After the tech incubator bubble

This article was taken from the June 2012 issue of Wired magazine. Be the first to read Wired's articles in print before they're posted online, and get your hands on loads of additional content by subscribing online.

Back in February 2009, Kristian Hiiemaa, then 29, bootstrapped Erply, a business-software startup, with three other developers in Tallinn, Estonia. They applied to Seedcamp, the London-based mentoring and investment programme, and were selected for Seedcamp Week -- in which startups are mentored -- the following September.

There, they dazzled investors at the pitching stage, known as Demo Day.

A tour of the US with other Seedcamp winners to meet investors and advisers ensued, and Hiiemaa's life changed. "Between the Monday and the Friday [of the US trip], Kris had two or three unsolicited term sheets to invest from tier-one Silicon Valley VC firms," recalls Saul Klein, Seedcamp's founder and a partner at Index Ventures. "He spent the next week or two in Tallinn fielding Skype calls from 20 VCs, including an investor in Microsoft."

In March 2010, less than a year after launch, Erply announced that it had closed a $2 million (£1.25m) funding round, led by Redpoint Ventures and Klein's Index. A year after that, they'd opened an office in New York and signed up clients including Elizabeth Arden, which now uses Erply's retailing platform across its chain. "From a snowbound Estonian office with 60s furniture to Elizabeth Arden on Fifth Avenue [in Manhattan]," says Klein. "It's like a fairy tale."

Erply's story demonstrates exactly what Seedcamp was set up to do, Klein says. "The core of this movement is to help talent and pull them in from the 'extremes', in remote locations like Estonia, or fragmented ecosystems, like Europe's, and get them from 0 to 60 in terms of advice and connections."

Since Seedcamp launched in 2006, a proliferation of accelerators has transformed the landscape for tech startups seeking early-stage investment in Europe. From Yandex Factory in Moscow to Dad.es in Madrid, Gamma Rebels in Warsaw, Copenhagen's roving Startupbootcamp, White Bear Yard in London and Birmingham's Oxygen Accelerator, Europe is now home to at least 50 programmes vying for entrepreneurs, with many pitching their own spins (shared office space, government cash, virtual incubation, intensive boot camps) on hot-housing the next coding entrepreneurs.

Inspired by the success of US programmes such as Y Combinator (Wired 07.11), founded in Silicon Valley in 2005, and Techstars, launched in 2006 in Boulder, Colorado, incubators are increasingly central to the tech ecosystem, because they sift thousands of companies in their selection each year (Seedcamp gets 3,000 applicants). This helps later investors find the right startups to back.

But is Europe producing the talent to meet the expectations of so many investors? Or could we be in a bubble where too many second-rate businesses are being nurtured? "I do think there's a surge," says Reshma Sohoni, a partner at Seedcamp. "And not only is the quality of the startups in question, but also the quality and experience of the folks running them."

Sohoni predicts the market will correct in the next few years.

The most effective programmes, argues Michael Jackson of Mangrove Capital Partners (a $500 million fund), are those that act as "a great filter". "Mentors throw projects up to us, but we're probably only going to take one of them," he says. "And they can't give us 198 pieces of junk, because we won't take their calls any more."

In its six years of existence, Mountain View-based Y Combinator (YC) has welcomed 316 startups. The pioneering fund designed a new way to back startups by investing an average of $18,000 twice a year (considered "living expenses") in lots of companies (around 60 in summer 2011) and inviting them to Silicon Valley for three months of intensive mentoring and project refining. For its investment, YC takes a stake of around six to seven per cent in each company. The process culminates in a nail-chewing Demo Day of pitching in front of power-broker financiers with billions at their disposal.

YC alumni consider the value of the programme to be twofold: the practical advice about fundraising, and the network of mentors -- which remains long after the startups disperse. "We were three tech guys who'd started companies successfully in the past, but had never raised funds before," says Allan Grant of social-marketing platform Curebit, part of YC's winter 2011 intake. "So learning how to fundraise helped us a lot."

For Ian Hogarth, cofounder of London-based Songkick [Wired 05.12], who attended the programme in summer 2007, it was the contacts book and instant induction to a tech elite that helped. "The alumni network has been hugely beneficial, and to this day we continue to receive advice and support from the YC partners," he says. "We are still close to many people from our batch of entrepreneurs, including Daniel Ha from Disqus, Sachin [Rekhi] from Anywhere.fm and Drew [Houston] and Arash [Ferdowsi] from Dropbox.

We've met more YC founders each year, as new folks go through it."

In November 2010 two investors, Ron Conway and Yuri Milner, offered $150,000 to every startup in the programme, ensuring that each business in that batch raised further cash after attending YC. So, as the programme's cofounder Paul Graham noted in Wired 07.11, a better gauge of YC's hit rate is to measure how many of the summer 2010 cycle -- 36 startups -- managed to raise more money after the process. "Of those, 33 raised more money after YC, one didn't bother because they were already profitable, and two were not able to," he noted on the YC website. "So 34/36 or 94.4 per cent either raised money or didn't need to."

Graham went on to say that, although funding was simple to gauge, it wasn't "the real test". He observed: "What matters is how the companies end up doing. The standard test of that is its value." By the summer of 2010, YC had funded 208 startups. By looking at "the acquisition prices of the companies that have been acquired and trying to estimate the values of the rest", Graham calculated that the mean value of the startups funded by YC up to that point was $22.4 million. He conceded that startup valuations are "volatile" and that "six years in, all we can say is the numbers look encouraging".

YC has had three breakout hits: file-hosting service Dropbox, valued at $4 billion at its last $250-million funding round; accommodation rentals network Airbnb, valued at $1.3 billion; and Heroku, a cloud application platform, which was acquired in 2010 for $212 million by Salesforce.com. The trio combined are worth about $6.5 billion. This, argues Tom Hulme, design director of IDEO with close links to another accelerator, HackFwd, means that YC and other accelerators have rebooted the early-stage investment model. "This shows that the investment game is about not missing out on the big wins, rather than screening out losers. This is affecting valuations, so smart investors, such as Yuri Milner, are willing to pay more to ensure they're not missing out on hot deals.

For the accelerators it's a reminder that it's a numbers game -- they need to cast the net wide to catch the rare big fish."

But Reddit cofounder Alexis Ohanian, a YC alumnus who's now an "East Coast ambassador" for the programme, sees nothing wrong with this policy, as it tilts power away from institutional investors in favour of startups, creating far more viable companies. "There is clearly a desire to hit huge home runs," he says. "But what's nice about the YC model is that it also succeeds when you have smaller exits. All of these VC firms, which were based around the idea that founders needed to beg for money, are now in this awkward position where YC and others have built a model that doesn't need tonnes of money to be dumped in, with the hope they'll get the home runs to make up for the failures. YC's model means they invest around $20,000 a startup, and the goal is to create lots of great companies."

Last November, Birmingham-based accelerator Oxygen held its inaugural investor demo day at Birmingham Science Park Aston. Nine startups pitched their business ideas to an audience of about 100, among them potential investors. Although at times it felt more like a regional sales conference than the nerve-fraying hell described by those who have attended demo days at the leading US programmes, there was much at stake for the fresh-faced entrepreneurs -- including a £75,000 cash prize, with no equity strings attached.

Those who claim there is a bubble in the European incubator/accelerator space have unproven schemes such as Oxygen in mind. So, four months on, what had Oxygen's hopeful's achieved? Cofounder Mark Hales, an entrepreneur who is also a founding investor in the Cambridge accelerator Springboard, told Wired that "five of the teams have raised or are in the final stages of raising" the investment they were seeking -- which varied between £100,000 and £200,000.

Mission accomplished, then. But does Hales think there is a bubble? After a pause, he agrees that programme quality is distinctly variable: "We have come across a few teams that have been through accelerators already but, frankly, it didn't consist of much more than a desk, a phone and a bit of broadband. That's not going to add much value. Startups need to ask hard questions of the people running the programmes about their track record and their mentor base. Find out how much real access they'll get. It is becoming more competitive [among accelerators] -- and that's fantastic for entrepreneurs."

Others familiar with the sector fear that an oversupply of programmes will lower the quality threshold for admitting entrepreneurs. More programmes will mean fewer top-tier mentors and dwindling interest from investors, which could lead to yet lower quality in the subsequent wave of firms, making it harder for the next truly disruptive tech company to be created in Europe. "There are definitely a lot more

[accelerators] than we need," says Tine Thygesen, Copenhagen-based CEO of Everplaces and cofounder of FoundersHouse, who also mentors at Seedcamp and StartupBootcamp. "In Europe, we see a lot of funds poured into 'incubators' both by the EU and national and local governments. A lot of these programmes basically amount to nothing -- because what do governments know? They hire the wrong kind of people. The programmes that have done really well are run by successful entrepreneurs like [Y Combinator's] Paul Graham, not some consultant or coach." Alex Halliday, CEO of east London-based SocialGo, notes that this tail-off in quality is becoming a talking point in London's tech community. "High-quality accelerators are fantastic places to be in," he says. "But I think there is a growing sentiment that the quality has dropped dramatically as more and more people have tried to do it."

Blogger Jed Christiansen has compiled a list of about 20 European companies that meet his definition of "seed-accelerators". According to Christiansen, who works for Google but doesn't speak for them, whether there is a bubble or not is moot. "To me a bubble is when supply outstrips demand and no one realises it," he says. "In that sense I don't think there's a bubble because there are vast numbers of people who would get into entrepreneurial ventures if they had access to programmes. So I think there's quite a bit more demand than programmes that have risen up to meet it."

Paradoxically, Christiansen argues that many won't survive anyway for economic reasons. "Will all these programmes still be around in five years? I don't think so," he says. "A lot of them are getting started without fully understanding the survival rates and economics of the companies that come out of these programmes. To stay in business as a startup accelerator, you need a certain return to keep the programme ticking over."

In the short term, however, the number of accelerators is still rising. In March Telefonica said it would launch its latest startup incubator, Wayra Academy, in London, where it will hothouse up to 20 companies. With ever more such programmes across Europe, it's little wonder that accelerator league tables (including one commissioned by the Kauffman Fellows Program in the US) are being published. Thygesen predicts that an Ivy League-style system will emerge, with the elite institutions sparking intense competition from aspiring entrepreneurs. "A couple of [the programmes] will become so good that they'll provide real advantages to the startups in them," she says.

Christiansen agrees. "You'll have the Ivy League, the top-level state schools and the regular state schools. And that's fine. Some people will want to stay home and go to the University of Wyoming rather than try for the Ivy League. But they're still better off than if they hadn't gone anywhere."

In this analogy, Seedcamp is Europe's Harvard. Voted the best incubator in Europe in 2011 in a year-long independent study commissioned in the US by the Kauffman Fellows Program, it is younger than YC, meaning that there is less data available. After three-and-a-half years, Seedcamp partner Reshma Sohoni crunched the numbers to see what she could learn.

Compared with YC, Seedcamp backs relatively few startups, although the number grew from six in 2007 to 25 in 2011. But its investment in winning teams -- at €50,000 -- is larger. Of the first 22 Seedcamp companies, which received a total of €1.1 million, 90 per cent raised follow-on funding of between €300,000 and €2 million in the next round. The total raised by all 22 was more than €30 million, meaning that they raised more than 27 times Seedcamp's initial investment.

Just three of those 22 companies had closed by the end of 2011; three have been acquired (including Mobclix, bought for $50 million by Velti); and of the 16 still trading and independent from Seedcamp Fund 1, all have a product launched and all but one are generating revenue. Six of the companies invested in by Seedcamp in the first three years had generated more than $1 million in revenues by the end of 2011, with two more generating more than $500,000.

While Seedcamp helps European tech entrepreneurs based outside hub cities, another accelerator is trying a new model. Launched in 2010 by German Lars Hinrichs, the founder and former CEO of business-networking site XING, HackFwd is billed as "a pre-seed investment company". HackFwd has some golden rules, chief among them that you cannot apply to join the programme. HackBoxes (selected startups) have to approach a HackFwd "referrer", defined as "experts in their fields, and eminent startup or technology figures in their areas", who then refers the company to Hinrichs and his team. "We have two hypotheses,"

Hinrichs says. "The first is that there is enough coding talent in Europe that wants to go the entrepreneurial way. We have 45,000 computer-science students graduating in Europe each year. There are another 45,000 studying maths, physics or who are self-trained. So that's 90,000 people new to the pool every year. The second is that it's much tougher for software engineers to get funding in Europe, compared to America."

HackFwd hopes to use the first thesis to help build businesses based on raw talent to address the second. Selected (very early-stage) startups are offered a single "take it or leave it" deal: if they give HackFwd a 27 per cent stake (plus three per cent more to mentors), they receive between

€91,000 and €200,000, depending on the number of team members -- effectively paying salaries for a year. "We are coming in as early as possible, when they can't get funding from anyone else," says Hinrichs. "Prototype stage. The significant stake allows us to put in up to €200,000, because of the level of risk. I think it is the best offer in Europe currently."

Two other elements set HackFwd apart from other accelerators. Mentoring is conducted at arm's length.

Startups live and work at home, meeting quarterly at Build events in Majorca and most recently Berlin. Here they update each other and the HackFwd team on their progress over the last three months, listen to "inspirational" guest speakers and attend intensive sessions with experts in areas such as finance, design, technology, strategy and product. The second element is the high emphasis HackFwd places on technical expertise. All applicants have to go through a Codility software test and can progress only if they achieve a high score. "We go for technical excellence because we believe that when combined with our support network, it creates a better outcome than other types of investment," Hinrichs says.

Twelve startups have been through or are going through the HackFwd programme, of which two are closing "significant" venture rounds, and three are being further funded by the programme. Hinrichs concedes it's too soon to say if the system is a success. "But if you look at the value drivers, I would say we've changed the game," he says. "We want to create companies in the range of $10-$20 million plus."

Despite the proliferation of incubators, a genuinely world-changing company has still to emerge from Europe. Tom Hulme, however, insists it's simply a matter of time: "I would be stunned if we didn't see more than one multi-billion-dollar business coming out of Europe's accelerators within five years."

Mangrove's Michael Jackson, a former COO of Skype, agrees: "We have a tendency to expect miracles, but things take time."

**Case Study No.

1**

Chris Prescott, cofounder, Fantasy Shopper (HackFwd)

The British startup Fantasy Shopper figured that women enjoy browsing fashion sites and shopping for clothes and handbags online, yet they also find the process somewhat sterile. They miss the buzz of shopping with friends -- the instant opinions and postmortems. Fantasy Shopping's solution: social gamification.

Launched in October 2011 by a team based at the University of Exeter's Innovation Centre, the platform enables players to collect "fantasy money". They can spend it on real-world items from a range of high-street and designer fashion brands -- including New Look, Miss Selfridge, Karen Millen and All Saints -- to build virtual wardrobes, which friends can view. "Fantasy purchases" are posted to their Facebook newsfeeds, creating feedback loops and competition within the wider community.

Clicking once will then take the fantasy shopper through to the retailer's site to buy the real-world item. "The big disruptive influence we think we'll have on retail is that while users appreciate the utility value of online shopping, they want a more hedonistic experience," says cofounder Chris Prescott. "With Fantasy Shopper they can interact with friends, hunt for bargains and get validation for their purchase decisions before they buy."

Last November the site won the Grand Prize at Amazon's AWS Start-up Challenge 2011, beating global competition from 1,500 startups to win $50,000. In December the company completed its first month in UK-only beta mode, attracting 25,000 unique visitors who viewed more than 1.5 million pages, spending, on average, 28 minutes on the site. More than 50,000 fantasy outfits were created and £7 million of fantasy money spent. "Fantasy Shopper has insane metrics," says Lars Hinrichs, founder of HackFwd, the pre-seed investment company backing the startup. "It's rare to see numbers like these."

To date, Fantasy Shopper is the only UK startup to have received an invitation to join HackFwd's network -- all companies need to be referred. "Fantasy Shopper is a big project and there's no way we'd have been able to build it without HackFwd," says Prescott. "It has a mind--blowing network of advisers. The year's salary [that is awarded with selection for the programme] allowed us to concentrate 100 per cent on our project, and the meet-ups with other HackBoxes [fellow HackFwd startups] on the programme every three months in Majorca, and last time in Berlin, gave us intensive workshops on every aspect required for startup success: product design, technical implementation, marketing and business strategy."

He also reckons that a key part of the programme's formula is the emphasis that it places on technical proficiency. "HackFwd really gets into the tech substance," he says. "A very high level of technical competency is expected from all of its startups and every founder has to take a coding test to prove they have strong skills." How did Prescott fare? <span class="s1">"The scores aren't published," he replies, "but on the grapevine I've heard that some HackBoxes achieve 100 per cent."

**Case Study No.

2**

Matija Kopic, CEO, Farmeron (Seedcamp; 500 startups)

While helping his father out on the family farm in Croatia, computer-engineering student Matija Kopic saw a way of combining his passion for coding with the family trade. Despite undergoing a host of mechanical and chemical innovations, agriculture had remained largely untouched by digital technology.

For Kopic's father this meant updating a farm-data-management Excel spread-sheet after a hard day in the fields. "I discovered that there were no online solutions for proper data management in farming," says Kopic. "So we decided to develop one."

With business partner Marko Dukmenic, Kopic founded Farmeron and set about building a web-based application that's been described by TechCrunch as "Google Analytics for farms". It enables farmers to build a live data stream on all aspects of their business, from wheat storage as a percentage of total warehousing, to the amount of protein in an individual cow's milk at any milking. A contract with a large Croatian agricultural corporation to develop dairy-farm-management analytics led to orders from other farmers in the region. "By the time we had developed a generic product, we were getting enquiries from the US, Australia, Brazil, New Zealand and India," Kopic says. "That's when we started to look for investment."

The Farmeron team applied to Seedcamp, the east London-based early-stage mentoring-and-investment programme, where startups compete for funding. In July 2011, they attended a Mini Seedcamp pitching event in Ljubljana, Slovenia, where positive feedback -- and an acquisition offer that they eventually declined -- led to a place at Seedcamp Week (for the 20 selected companies) in London in September. Farmeron came fourth at the event, receiving 50,000 euros in investment and ongoing mentoring from the fund. The team received further (undisclosed) financial backing from the Silicon Valley accelerator 500 Startups, as well as an invitation to join that programme's "Fall 2011" batch of 30 companies. "We estimate there are 164 million farmers worldwide with access to the internet who are our target market," says Kopic. Both Seedcamp and 500 Startups' mentor networks are advising them on how to expand their business internationally. "As we already had some revenue, the investments weren't crucial for our immediate survival. But we didn't know anything about the global market and how to scale the --business to accommodate different geographies and jurisdictions. So we needed that input from our mentors, people who have built these kinds of businesses."

By the end of 2011, Farmeron had been used by 200 corporate farms in 14 countries. "Agriculture still lags really far behind technology," says Seedcamp partner Reshma Sohoni. "What attracted us to Matija was his vision for bringing agriculture into the 21st Century. To him, farming's a sexy world."

This article was originally published by WIRED UK