due to my health and the fact that there was irreconcilable differences,” he says. “I felt that things I was very passionate about weren’t being shared by the rest of the company. The direction I felt things needed to go, that’s not how things were going and I couldn’t be part of it anymore.”

Reedman provided me with copies of documentation that confirm his disability status but would not be more specific about the irreconcilable differences, citing a “very restrictive employment contract and shareholders agreement.” He says that he walked away with no shares, no test equipment, no computers, no hand tools, “not even a screwdriver. I left with nothing.”

He left a group with shaky finances, around £1m worth of unfinished stock, a horde of unhappy customers, a single product that barely worked — and now no one even theoretically capable of addressing its numerous technical issues. “Without Reedman, they were finished,” says Kestrel’s David Grant. “They didn’t have a clue what to do without him. His resignation caused the death of Torquing Robotics.”

Its demise was rapid. On 13 November, the remaining directors voted to wind up the Torquing companies. On the evening of 16 November, Phil Busby met with a liquidator in Swansea, Gary Stones, then signed all the necessary paperwork the next morning. On 20 November, Stones sent out a notice of liquidation, and three days later, Kickstarter sent a private letter to backers announcing that it was banning Zano’s creators from future projects.

Up until the very last minute, many backers still believed that they would eventually receive their reward drones. Some vented their spleen online, and a few even lashed out at Gary Stones.

“A small percentage of Kickstarter backers were hostile,” he says. “I’ve had bad language and threats, but not that many and it has blown over. I did have a phone call from the police tipping me off that there was going to be a demonstration outside my office.”

That would have been on 4 December, when the one and only creditor’s meeting took place at Stones’ Swansea premises. No protesters showed up.

Zano was officially dead.

First and only annual profit and loss statement of Torquing Robotics Ltd / Mark Harris

Each of the Torquing companies has a so-called Statement of Affairs that details its assets and creditors at the time of liquidation. With a list of up to 17,000 possible creditors spanning 68 countries, Stones thinks that Torquing Robotics Limited’s Statement of Affairs is probably the longest such document ever filed in the UK. (This section may be tedious to readers without a vested interest in the precise financial details. Skip ahead a section if that fits you.)

A Statement of Affairs is a bit like a black box flight recorder. It can give you a good idea of what was happening at the precise moment of impact, but has little to say about how smooth or bumpy the ride was to get there. Only the companies’ audited accounts, which were not available to me, would show every penny coming in or going out over the preceding year.

Nevertheless, these documents, along with a single-sheet profit and loss account from TRL given to me by Gary Stones, are the most objective source of financial information about the rise and fall of the Torquing companies. I have rounded some of the figures below for clarity.

Let’s start with Torquing Technology Limited (TTL). It seems that this was Reedman’s primary company for the development work on the Sparrow drone with BCB International. Its gross revenue decreased from a high of over £80,000 in 2012 to £19,000 in 2014, probably representing the last of the BCB monies. But the company still had “administrative expenses” of nearly £69,000 in 2014, the nature of which are unclear. TTL owns the Sparrow patent (jointly with Barry Davies), and leased a Volkswagen Tiguan compact SUV, for which it still owes around £2,500.

(Just in case VW Financial Services (UK) Ltd of Milton Keynes is looking for it, this Tiguan was the only vehicle in the growth units car park when I visited the Bridge Innovation Centre in January.)

Torquing Group Limited (TGL) had a similarly low turnover — just £15,600 in 2014. While its expenses of £45,000 were less than TTL, it appears from the creditors list that TTL was responsible for paying for the entire group’s utilities, rent, waste disposal, lawyers, and accountants. TGL also owed £10,800 on the lease of two VW Golf cars, and £16,800 on the lease of a colour photocopier and “various Apple computer equipment”.

The trading accounts of Torquing Robotics Limited (TRL) are the most relevant to Zano. These run from December 2014 to November 2015, covering the full Kickstarter period. During that time, TRL earned £2,424,688 from product sales, and spent £1,556,628 on purchases. This would include all stock as well as any tools or machinery for design and assembly. Expenses and overheads swallowed up another £1,103,581.

It’s worth drilling down into these figures. Labour and wages accounted for nearly £540,000. If TGL did in fact have 30 employees, that would be an average salary of £18,000 — low even for rural Wales. Twenty staff would average to £27,000 each. A combined total for rent, rates, heat, light, and power of just £1,620 implies that TTL did, in fact, pick up most of these bills, while £3,500 for motor expenses, £11,300 for travelling, and even £51,000 for phone and computer charges are all plausible.

Several lines in the accounts do jump out, however. A “general expenses” figure of £186,293 could hide all manner of things, while professional fees of over £220,000 might be typical of a larger company. Reedman can shed no light on these. “I can’t think of anything. Fees relating to contractors [I] suspect would be under different areas,” he says.

The list of creditors is also illuminating. TRL took a loan of £7,000 from Bluestone Industries Limited, a company owned by Philip Busby, and one of £6,000 from Rockford Lane Limited, controlled by Crowther. Busby’s Intelligent Trucking Solutions was owed nearly £40,000 and his Mackenzie Corporation another £62,400. The Busbys were also personally owed £1,600 each as employees. Reedman was uncertain how the larger debts might have been incurred, telling me, “I am not aware what those amounts are for. No services had been paid for to either in the past.”

Stones says that Mackenzie paid some rent for the Torquing companies, charged for some of its own personnel that had worked at TRL, and even repaid some loans. “As far as I can see, no fraud has been committed and there were no signs of extravagance,” says Stones during an interview at his offices. “Phil Busby was throwing money at the company. [He] told me that he is £250,000 down.”

Some valuable assets do remain. Unfinished stock that cost around £1m to buy remains at Camtronics’ facility in Tredegar. Camtronics claims that the chips and components belong to them, to offset unpaid invoices and consequential losses as a result of buying machinery to build the Zano. There is also the matter of the unpaid monies sitting in PayPal’s bank accounts: Around £200,000 but shrinking all the time as pre-order customers request refunds. Gary Stones is considering legal action to recover some or all of both.

My editor Glenn subsequently explored the PayPal issue in more detail here, as some pre-order customers are currently in a Catch 22 situation of having had neither a drone (working or otherwise) nor a refund. Incredibly, it seems that their funds may now end up in the pocket of the liquidator or trade creditors rather than being returned to them.

Stones’ expenses in liquidating the business will likely swallow up all of the money from selling off Torquing’s intellectual property, lab equipment, and office supplies. If there are funds left over, the UK government will seize a slice for unpaid tax and social security contributions, leaving no more than pennies on the pound for trade creditors. The position of Kickstarter backers is murky. Without an actual sales contract, they may not legally be considered creditors and thus have virtually no chance of them receiving anything. As Kickstarter itself likes to say, Kickstarter is not a store.

Craig Holloway’s Zano attempting to connect to Torquing’s server via smartphone app / Mark Harris

At liquidation, the three Torquing companies had a total deficiency (debts over assets) of £1,262,608. Even had Reedman stayed in his post and all the Kickstarter rewards and pre-orders shipped successfully, with not a single one returned for a refund, Torquing would have been facing a deficit of nearly a million pounds.

So what went wrong? Almost everyone I spoke with had an opinion, and (almost) none suspects foul play. Kestrel’s Grant says, “I don’t think they were criminal. Why would you buy all the parts if you didn’t intend to build the product? Why would you employ all those people? I think they were small businesspeople who bit off more than they could chew.”

With the sceptical eye of a liquidator, Gary Stones says, “The company prepared accounts to the bitter end, which is always a good sign. I think they were chronically overstocked…but the product had not been perfected. They could have waited until it was ready.”

Vernon Kerswell of Extreme Fliers goes further: “I knew it was a scam from the beginning. I don’t know how they managed to spend all that money and have none left. It would be hard to spend all that on manufacturing if you were doing it properly. $3m has definitely been enough for what we need [to build the Micro Drone 3.0]. It’s actually given us a massive war chest for future developments.”

Ivan’s brother, Stuart, says that he warned Torquing not to go hell for leather to meet the Kickstarter deadline. “I advised the operations team quite early on not to place large orders for manufacturing until smaller runs…of Zanos had been completed and all production issues ironed out,” he says. “I believe my advice was not followed, primarily as the Kickstarter interest was much larger than anticipated. Without additional funding, there was no hope of success from the moment the purchase orders for 20,000 units were placed.”

Reedman himself has a similar view. “In no uncertain terms, I believe the reason TRL failed was basically poor financial planning,” he says. “It was a commitment to stock before the product was ready, which ran the project into financial difficulties, which meant we didn’t have the resources to re-engineer or fix certain things.”

This is almost certainly true. But it raises the question of whether an extra week, month, or even a year would have resulted in a Zano capable of anything close to what was originally promised. On two occasions, Reedman had been given virtual free rein to build an autonomous Wi-Fi drone. Both times, and even with millions at his disposal, his achievements fell significantly short of his vision.

The truth seems to be that almost everyone at Torquing was out of their depth. It turns out that building a competitive, multinational consumer product company is difficult, and that making a niche-market widget for truckers does little to prepare you for developing, producing, and marketing an innovative drone.

Torquing’s directors managed their business poorly and spent the Kickstarter money too freely, but I’ve found no evidence that any of them ended up rich on the backs of the crowd.

The fancy leased cars and rented computers have been returned. Ivan Reedman lost his company, his reputation, and his tools, and is currently looking for work. Phil Busby jeopardised his other businesses in a failed attempt to prop up TRL, and (according to one person at the Technology Park) has had to lay off several people from those. The other directors, Anna Reedman, Thomas Dietrich, and Reece Crowther, will never see a penny of their investments again.

I’ll leave the last word here to Barry Davies of BCB. “I started to feel…that where we had gone wrong was listening to Ivan saying he could do this. He can’t do it. He just can’t,” he says. “Ivan was a good friend to me. If I saw him tomorrow, I would have more condolences for him than I would have anger. But at the end of the day, thinking of all the people who invested in his Kickstarter, he shouldn’t really have done it.”

The story doesn’t end here, however.

A still from Zano’s promotional video on Kickstarter / Torquing Group

A failure of this magnitude makes people sit up and take notice. Although the liquidation is likely to rumble on for months, Vernon Kerswell noticed a change almost immediately in interactions with his backers and the business community. “Zano… affected lots of people [and] it’s really put the heat on us. People have got their knives out now,” he says.

Many of the backers that got in touch with me, even ones who had previously lost money in unsuccessful crowdfunding projects, say that their experiences with Zano soured them on Kickstarter itself and crowdfunding in general.

There remains a possibility of criminal prosecutions. The Trading Standards division of Pembrokeshire County Council (which falls, ironically, under Steven Jones’ purview) opened an investigation into Torquing Robotics in October, shortly after the first Zanos were shipped out. So far, it has received over 250 complaints from owners around the world that the drone did not match the description on the original Kickstarter page, and in particular the promotional video.

Council investigators I spoke to believe that while crowdfunding backers might not legally be considered creditors, they do count as consumers under UK trading regulations. Trading Standards hopes to conclude its investigation by April 2016.

Beyond Torquing itself, what lessons can be learned from the collapse of the Zano project? Inevitably, much of the discussion focuses on Kickstarter’s role in the affair. Could the crowdfunding platform have done anything to avoid such a monumental failure, or even prevent such a dubious project from funding in the first place?

Just before Christmas, I had a long phone conversation with Yancey Strickler, co-founder and CEO of Kickstarter. I put it to him that many Zano backers I had heard from felt that Kickstarter had no skin in the game, that it was content to bank its 5% fee whether projects succeeded or failed.

“I feel Kickstarter shares in the risks, absolutely,” he bristles. “It’s our reputation on the line every time. The only reason that Zano was able to raise this amount of money is because we have done such a good job administering this platform and working with trusted creators over the last six years. And when something like Zano happens, that burns. And [it] affects how someone looks at the next Kickstarter project.”

Strickler says that he has read every email from disgruntled Zano backers, and was often surprised by what he read. “[There’s] seemingly some belief that we require creators to mail us their single prototype, that we spend a week play-testing it and mail it back to them,” he says. “We have these rules for no photo-realistic renderings…but practically speaking, those are hard things to enforce. The system is reliant on backers to make a decision.”

While Kickstarter does have some rules — the word “pre-order” is particularly taboo — the premise of the platform, he says, is that the crowd decides which projects get funded and which do not. “The worlds of technology and design have the lowest [funding] success rates of any on Kickstarter, somewhere in the 20% range,” says Strickler. “Also the system allows for backers to come to us…and say, hey, we think there’s something weird happening here, check it out.”

Strickler points to concerns raised about the feasibility of a laser-powered razor called Skarp, leading Kickstarter to remove it from the site after it had raised over $4m (but before it had funded). “That’s like a quarter of a million dollars that we said, OK, we will not collect on this because we think there are issues there,” says Strickler. Skarp, incidentally, went on to successfully raise $450,000 on Indiegogo, which has fewer rules and restrictions.

How about when a project massively overfunds, as with Zano? Strickler admits that there are no special procedures for that, but mentions a new “creator-only knowledge sharing space” called Campus that is meant to help newly funded creators find their feet.

Perhaps it would make sense to tag on an extra 1% fee in such cases, I suggest, to pay for an independent consultant or experienced project manager? Strickler is quiet for a moment. “I think that that’s the sort of thing that I do find compelling because it is on mission,” he says. “[But] I don’t know if that would have any effect on failure. Or if backers would feel better about the fact that Zano didn’t deliver even though they had Joe Fancypants as their mentor.”

Vernon Kerswell told me that he received his Micro Drone millions from Indiegogo in two batches, the second only once he had started production and could show invoices and plastic mouldings. “We’ve done some research on this,” says Strickler. “Hardware especially is very capital intensive right at the start. We found that it’s possible that in terms of failures there could be money returned to backers, but it also seemed quite likely that it would greatly increase the number of failures overall, because people wouldn’t have the money to actually get started making their thing.”

Strickler also rejects suggestions from backers for shipping fees and even Kickstarter’s 5% commission to be placed in an escrow fund and refunded to them if a project fails to deliver.

“If someone pledged $300 and they got $15 after it failed, what has been solved there?” he wonders. “It’s decent PR but I don’t really know that that is helping to solve the issues. In reality…at a certain scale, it could put the world of crowdfunding at risk, period.”

His final statement might sound hyperbolic, but Strickler does make some good points. If we as a society value the benefits of crowdfunding — supporting projects that might not otherwise happen, feeling a tangible part of something we believe in, and possibly, just possibly, getting something cool at the end — then we have to be prepared to assume some risk.

In December, Kickstarter published research from Professor Ethan Mollick from the Wharton School of the University of Pennsylvania. Mollick surveyed over 47,000 backers of over 65,000 projects successfully funded on Kickstarter. He found that 9% of projects failed to deliver awards, and that failure rates for hardware projects were not significantly higher than other categories.

“If you want 100% success with hardware and new products, I think the only solution is that you just shop on Amazon,” says Strickler. “ We’re all still searching for what is the right playbook for these moments of extreme catastrophe. Thankfully, it’s been six years and we’re experiencing one on the scale of Zano really for the first time…Let it be an educational moment for a lot of people, and really very clearly set the expectations here.”

That might sound harsh to the over 12,000 backers of Zano; it certainly sounds harsh to me. But if crowdfunding is going to have a future in the mainstream, beyond toys, gadgets and GarageBand albums, the crowd will need to grow and mature along with it.

UK market analyst firm Juniper Research predicts that that investments made in technology via crowdfunding platforms will increase sevenfold from an estimated $1.1 billion last year to $8.2 billion by 2020. If we want a democratic, open, freely accessible alternative to banks and venture capitalists, then we will have to accept occasional failures like Zano along with runaway successes like Pebble, Oculus Rift, and Veronica Mars.

A still from Zano’s promotional video on Kickstarter / Torquing Group

I think we can do better than we are right now. For what it’s worth, and as requested by Kickstarter, here a few observations and suggestions from my time looking into Zano.

Creators should go in the crowdfunding process with as few illusions as possible. I would urge them to be as objective as possible about their project, to seek expert advice from outside their bubble, and above all to be as honest and transparent about their failures as they are celebratory with their successes. Also, avoid relying exclusively on the technical abilities of a single person.

Crowdfunding platforms can do more. If they are serious about not being considered stores, they should resist the temptation to mimic retail websites. One glance at the home pages of Amazon, Kickstarter, and Indiegogo is enough to see why a first-time backer might be confused. Project pages should explicitly state that Kickstarter has not seen, nor cares to see, product prototypes, and that Top Picks or Project We Love badges come with no endorsement.

It is in everyone’s interest — creators, backers, and platforms alike — that weak projects are weeded out before they are funded. The Report This Project button at the very bottom of each Kickstarter page is vague and easy to overlook. I would suggest replacing it with a Convince Me button in a prominent location. If enough registered users click on this, the project would receive additional scrutiny from the platform’s Trust or Integrity teams. (Also, platforms should fund robust Trust and Integrity teams.)

External support and advice is invaluable. Once funded, platforms could automatically assign every creator a mentor from a project in the same category that had already delivered rewards. There are now tens of thousands of such projects across the platforms. Mentors would be volunteers at first, but after one or two iterations, this would hopefully become an accepted and valued part of the fabric of crowdfunding, and organically build into a supportive community.

High value and massively overfunded projects require special attention. Platforms should recognise the difficulties inherent in manufacturing physical items, and in dealing with tens of thousands of backers. That could mean managing the funding process, requiring projects to hire external experts, or even automatically pushing estimated delivery dates out into the future for rewards in batches as orders increase — something the site Crowd Supply already builds into its process.

Kickstarter and others could also impose review requirements on videos and specifications once a threshold of, say, $1m, is passed. This might have revealed the problems with Zano’s promotion, and required it to be re-edited or demonstrate through raw footage that no simulations or CGI was used.

Journalists need to bring the same scepticism to crowdfunding they do to established companies that announce products long before they ship (known as vaporware), without the proven capability or any apparent rush to bring it to market.

The crowd itself needs to step up. Potential backers should do more to educate themselves and each other. While unbridled enthusiasm is always welcome, the most useful thing backers can bring to any project is helpful, thoughtful criticism. And before anyone hits a button to Back This Project, they need to remember that pledges are at best a gamble and just as likely an outright donation.

Finally, crowds are great but mobs are not. Having ten thousand people actively following your every move can be overwhelming even when things are going well. Rants, accusations, and personal threats are hurtful and invariably counterproductive.

Graphics for a Zano card game, to be packaged in Zano clamshells / Alex Mühlhölzl

The Zano story still has a host of loose ends. A better picture of the personal and professional inter-relationships of the Torquing directors could do much to explain why the company made what were, in hindsight, truly terrible business decisions. I wish I had uncovered more concrete details of the Torquing companies’ financial dealings, although more information could emerge once the liquidation process is complete. I also have my fingers crossed that Pembrokeshire Trading Standards will get to the bottom of the, I believe, quite misleading Kickstarter video.

I would very much like to have talked properly to the other Torquing directors, in particular Reece Crowther and the Busbys. This very nearly happened. As I was sitting in the car park of the Bridge Innovation Centre, having knocked on every door in the growth units and fruitlessly peered at empty Apple Mac boxes through the dusty window of Torquing’s, I heard a tap on the door of my rented Ford Focus.

It was Phil Busby (junior) himself, who had seen me nosing around. He explained that he did not want me to think that he was avoiding me because he had anything to hide. Although he declined to be interviewed or quoted on the record, Busby was insistent that he had done nothing wrong. He gave the impression that he might be more forthcoming after the Torquing companies had gone through liquidation, and the Trading Standards investigation was over. I do hope so.

Even more intriguing, there is the faintest of possibilities that Zanos might one day fly again — albeit in their own wobbly way. “Torquing Robotics has failed. But I don’t think the Zano project has in any way,” says Reedman. “It’s simply on hiatus whilst all the regulatory and all the other issues are sorted out. The technology was so close. I think there’s a very distinct possibility that the project itself will continue.”

Reedman tells me that he has had a number of discussions, both commercial and open source, about resurrecting Zano: “If somebody comes in and buys all the intellectual property and continues the project, or if it goes down open source route, that would be absolutely fantastic. I’d be the first person to put my hand up to be a contributor.”

More wishful thinking perhaps. But if you are one of the “lucky” few ever to receive a Zano, you might not want to recycle it just yet.

In the meantime, there could be one last chance for backers to get their hands on something made by Zano. Entrepreneur Alex Mühlhölzl co-founded Jellagen, the jellyfish harvesting company also based at Pembroke Dock. He was around to witness the rise and fall of Torquing and has now decided, with the help of some former employees, to create a Zano card game. “The game is a mockumentary of the business,” he tells me. “[It will have] things like fund raising, spin doctoring, ordering the wrong parts, staff fleeing and liquidators, in card form.”

Mühlhölzl has acquired from the liquidator a number of plastic Zano clamshell cases to use as packaging for the game. If all goes well, he plans to launch it, on Kickstarter of course, in the next couple of weeks.

How many backers, I wonder, are ready to crowdfund another Zano?

Way too long; didn’t read

I can only apologise for the extreme length of this report. While no magazine or newspaper would have allowed me go into so much detail, I think Zano’s backers deserve the fullest possible picture of events. I hope that it remains readable and interesting, and that the minutiae of propeller plastics and inter-company loans were not too tedious.

If you couldn’t make it all the way through, I don’t blame you. Here’s a brief and partial summary.

  • Ivan Reedman’s Torquing brand traded variously as an IT consultancy, golfing tour operator, and software developer. The only product any of these enterprises made that I could identify as reaching the market is a wireless device that controls raising and lower trailer legs.
  • Torquing’s previous drone development work, a military surveillance quadcopter, was not completed to the satisfaction of its client. The drone never flew properly and did not enter production.
  • In the spring 2014, Reedman received private investment of around £150,000 to develop a palm-sized consumer drone called the Zano.
  • There is convincing evidence that the Kickstarter campaign video, released in November 2014, was misleading as to the existing capabilities and readiness level of the Zano.
  • The reason that the Zano was not shown flying at the CES trade show in January 2015 is that it was not capable of performing adequately.
  • The massive success of the Kickstarter campaign (20 times Torquing’s target) caused enormous difficulties for the Zano team, obliging them to develop additional features, as well as scale up communications and production by an order of magnitude.
  • Torquing directors may have awarded themselves higher salaries than necessary and spent money on superfluous items like cars, but there is no sign of sustained extravagance or criminal fraud.
  • Torquing did mount a serious, well-intentioned attempt to develop, manufacture, and deliver an intelligent autonomous consumer drone along the lines of their promises in the Kickstarter campaign. A seemingly dedicated staff couldn’t, in any case, meet the over-ambitious deadlines and specifications.
  • Torquing directors made a series of serious errors in committing the business to extremely high levels of stock in the absence of proven production models, or even fully functional prototypes.
  • Communications from the project creators to backers were, on the whole, regular and fairly honest. However, they were also incomplete, overconfident, and reflected a dangerous lack of self-awareness of the problems the company was making for itself.
  • Financial pressures led the creators to ship Zano units that they knew were not ready, and additionally to favour pre-order customers in the hope of receiving additional revenues.
  • The resignation of Ivan Reedman was the immediate cause of the directors of Torquing seeking liquidation, but the business was already on its last legs, with a shortfall of over £1m.
  • The liquidation is proceeding in a professional manner, but is unlikely to result in any refund, however small, to any Kickstarter backer.
  • Personally, I do not believe that the creators possessed the technical or commercial competencies necessary to deliver the Zano as specified in the original campaign.
  • Kickstarter, and other crowdfunding platforms, should reconsider the way that they deal with projects involving complex hardware, massive overfunding, or large sums of money. There should be better mechanisms to identify weak projects before they fund, as well as new processes to provide mentorship, support and expert advice to newly-funded projects.