London
Auto Added by WPeMatico
Fabula AI is using social spread to spot ‘fake news’
UK startup Fabula AI reckons it’s devised a way for artificial intelligence to help user generated content platforms get on top of the disinformation crisis that keeps rocking the world of social media with antisocial scandals.
Even Facebook’s Mark Zuckerberg has sounded a cautious note about AI technology’s capability to meet the complex, contextual, messy and inherently human challenge of correctly understanding every missive a social media user might send, well-intentioned or its nasty flip-side.
“It will take many years to fully develop these systems,” the Facebook founder wrote two years ago, in an open letter discussing the scale of the challenge of moderating content on platforms thick with billions of users. “This is technically difficult as it requires building AI that can read and understand news.”
But what if AI doesn’t need to read and understand news in order to detect whether it’s true or false?
Step forward Fabula, which has patented what it dubs a “new class” of machine learning algorithms to detect “fake news” — in the emergent field of “Geometric Deep Learning”; where the datasets to be studied are so large and complex that traditional machine learning techniques struggle to find purchase on this ‘non-Euclidean’ space.
The startup says its deep learning algorithms are, by contrast, capable of learning patterns on complex, distributed data sets like social networks. So it’s billing its technology as a breakthrough. (Its written a paper on the approach which can be downloaded here.)
It is, rather unfortunately, using the populist and now frowned upon badge “fake news” in its PR. But it says it’s intending this fuzzy umbrella to refer to both disinformation and misinformation. Which means maliciously minded and unintentional fakes. Or, to put it another way, a photoshopped fake photo or a genuine image spread in the wrong context.
The approach it’s taking to detecting disinformation relies not on algorithms parsing news content to try to identify malicious nonsense but instead looks at how such stuff spreads on social networks — and also therefore who is spreading it.
There are characteristic patterns to how ‘fake news’ spreads vs the genuine article, says Fabula co-founder and chief scientist, Michael Bronstein.
“We look at the way that the news spreads on the social network. And there is — I would say — a mounting amount of evidence that shows that fake news and real news spread differently,” he tells TechCrunch, pointing to a recent major study by MIT academics which found ‘fake news’ spreads differently vs bona fide content on Twitter.
“The essence of geometric deep learning is it can work with network-structured data. So here we can incorporate heterogenous data such as user characteristics; the social network interactions between users; the spread of the news itself; so many features that otherwise would be impossible to deal with under machine learning techniques,” he continues.
Bronstein, who is also a professor at Imperial College London, with a chair in machine learning and pattern recognition, likens the phenomenon Fabula’s machine learning classifier has learnt to spot to the way infectious disease spreads through a population.
“This is of course a very simplified model of how a disease spreads on the network. In this case network models relations or interactions between people. So in a sense you can think of news in this way,” he suggests. “There is evidence of polarization, there is evidence of confirmation bias. So, basically, there are what is called echo chambers that are formed in a social network that favor these behaviours.”
“We didn’t really go into — let’s say — the sociological or the psychological factors that probably explain why this happens. But there is some research that shows that fake news is akin to epidemics.”
The tl;dr of the MIT study, which examined a decade’s worth of tweets, was that not only does the truth spread slower but also that human beings themselves are implicated in accelerating disinformation. (So, yes, actual human beings are the problem.) Ergo, it’s not all bots doing all the heavy lifting of amplifying junk online.
The silver lining of what appears to be an unfortunate quirk of human nature is that a penchant for spreading nonsense may ultimately help give the stuff away — making a scalable AI-based tool for detecting ‘BS’ potentially not such a crazy pipe-dream.
Although, to be clear, Fabula’s AI remains in development at this stage, having been tested internally on Twitter data sub-sets at this stage. And the claims it’s making for its prototype model remain to be commercially tested with customers in the wild using the tech across different social platforms.
It’s hoping to get there this year, though, and intends to offer an API for platforms and publishers towards the end of this year. The AI classifier is intended to run in near real-time on a social network or other content platform, identifying BS.
Fabula envisages its own role, as the company behind the tech, as that of an open, decentralised “truth-risk scoring platform” — akin to a credit referencing agency just related to content, not cash.
Scoring comes into it because the AI generates a score for classifying content based on how confident it is it’s looking at a piece of fake vs true news.
A visualisation of a fake vs real news distribution pattern; users who predominantly share fake news are coloured red and users who don’t share fake news at all are coloured blue — which Fabula says shows the clear separation into distinct groups, and “the immediately recognisable difference in spread pattern of dissemination”.
In its own tests Fabula says its algorithms were able to identify 93 percent of “fake news” within hours of dissemination — which Bronstein claims is “significantly higher” than any other published method for detecting ‘fake news’. (Their accuracy figure uses a standard aggregate measurement of machine learning classification model performance, called ROC AUC.)
The dataset the team used to train their model is a subset of Twitter’s network — comprised of around 250,000 users and containing around 2.5 million “edges” (aka social connections).
For their training dataset Fabula relied on true/fake labels attached to news stories by third party fact checking NGOs, including Snopes and PolitiFact. And, overall, pulling together the dataset was a process of “many months”, according to Bronstein, He also says that around a thousand different stories were used to train the model, adding that the team is confident the approach works on small social networks, as well as Facebook-sized mega-nets.
Asked whether he’s sure the model hasn’t been trained to identified patterns caused by bot-based junk news spreaders, he says the training dataset included some registered (and thus verified ‘true’) users.
“There is multiple research that shows that bots didn’t play a significant amount [of a role in spreading fake news] because the amount of it was just a few percent. And bots can be quite easily detected,” he also suggests, adding: “Usually it’s based on some connectivity analysis or content analysis. With our methods we can also detect bots easily.”
To further check the model, the team tested its performance over time by training it on historical data and then using a different split of test data.
“While we see some drop in performance it is not dramatic. So the model ages well, basically. Up to something like a year the model can still be applied without any re-training,” he notes, while also saying that, when applied in practice, the model would be continually updated as it keeps digesting (ingesting?) new stories and social media content.
Somewhat terrifyingly, the model could also be used to predict virality, according to Bronstein — raising the dystopian prospect of the API being used for the opposite purpose to that which it’s intended: i.e. maliciously, by fake news purveyors, to further amp up their (anti)social spread.
“Potentially putting it into evil hands it might do harm,” Bronstein concedes. Though he takes a philosophical view on the hyper-powerful double-edged sword of AI technology, arguing such technologies will create an imperative for a rethinking of the news ecosystem by all stakeholders, as well as encouraging emphasis on user education and teaching critical thinking.
Let’s certainly hope so. And, on the educational front, Fabula is hoping its technology can play an important role — by spotlighting network-based cause and effect.
“People now like or retweet or basically spread information without thinking too much or the potential harm or damage they’re doing to everyone,” says Bronstein, pointing again to the infectious diseases analogy. “It’s like not vaccinating yourself or your children. If you think a little bit about what you’re spreading on a social network you might prevent an epidemic.”
So, tl;dr, think before you RT.
Returning to the accuracy rate of Fabula’s model, while ~93 per cent might sound pretty impressive, if it were applied to content on a massive social network like Facebook — which has some 2.3BN+ users, uploading what could be trillions of pieces of content daily — even a seven percent failure rate would still make for an awful lot of fakes slipping undetected through the AI’s net.
But Bronstein says the technology does not have to be used as a standalone moderation system. Rather he suggests it could be used in conjunction with other approaches such as content analysis, and thus function as another string on a wider ‘BS detector’s bow.
It could also, he suggests, further aid human content reviewers — to point them to potentially problematic content more quickly.
Depending on how the technology gets used he says it could do away with the need for independent third party fact-checking organizations altogether because the deep learning system can be adapted to different use cases.
Example use-cases he mentions include an entirely automated filter (i.e. with no human reviewer in the loop); or to power a content credibility ranking system that can down-weight dubious stories or even block them entirely; or for intermediate content screening to flag potential fake news for human attention.
Each of those scenarios would likely entail a different truth-risk confidence score. Though most — if not all — would still require some human back-up. If only to manage overarching ethical and legal considerations related to largely automated decisions. (Europe’s GDPR framework has some requirements on that front, for example.)
Facebook’s grave failures around moderating hate speech in Myanmar — which led to its own platform becoming a megaphone for terrible ethnical violence — were very clearly exacerbated by the fact it did not have enough reviewers who were able to understand (the many) local languages and dialects spoken in the country.
So if Fabula’s language-agnostic propagation and user focused approach proves to be as culturally universal as its makers hope, it might be able to raise flags faster than human brains which lack the necessary language skills and local knowledge to intelligently parse context.
“Of course we can incorporate content features but we don’t have to — we don’t want to,” says Bronstein. “The method can be made language independent. So it doesn’t matter whether the news are written in French, in English, in Italian. It is based on the way the news propagates on the network.”
Although he also concedes: “We have not done any geographic, localized studies.”
“Most of the news that we take are from PolitiFact so they somehow regard mainly the American political life but the Twitter users are global. So not all of them, for example, tweet in English. So we don’t yet take into account tweet content itself or their comments in the tweet — we are looking at the propagation features and the user features,” he continues.
“These will be obviously next steps but we hypothesis that it’s less language dependent. It might be somehow geographically varied. But these will be already second order details that might make the model more accurate. But, overall, currently we are not using any location-specific or geographic targeting for the model.
“But it will be an interesting thing to explore. So this is one of the things we’ll be looking into in the future.”
Fabula’s approach being tied to the spread (and the spreaders) of fake news certainly means there’s a raft of associated ethical considerations that any platform making use of its technology would need to be hyper sensitive to.
For instance, if platforms could suddenly identify and label a sub-set of users as ‘junk spreaders’ the next obvious question is how will they treat such people?
Would they penalize them with limits — or even a total block — on their power to socially share on the platform? And would that be ethical or fair given that not every sharer of fake news is maliciously intending to spread lies?
What if it turns out there’s a link between — let’s say — a lack of education and propensity to spread disinformation? As there can be a link between poverty and education… What then? Aren’t your savvy algorithmic content downweights risking exacerbating existing unfair societal divisions?
Bronstein agrees there are major ethical questions ahead when it comes to how a ‘fake news’ classifier gets used.
“Imagine that we find a strong correlation between the political affiliation of a user and this ‘credibility’ score. So for example we can tell with hyper-ability that if someone is a Trump supporter then he or she will be mainly spreading fake news. Of course such an algorithm would provide great accuracy but at least ethically it might be wrong,” he says when we ask about ethics.
He confirms Fabula is not using any kind of political affiliation information in its model at this point — but it’s all too easy to imagine this sort of classifier being used to surface (and even exploit) such links.
“What is very important in these problems is not only to be right — so it’s great of course that we’re able to quantify fake news with this accuracy of ~90 percent — but it must also be for the right reasons,” he adds.
The London-based startup was founded in April last year, though the academic research underpinning the algorithms has been in train for the past four years, according to Bronstein.
The patent for their method was filed in early 2016 and granted last July.
They’ve been funded by $500,000 in angel funding and about another $500,000 in total of European Research Council grants plus academic grants from tech giants Amazon, Google and Facebook, awarded via open research competition awards.
(Bronstein confirms the three companies have no active involvement in the business. Though doubtless Fabula is hoping to turn them into customers for its API down the line. But he says he can’t discuss any potential discussions it might be having with the platforms about using its tech.)
Focusing on spotting patterns in how content spreads as a detection mechanism does have one major and obvious drawback — in that it only works after the fact of (some) fake content spread. So this approach could never entirely stop disinformation in its tracks.
Though Fabula claims detection is possible within a relatively short time frame — of between two and 20 hours after content has been seeded onto a network.
“What we show is that this spread can be very short,” he says. “We looked at up to 24 hours and we’ve seen that just in a few hours… we can already make an accurate prediction. Basically it increases and slowly saturates. Let’s say after four or five hours we’re already about 90 per cent.”
“We never worked with anything that was lower than hours but we could look,” he continues. “It really depends on the news. Some news does not spread that fast. Even the most groundbreaking news do not spread extremely fast. If you look at the percentage of the spread of the news in the first hours you get maybe just a small fraction. The spreading is usually triggered by some important nodes in the social network. Users with many followers, tweeting or retweeting. So there are some key bottlenecks in the network that make something viral or not.”
A network-based approach to content moderation could also serve to further enhance the power and dominance of already hugely powerful content platforms — by making the networks themselves core to social media regulation, i.e. if pattern-spotting algorithms rely on key network components (such as graph structure) to function.
So you can certainly see why — even above a pressing business need — tech giants are at least interested in backing the academic research. Especially with politicians increasingly calling for online content platforms to be regulated like publishers.
At the same time, there are — what look like — some big potential positives to analyzing spread, rather than content, for content moderation purposes.
As noted above, the approach doesn’t require training the algorithms on different languages and (seemingly) cultural contexts — setting it apart from content-based disinformation detection systems. So if it proves as robust as claimed it should be more scalable.
Though, as Bronstein notes, the team have mostly used U.S. political news for training their initial classifier. So some cultural variations in how people spread and react to nonsense online at least remains a possibility.
A more certain challenge is “interpretability” — aka explaining what underlies the patterns the deep learning technology has identified via the spread of fake news.
While algorithmic accountability is very often a challenge for AI technologies, Bronstein admits it’s “more complicated” for geometric deep learning.
“We can potentially identify some features that are the most characteristic of fake vs true news,” he suggests when asked whether some sort of ‘formula’ of fake news can be traced via the data, noting that while they haven’t yet tried to do this they did observe “some polarization”.
“There are basically two communities in the social network that communicate mainly within the community and rarely across the communities,” he says. “Basically it is less likely that somebody who tweets a fake story will be retweeted by somebody who mostly tweets real stories. There is a manifestation of this polarization. It might be related to these theories of echo chambers and various biases that exist. Again we didn’t dive into trying to explain it from a sociological point of view — but we observed it.”
So while, in recent years, there have been some academic efforts to debunk the notion that social media users are stuck inside filter bubble bouncing their own opinions back at them, Fabula’s analysis of the landscape of social media opinions suggests they do exist — albeit, just not encasing every Internet user.
Bronstein says the next steps for the startup is to scale its prototype to be able to deal with multiple requests so it can get the API to market in 2019 — and start charging publishers for a truth-risk/reliability score for each piece of content they host.
“We’ll probably be providing some restricted access maybe with some commercial partners to test the API but eventually we would like to make it useable by multiple people from different businesses,” says requests. “Potentially also private users — journalists or social media platforms or advertisers. Basically we want to be… a clearing house for news.”
Powered by WPeMatico
pi-top’s latest edtech tool doubles down on maker culture
London-based edtech startup, pi-top, has unboxed a new flagship learn-to-code product, demoing the “go anywhere” Pi-powered computer at the Bett Show education fare in London today.
Discussing the product with TechCrunch ahead of launch, co-founder and CEO Jesse Lozano talked up the skills the company hopes students in the target 12-to-17 age range will develop and learn to apply by using sensor-based connected tech, powered by its new pi-top 4, to solve real world problems.
“When you get a pi-top 4 out of the box you’re going to start to learn how to code with it, you’re going to start to learn and understand electronic circuits, you’re going to understand sensors from our sensor library. Or components from our components library,” he told us. “So it’s not: ‘I’m going to learn how to create a robot that rolls around on wheels and doesn’t knock into things’.
“It’s more: ‘I’m going to learn how a motor works. I’m going to learn how a distance sensor works. I’m going to learn how to properly hook up power to these different sensors. I’m going to learn how to apply that knowledge… take those skills and [keep making stuff].”
The pi-top 4 is a modular computer that’s designed to be applicable, well, anywhere; up in the air, with the help of a drone attachment; powering a sensing weather balloon; acting as the brains for a rover style wheeled robot; or attached to sensors planted firmly in the ground to monitor local environmental conditions.
The startup was already dabbling in this area, via earlier products — such as a Pi-powered laptop that featured a built in rail for breadboarding electronics. But the pi-top 4 is a full step outside the usual computing box.
The device has a built-in mini OLED screen for displaying project info, along with an array of ports. It can be connected to and programmed via one of pi-top’s other Pi-powered computers, or any PC, Mac and Chromebook, with the company also saying it easily connects to existing screens, keyboards and mice. Versatility looks to be the name of the game for pi-top 4.
pi-top’s approach to computing and electronics is flexible and interoperable, meaning the pi-top 4 can be extended with standard electronics components — or even with Littlebits‘ style kits’ more manageable bits and bobs.
pi-top is also intending to sell a few accessories of its own (such as the drone add-on, pictured above) to help get kids’ creative project juices flowing — and has launched a range of accessories, cameras, motors and sensors to “allow creators of all ages to start learning by making straight out of the box”.
But Lozano emphasizes its platform play is about reaching out to a wider world, not seeking to lock teachers and kids to buying proprietary hardware. (Which would be all but impossible, in any case, given the Raspberry Pi core.)
“It’s really about giving people that breadth of ability,” says Lozano, discussing the sensor-based skills he wants the product to foster. “As you go through these different projects you’re learning these specific skills but you also start to understand how they would apply to other projects.”
He mentions various maker projects the pi-top can be used to make, like a music synth or wheeled robot, but says the point isn’t making any specific connected thing; it’s encouraging kids to come up with project ideas of their own.
“Once that sort of veil has been pierced in students and in teachers we see some of the best stuff starts to be made. People make things that we had no idea they would integrate it into,” he tells us, pointing by way of example to a solar car project from a group of U.S. schoolkids. “These fifteen year olds are building solar cars and they’re racing them from Texas to California — and they’re using pi-tops to understand how their cars are performing to make better race decisions.”
pi-top’s new device is a modular programmable computer designed for maker projects
“What you’re really learning is the base skills,” he adds, with a gentle sideswipe at the flood of STEM toys now targeting parents’ wallets. “We want to teach you real skills. And we want you to be able to create projects that are real. That it’s not block-based coding. It’s not magnetized, clipped in this into that and all of a sudden you have something. It’s about teaching you how to really make things. And how the world actually works around you.”
The pi-top 4 starts at $199 for a foundation bundle which includes a Raspberry Pi 3B+,16GB SD card, power pack, along with a selection of sensors and add-on components for starter projects.
Additional educational bundles will also launch down the line, at a higher price, including more add ons, access to premium software and a full curriculum for educators to support budding makers, according to Lozano.
The startup has certainly come a long way from its founders’ first luridly green 3D printed laptop which caught our eye back in 2015. Today it employs more than 80 people globally, with offices in the UK, US and China, while its creative learning devices are in the hands of “hundreds of thousands” of schoolkids across more than 70 countries at this stage. And Lozano says they’re gunning to pass the million mark this year.
So while the ‘learn to code’ space has erupted into a riot of noise and color over the past half decade, with all sorts of connected playthings now competing for kids’ attention, and pestering parents with quasi-educational claims, pi-top has kept its head down and focused firmly on building a serious edtech business with STEM learning as its core focus, saving it from chasing fickle consumer fads, as Lozano tells it.
“Our relentless focus on real education is something that has differentiated us,” he responds, when asked how pi-top stands out in what’s now a very crowded marketplace. “The consumer market, as we’ve seen with other startups, it can be fickle. And trying to create a hit toy all the time — I’d rather leave that to Mattel… When you’re working with schools it’s not a fickle process.”
Part of that focus includes supporting educators to acquire the necessary skills themselves to be able to teach what’s always a fast-evolving area of study. So schools signing up to pi-top’s subscription product get support materials and guides, to help them create a maker space and understand all the ins and outs of the pi-top platform. It also provides a classroom management backend system that lets teachers track students’ progress.
“If you’re a teacher that has absolutely no experience in computer science or engineering or STEM based learning or making then you’re able to bring on the pi-top platform, learn with it and with your student, and when they’re ready they can create a computer science course — or something of that ilk — in their classroom,” says Lozano.
pi-top wants kids to use tech to tackle real-world problems
“As with all good things it takes time, and you need to build up a bank of experience. One of the things we’ve really focused on is giving teachers that ability to build up that bank of experience, through an after school club, or through a special lesson plan that they might do.
“For us it’s about augmenting that teacher and helping them become a great educator with tools and with resources. There’s some edtech stuff they want to replace the teacher — they want to make the teacher obsolete. I couldn’t disagree with that viewpoint more.”
“Why aren’t teachers just buying textbooks?” he adds. “It takes 24 months to publish a textbook. So how are you supposed to teach computer science with those technology-based skills with something that’s by design two years out of date?”
Last summer pi-top took in $16M in Series B funding, led by existing founders Hambro Perks and Committed Capital. It’s been using the financing to bring pi-top 4 to market while also investing heavily in its team over the past 18 months — expanding in-house expertise in designing learning products and selling in to the education sector via a number of hires. Including the former director of learning at Apple, Dr William Rankin.
The founders’ philosophy is to combine academic expertise in education with “excellence in engineering”. “We want the learning experience to be something we’re 100% confident in,” says Lozano. “You can go into pi-top and immediately start learning with our lesson plans and the kind of framework that we provide.”
“[W]e’ve unabashedly focused on… education. It is the pedagogy,” he adds. “It is the learning outcome that you’re going to get when you use the pi-top. So one of the big changes over the last 18 months is we’ve hired a world class education team. We have over 100 years of pedagogical experience on the team now producing an enormous amount of — we call them learning experience designers.”
He reckons that focus will stand pi-top in good stead as more educators turn their attention to how to arm their pupils with the techie skills of the future.
“There’s loads of competition but now the schools are looking they’re [asking] who’s the team behind the education outcome that you’re selling me?” he suggests. “And you know what if you don’t have a really strong education team then you’re seeing schools and districts become a lot more picky — because there is so much choice. And again that’s something I’m really excited about. Everybody’s always trying to do a commercial brand partnership deal. That’s just not something that we’ve focused on and I do really think that was a smart choice on our end.”
Lozano is also excited about a video the team has produced to promote the new product — which strikes a hip, urban note as pi-top seeks to inspire the next generation of makers.
“We really enjoy working in the education sector and I really, really enjoy helping teachers and schools deliver inspirational content and learning outcomes to their students,” he adds. “It’s genuinely a great reason to wake up in the morning.”
Powered by WPeMatico
London’s transport regulator looks to startups to help fix urban mobility
London’s transport regulator, TfL has announced a partnership with Bosch for its forthcoming co-working space in Shoreditch.
The civic tech project is intended to run for 18 months as a pilot — though Bosch’s ‘Connectory’ co-working facility won’t open until the end of January. A company spokeswoman confirmed the partnership is nonetheless up and running now.
The aim of the collaborative project is to share data and expertise, including by tapping into London’s startup ecosystem, to land on new ideas for tackling urban mobility issues — from traffic jams to awful air quality.
Transport issues are especially pressing for the city as London’s population is forecast to reach a staggering 10.8 million by 2041 — which would mean around six million additional trips being generated per day.
Specific issues TfL is looking for help with include developing more efficient, greener and safer vehicles; reducing congestion; and encouraging more people to walk, cycle and take public transport across London, it said today.
TfL will be providing technical knowledge and “a wide range” of datasets throughout the pilot to allow participating companies to test ideas and “understand patterns in more detail than has previously been possible”, it added.
The data will be based on its existing Unified API and open data platform, which it notes is already underpinning nearly 700 apps used by approaching half (42 per cent) of Londoners.
Startups selected for the collaboration will be provided with dedicated space within Bosch’s Connectory, alongside TfL staff who will also be based there during the pilot.
Commenting in a statement, Arun Srinivasan, executive VP and head of mobility solutions at Bosch UK said: “We believe that the collaboration between Bosch and TfL will enable us to accelerate the development of technologies, products and services that have a positive impact on city life.”
Startups will be selected by Bosch, according to a TfL spokesman. We’ve asked for more details on selection criteria.
Update: A Bosch spokeswomen told us: “There will be a number of programmes running for start ups in the Connectory. These programmes will be around specific mobility challenges and many will have open calls for start ups to enter. We also welcome direct approaches by small business/start ups who want to be part of the Connectory community feel they have something to offer that will help solve London’s transport challenges. Get in touch!”
She said there is no fixed number of startup planned to be selected for the pilot, saying they will have rolling cohorts “designed around specific London mobility challenges” — launching this process in the New Year.
“This new ‘urban mobility’ lab is the first of its kind with a primary focus on urban mobility, and will provide the forum for private sector partners, academia and public sector to work together to tackle a range of problems facing Londoners in years to come,” the pair added in a press release today.
“By facilitating closer collaboration, TfL and Bosch hope to support start-ups to develop a range of smart products and help them identify ways to bring them to market more quickly through open procurement.”
The entire co-working facility is focused on urban mobility — but will also be open to other interested companies and startups to rent or bag a space (i.e. via Bosch’s scouting programs where it does take equity), not just to the startups selected for the TfL pilot.
Bosch’s network of Connectory co-innovation spaces also links out to cities internationally, including Chicago and Stuttgart, further expanding potential knowledge-sharing opportunities.
Commenting in a statement, the mayor of London, Sadiq Khan, said: “This initiative will foster closer working between London’s tech sector and other leading tech cities. If we are to use data and smart technology to help solve the biggest problems our city faces, it’s crucial we take a more collaborative approach. I see London’s future as a global ‘test-bed city’ for civic innovation, where the best ideas are developed, amplified and scaled.”
Depending on the outcome of the pilot, TfL said the Greater London Authority may seek similar collaborative approaches to support other aspects of its work — including housing, environment and policing, aligning with the mayor of London’s strategic priorities.
“I’ve been clear I want London to become the world’s smartest city and this is a further step towards realising that ambition,” Khan added.
This report was updated with additional detail about startup selection; and to correct that the lab will focus exclusively on urban mobility — but is also open to interested companies to rent space, as well as to startups Bosch selects to take a stake in
Powered by WPeMatico
Jane VC, a new fund for female entrepreneurs, wants founders to cold email them
Want to pitch a venture capitalist? You’ll need a “warm introduction” first. At least that’s what most in the business will advise.
Find a person, typically a man, who made the VC you’re interested in pitching a whole bunch of money at some point and have them introduce you. Why? Because VCs love people who’ve made them money; naturally, they’ll be willing to hear you out if you’ve got at least one money-maker on your side.
There’s a big problem with that cycle. Not all entrepreneurs are friendly with millionaires and not all entrepreneurs, especially those based outside Silicon Valley or from underrepresented backgrounds, have anyone in their network to provide them that coveted intro.
Jane VC, a new venture fund based out of Cleveland and London, wants entrepreneurs to cold email them. Send them your pitch, no wealthy or successful intermediary necessary. The fund, which has so far raised $2 million to invest between $25,000 and $150,000 in early-stage female-founded companies across industries, is scrapping the opaque, inaccessible model of VC that’s been less than favorable toward women.
“We like to say that Jane VC is venture for every woman,” the firm’s co-founder Jennifer Neundorfer told TechCrunch.
Neundorfer, who previously co-founded and led an accelerator for Midwest startups called Flashstarts after stints at 21st Century Fox and YouTube, partnered with her former Stanford business school classmate Maren Bannon, the former chief executive officer and co-founder of LittleLane. So far, they’ve backed insurtech company Proformex and Hatch Apps, an enterprise software startup that makes it easier for companies to create and distribute mobile and web apps.
“We are going to shoot them straight.” — Jennifer Neundorfer
Jane VC, like many members of the next generation of venture capital funds, is bucking the idea that the best founders can only be found in Silicon Valley. Instead, the firm is going global and operating under the philosophy that a system of radical transparency and honesty will pay off.
“Let’s be efficient with an entrepreneur’s time and say no if it’s not a hit,” Neundorfer said. “I’ve been on the opposite end of that coaching. So many entrepreneurs think a VC is interested and they aren’t. An entrepreneur’s time is so valuable and we want to protect that. We are going to shoot them straight.”
Though Jane VC plans to invest across the globe, the firm isn’t turning its back on Bay Area founders. Neundorfer and Bannon will leverage their Silicon Valley network and work with an investment committee of nine women based throughout the U.S. to source deals.
“We are women that have raised money and have been through the ups and downs of raising money in what is a very male-dominated world,” Neundorfer added. “We believe that investing in women is not only the right thing to do but that you can make a lot of money doing it.”
Powered by WPeMatico
VCs say Silicon Valley isn’t the gold mine it used to be
In the days leading up to TechCrunch Disrupt SF 2018, The Economist published the cover story, ‘Why Startups Are Leaving Silicon Valley.’
The author outlined reasons why the Valley has “peaked.” Venture capital investors are deploying capital outside the Bay Area more than ever before. High-profile entrepreneurs and investors, Peter Thiel, for example, have left. Rising rents are making it impossible for new blood to make a living, let alone build businesses. And according to a recent survey, 46 percent of Bay Area residents want to get the hell out, an increase from 34 percent two years ago.
Needless to say, the future of Silicon Valley was top of mind on stage at Disrupt.
“It’s hard to make a difference in San Francisco as a single entrepreneur,” said J.D. Vance, the author of ‘Hillbilly Elegy’ and a managing partner at Revolution’s Rise of the Rest Fund, which backs seed-stage companies based outside Silicon Valley. “It’s not as a hard to make a difference as a successful entrepreneur in Columbus, Ohio.”
In conversation with Vance, Revolution CEO Steve Case said he’s noticed a “mega-trend” emerging. Founders from cities like Pittsburgh, Detroit or Portland are opting to stay in their hometowns instead of moving to U.S. innovation hubs like San Francisco.
“The sense that you have to be here or you can’t play is going to start diminishing.”
“We are seeing the beginnings of a slowing of what has been a brain drain the last 20 years,” Case said. “It’s not just watching where the capital flows, it’s watching where the talent flows. And the sense that you have to be here or you can’t play is going to start diminishing.”
J.D. Vance says that most entrepreneurs don’t need to move to Silicon Valley.
Here’s why. #TCDisrupt pic.twitter.com/0mFPeTuHLe
— TechCrunch (@TechCrunch) September 6, 2018
Farewell, San Francisco
“It’s too expensive to live here,” said Aileen Lee, the founder of seed-stage VC firm Cowboy Ventures, amid a conversation with leading venture capitalists Spark Capital general partner Megan Quinn and Benchmark general partner Sarah Tavel .
“I know that there are a lot of people in the Bay Area that are trying to work on that problem and I hope that they are successful,” Lee added. “It’s an amazing place to live and we’ve made it really challenging for people to live here and not worry about making ends meet.”
One of Cowboy’s portfolio companies opted to relocate from Silicon Valley to Colorado when it came time to scale their business. That kind of move would’ve historically been seen as a failure. Today, it may be a sign of strong business acumen.
Quinn said that of all 28 of Spark’s growth-stage portfolio companies, Raleigh, North Carolina-based Pendo has the easiest time recruiting folks locally and from the Bay Area.
She advises her Bay Area-based late-stage companies to open a second office outside of the Valley where lower-cost talent is available.
“We often say go to [flySFO.com], draw a three-hour circle around San Francisco where they have direct flights, find a city that has a university and open up a second office as quickly as possible,” Quinn said.
Still, all three firms invest in a lot of companies based in San Francisco. Of Benchmark’s 10 most recent investments, for example, eight were based in SF, according to Crunchbase.
“I used to believe really strongly if you wanted to build a multi-billion dollar company you had to be based here,” Tavel said. “I’ve stopped giving that soap speech.”
Aileen Lee (Cowboy Ventures), Megan Quinn (Spark Capital), and Sarah Tavel (Benchmark Capital) on whether or not Silicon Valley is on the wane for investors #TCDisrupt pic.twitter.com/SOpn7p0eNQ
— TechCrunch (@TechCrunch) September 5, 2018
Underestimated talent
A lot of Bay Area VCs have been blind to the droves of tech talent located outside the region. Believe it or not, there are great engineers in America’s small- and medium-sized markets too.
At Disrupt, Backstage Capital founder Arlan Hamilton announced the firm would launch an accelerator to further amplify companies led by underestimated founders. The program will have cohorts based in four cities; San Francisco was noticeably absent from that list.
Instead, the firm, which invests in underrepresented founders and recently raised a $36 million fund, will work with companies in Philadelphia, Los Angeles, London and one more city, which will be determined by a public vote. Aniyia Williams, the founder of Tinsel and Black & Brown Founders, will spearhead the Philadelphia effort.
“For us, it’s about closing that wealth gap to address inequity in tech,” Williams said. “There needs to be more active participation from everyone.”
Hamilton added that for her, the tech talent in LA and London is undeniable.
“There is a lot of money and a lot of investors … it reminds me of three years ago in Silicon Valley,” Hamilton said.
Silicon Valley vs. China
Silicon Valley’s demise may not be just as a result of increased costs of living or investors overlooking talent in other geographies. It may be because of heightened competition abroad.
Doug Leone, an early- and growth-stage investor at Sequoia Capital, said at Disrupt that he’s noticed a very different work ethic in China.
Chinese entrepreneurs, he explained, are more ruthless than their American counterparts and they’re putting in a whole lot more hours.
Doug Leone of Sequoia Capital says founders in the US and China both want to change the world, but Chinese founders are a little more desperate (and you see it in the crazy work ethic they have).#TCDisrupt pic.twitter.com/dPxsRTbJoq
— TechCrunch (@TechCrunch) September 6, 2018
“I’ve had dinner in China until after 10 p.m. and people go to work after 10 p.m.,” Leone recalled.
“We don’t see that in the U.S. I’m not saying the U.S. founders oughta do that but those are the differences. They are similar in character. They are similar in dreams. They are similar in how they want to change the world. They are ultra-driven … The Chinese founders have a half other gear because I think they are a little more desperate.”
Much of this, however, has been said before and still, somehow, Silicon Valley remained the place to be for investors and startup entrepreneurs.
The reality is, those engaged in tech culture are always anxiously awaiting for the bubble to pop, the market to crash and for “peak Valley” to finally arrive.
Maybe, just maybe, Silicon Valley is forever.
Here’s more of our coverage of Disrupt 2018.
Powered by WPeMatico
Kano’s latest computer kit for kids doubles down on touch
Learn-to-code startup Kano, whose products aim to turn kids into digital makers, has taken the wraps off the latest incarnation of its build-it-yourself computer kit.
With the new flagship Kano is doubling down on touch interactions — urging kids to “make your own tablet”. The Computer Kit Touch packs a 10.1″ HD touchscreen, along with Kano’s now familiar bright orange wireless keyboard which comes with a built in trackpad.
While touch is becoming increasingly central to its products, Kano says the keyboard remains an important component of the product — supporting text-based coding apps which its platform also provides access to, as well as the more approachable drag-and-drop block-based coding systems that do really benefit from having a touchscreen to hand.
The kit, which Kano says is generally (but not exclusively) aimed at the 6-13 age range, is on sale from today, priced at $279.99 — via its website (Kano.me), as well as from selected retailers and e-tailers.
The Raspberry Pi powered computer is also getting increased storage capacity in this upgrade — of 16GB. But the main refresh is around updating Kano OS, Kano’s kid-friendly Pi topper, with expanded support for touch controls, according to founder Alex Klein .
Last year Kano combined touch and keyboard based interaction into a single product, the Computer Kit Complete — calling that a DIY laptop.
The 2018 refreshed version looks much the same, with enhancements generally behind the scenes and/or under the hood.
“The big moves this year are advancing the software and content ecosystem,” says Klein. “How it’s all integrated together.”
He points to another coding kit the team has up for pre-order, slated to ship next month — a co-branded Harry Potter gizmo in which kids get to build a motion-sensitive “coding wand” and use it to cook up their own digital spells, helped along by Kano’s software — adding: “With the Potter kit we’re bringing Kano code — to create a system, the ability to blend and change physics engines and sounds and particle systems — to tablets. So we’ve now got a touch-based interaction model for that e-product, as well as mouse and keyboard, and so we’ve brought that software system now to the Computer Kit Touch.
“You can code by dragging and dropping blocks with your fingers, you can paint and draw. You can change the pitch of a loop or a melody by running your fingers up and down and then using a change of a parameter mess with how quickly that melody changes, mess with the number of layers, you can make a beat or a loop using a touch-based digital audio workstation style X-Y plane. You can go into any one of our creative coding apps and pull in touch-based interactions, so instead of just using a mouse, a click and point, you can make an app that responds to swipes and taps, and different speeds, and in different locations.”
“On the touch kit itself there’s also a set of new content that demystifies how touchscreens work and peels back the layer of the screen and shows you what’s behind, and you’re kind of touching the intersection of the different copper wires and seeing what’s happening beneath,” he adds.
“There’s obviously a big hardware upgrade with the new ability to touch it, to take it with you. We’ve refined a lot of the components, we’ve improved the speed, the battery life. But really the core of it is this upgraded software that integrates with all the other kit.”
Talking of other kit, the learn-to-code space is now awash with quasi-educational gizmos, leaving parents in Western markets spoiled for choice of what to buy a budding coder.
Many more of these gizmos will be unboxed as we head into the holiday season. And while Kano was something of a startup pioneer here — a category creator, as Klein tells it — there’s now no shortage of tech for kids promising some kind of STEM-based educational benefit. So it’s facing an ever-growing gaggle of competition.
Kano’s strategy to stand out in an increasingly contested space is to fix on familiar elements, says Stein — flagging for example the popular game Minecraft — which runs on the Kano kit, and for which there’s a whole subsection of the Kano World community given over to hacking Minecraft.
And, well, aside from block-headed Minecraft characters it’s hard to find a character more familiar to children than the fictional wizard Harry Potter. So you can certainly see where Kano’s trying to get with the coding wand.
“We broke our first month pre-order target in one day,” he says of that forthcoming e-product (RRP ~$130). “There was massive coverage, massive traffic on our site, it was picked up all over the place and we’re very happy with the pre-orders so far. As are our retail partners.”
The Potter co-branding play is certainly Kano trying to make its products cast a wider spell by expanding the appeal of coding from nerdy makers to more mainstream child consumers. But how successful that will be remains to be seen. Not least because we’ve seen this sort of tactic elsewhere in this space.
Sphero, for example, is now rolling back the other way — shifting away from Star Wars co-branded bots to a serious education push focused on bringing STEM robotics to schools. (Although Kano would doubtless say a programmable bot that rolls is not the same as a fully fledged kit computer that can run all manner of apps, including familiar and fashionable stuff like Minecraft and YouTube.)
“We’re very pleased to see that this category that we created, with that Kickstarter campaign in 2013 — it’s become more than what some people initially feared it would be which was niche, maker ‘arcanery’; and it’s becoming a major consumer phenomenon,” he says. “This notion that people want to make their own technology, learn how to code and play in that way. And not just kids — people of all ages.”
On the hard sales front, Klein isn’t breaking out numbers for Potter kit pre-sales at this stage. But says the various incarnations of its main computer kit have shipped ~360,000 units since September 2014. So it’s not Lego (which has also moved into programmable kits) — but it’s not bad either.
In recent years Kano has also branched out into offering Internet of Things kits, previewing three code-your-own connected devices in 2016 — and launching Kickstarter campaigns to get the products to market.
It’s since shipped one (the Pixel kit) but the other two (a build-it-yourself camera kit and a DIY speaker) remain delayed — leaving crowdfunder backers waiting for their hardware.
Why the delay? Have Kano’s priorities shifted — perhaps because it’s focusing efforts on cobranded products (like the Potter wand) vs creating more of its own standalone devices?
“We are still committed to shipping the speaker kit, the camera kit,” Klein tells TechCrunch. “A big reason for [the delay] is not only the fact that the company is in a position now where we have mass distribution, we have great partners — perennially testing new product ideas — and we want to make sure that products are going to resonate with, not just a small group of people but many, many people, of many different age groups and interests before we release them.”
He also points out that any backers of the two devices who want refunds can get them in full.
Though he also says some are choosing to wait — adding that Kano remains committed to shipping the devices, and saying for those that do wait there will be a few extra bells and whistles than originally specced out in the crowdfunder campaign.
The delay itself looks like the market (and consumer tastes) moving quicker than Kano predicted — and so it finds itself wishing its products could deliver more than it originally planned (but without a wand to wave to instantly achieve that).
This is also a pitfall with previewing anything months or years ahead of time, of course. But the expense and complexity of building hardware makes crowdfunding platforms attractive — even for a relatively established brand like Kano.
“The delay is really unfortunate,” he adds. “We did say they would ship earlier but what we have done is we’ve offered any backer a full refund on the camera and the speaker if they don’t want to wait. But if they do wait they will receive incredible camera, incredible speaker. Both of them are going to benefit from the advancements made in low cost computing in the last year.
“The speaker as well is going to have elements that weren’t even part of the original campaign. On our side it’s critical that we get those products absolutely right and that they feel mass, and that they demystify not only coding and the Internet of Things, which was part of the original purpose, but in the case of the camera and the speaker there are elements that have come to the fore in more recent months like voice interaction and image recognition that we feel if our mandate is to demystify technology and we’re shipping a camera and a speaker… that’s kind of part of it. Make it perfect, make it of the moment. And for any backer who doesn’t want to wait for that, no problem at all — we’ll refund you 100%.”
Beyond reworking its approach with those perhaps overly ambitious connected devices, Kano has additional release plans in its pipeline — with Klein mentioning that additional co-branded products will be coming next year.
He says Kano is also eyeing expanding into more markets. “There’s a significant market for Kano even beyond our traditional leading position amongst 6-13 year olds in the US and the UK. There’s a really strong market for people who are beyond the US and the UK and we’re now at a scale where we can start really investing in these distribution and localization relationships that have come our way since year one,” he says.
And he at least entertains the idea of a future Kano device that does away with a keyboard entirely — and goes all in on touch — when we suggest it.
“Would we move to a place where we have no keyboard in a Kano computer? I think it’s very possible,” he says. “It might be a different form factor, it might be smaller, it might fit in your pocket, it might have connectivity — that kind of stuff.”
Which sort of sounds like Kano’s thinking about making a DIY smartphone. If so, you heard it here first.
The five and a half year old London-based startup is not yet profitable but Klein flags growth he dubs “fast enough” (noting it doubled sales year-over-year last year, a “trend” he says continued in the first half of this year), before adding: “It’s not impossible for us to get to profitability. We have a lot of optionality. But at the moment we are making investments — in software, in team — we have partner products coming out like Harry, we’ll have more coming out next year. So in terms of absolute positive EBITDA not yet but we are profitable on a units basis.”
Kano closed a $28M Series B last year — and has raised some $44.5M in all at this stage, according to Crunchbase. Is it raising more funding now? “I think any entrepreneur who is looking to do something big is always in some sense keeping an eye out for sources of capital,” replies Klein. “As well as sources of talent.”
He points by way of a connected aside to this study of C-suite execs, carried out by Stripe and Harris poll, which found that access to software developers is a bigger constraint than access to capital, saying: “I read that and I thought that that gap — between the 1% of 1% who can develop software or hardware and the rest of us — is exactly the challenge that Kano set out to solve from a consumer and education perspective.”
“In terms of fundraising we do get a lot of inbound, we have great investors at the moment,” he adds. “We do know that the scale of this particular challenge — which is demystify technology, become synonymous with learning to code and making your own computers — that requires significant support and we’ll be continuing to keep our eyes out as we grow.”
Powered by WPeMatico
Peep the future of distributed ledgers with the leaders of Hyperledger, Parity Technologies and Tradeshift
As cryptocurrencies emerge from the speculative bloodletting of the past months, believers in the promise of distributed ledger technologies for business and consumer applications are casting about for what comes next.
On our stage at Disrupt San Francisco we’ll be welcoming some of the leading thinkers in how distributed ledgers can create an entirely new architecture for computing and new processes for almost every conceivable transaction framework.
For Brian Behlendorf, the executive director of Hyperledger, distributed ledger technologies represent a powerful path for the future of networked computing — no matter the underlying technology. That’s why Behlendorf –through the Linux Foundation — is investing resources in ensuring that viable open source distributed ledger projects are supported and coming to market for any number of applications for businesses and consumers.
One of the leading lights of the internet revolution, Behlendorf’s career shaping the future of the networked world began in 1993 when he co-founded Organic Inc. — the first business dedicated to building commercial websites. Going on to become one of the foundational architects of the Apache http protocol, Behlendorf has served as the chief technology officer of the World Economic Forum and as an executive director for the technology investment fund, Mithril Capital.
Meanwhile, Parity Technologies is attempting to ensure that businesses don’t need to worry about the underlying technologies at all. Selling a suite of services that are all enabled by distributed ledger technologies and cryptographic computing, Jutta Steiner is giving businesses a way through the maze of competing protocols with a service that can enable the creation and adoption of distributed apps for businesses.
“We see it as a way for people to build blockchains that fulfill their particular needs,” Steiner told our own Samantha Stein at our Blockchain event earlier this year in Zug. “One of the challenges we’re addressing in this is to come up with a scalable framework.”
Before Parity, Steiner was responsible for security and partner integration within the Ethereum Foundation when the public blockchain first launched in 2015. Steiner also co-founded Project Provenance — a London based start-up that employs blockchain technology to make supply chains more transparent.
Supply chains are at the heart of Tradeshift’s offerings — and the company is hoping that distributed ledgers will be too. That’s why the company created Tradeshift Frontiers, an innovation lab and incubator that will focus on transforming supply chains through emerging technologies, such as distributed ledgers, artificial intelligence and the Internet of Things.
“The use cases we’re working through Frontiers cover a very wide variety of themes, including supply chain financing, asset liquidity, and supply chain transparency,” said Gert Sylvest, co-founder and GM of Tradeshift Frontiers, at the time. “There is so much more potential than just cryptocurrencies.”
That potential will be one of the things that Sylvest, Steiner, and Behlendorf discuss. We’ll hope you’ll be in the audience to listen.
Disrupt SF will take place in San Francisco’s Moscone Center West from September 5 to 7. The full agenda is here, and you can still buy tickets right here.
Powered by WPeMatico
What every startup founder should know about exits
The dream of a startup founder can often be summarized by the following well-intentioned, and mostly delusional, quote: “We’ll raise a few rounds and in a few years we’ll IPO on Nasdaq.”
But a more likely scenario looks something like this:
You invest a few years of hard work to build something of value. One day you receive an acquisition offer out of the blue. You’re elated. And you’re not prepared. You drop everything to focus on this opportunity. Exclusive due diligence starts. Your company is a mess (IP, contracts, burn). Days become weeks; weeks become months. You’ve neglected business and fundraising. You’re running out of money. M&A is now your one and only option. The buyer says they found a bunch of cockroaches in the walls and drops the price. Now what?
Sound unlikely?
This is still a favorable situation: You had an offer! Think about how much time you invested in your various funding rounds. The hundreds of names and Google spreadsheet or Streak-powered quasi-CRM process.
Have you spent even a fraction of that on understanding exit paths? If you’d rather not live the situation described above, read along.
The E-word: A strange taboo
Investors live by exits, but many founders keep dreaming of unicornization and avoid the “E-word” until it’s too late. Yet, in 2016, 97 percent of exits were M&As. And most happened before Series B.
Exits matter because that’s when you, your team and your investors get paid. Oddly enough, and to use a chess metaphor, we hear a lot about the “opening game” (lean startup) and the “mid-game” (growth), but very little about this “end game.”
As a result, founders miss opportunities or leave money on the table. This is a shame. Our fund has more than 700 companies in portfolio. We want the best possible exit for each of them. And fortune favors the prepared! Now, how to get 700 exits (and counting)?
To explore the topic, we organized a series of Master Classes tapping corporate buyers, bankers, investors, lawyers and startup CEOs with M&A or IPO experience in San Francisco. It was a group that included the founders of Guitar Hero — bought by Activision; JUMP Bikes — a SOSV portfolio company bought by Uber, Ubiquisys — bought by Cisco and Withings — bought by Nokia. Each one for hundreds of millions.
Their observations can be summarized below.
Maximize optionality
“Founders must be aware of what contributes to an exit. This means understanding partnerships and how they are formed in the business space the entrepreneur is working in,” said one Master Class participant.
As founders, you build your product, your company and… optionality. You need to understand the options open to your company, and take steps to enable them.
The most likely one is an acquisition, but there are others like IPO (including small cap), RTO, SBO, LBO, Equity Crowdfunding and even ICO.
“Exit is not a goal per se, but as a CEO it is something you should think about as early in your cycle as possible, while being business-focused,” said the London-based investor Frederic Rombaut, of Seraphim Capital.
Indeed, most participants said that exits should always be on the chief executive’s agenda, no matter how early in the process. “Exits should be on the CEO agenda. Not front and center, but on the agenda. M&A is a by-product of a great business and targeted BD. IPOs are always an option once you’ve built significant cashflow forecasting.”
It’s important to ask questions like: How many “strategic engagements” with potential buyers have you had this month? Is your message and value clear in their eyes? Have you considered an acquisition track in parallel to a fundraise?
It doesn’t stop there:
- Equity crowdfunding might help close some gaps at seed stage.
- Early IPOs on smaller exchanges can be an option to raise over $10 million — the robotics startup Balyo went public and raised €40 million on Euronext to get rid of a critical “right of first refusal” option held by one of its corporate investors.
- Reverse mergers can work too: the medical exoskeleton company EKSO Bionics went public this way.
One thing is sure: The time to exit is not when you’re running out of money.
Companies are bought, not sold
Unicorn or not, the most likely exit is an acquisition.
As George Patterson, managing director at HSBC in New York said, “Good tech companies are bought, not sold. The question is thus: how to get bought?”
Patterson says it’s important to understand how mergers and acquisitions actually work; how to prepare a startup for an exit; and how to develop a “feel” for the market you’re exiting through and into.
How M&A works
Hearing from corp dev veterans from Cisco, Logitech, Dassault and IBM, a few key ideas emerged:
Motivations vary
It could be from least to most expensive, or as a mix, as listed by Mark Suster, managing partner at Upfront Ventures:
- Talent hire ($1 million/dev as a rule of thumb — location matters)
- Product gap
- Revenue driver
- Strategic threat (avoid or delay disruption)
- Defensive move (can’t afford a competitor to own it)
How corporates find you
Corporates find deals via the development of partnerships, investment (CVC), their business units, corp dev research, media and investor connections.
Asked about the best approach, Todd Neville, manager of Corporate Business Development and Strategy at IBM (who gave the most detailed description of the corp dev process), said, “Do something cool to one of the IBM customers. If they rave about even a POC, we’re interested.”
In other words, business development is corporate development.
Get the house in order
Buyers typically want to know three things:
- Is your IP really yours?
- Is your team capable?
- Will your customers stick around?
For IP, they will check your contracts (staff and contractors), and run some automated code analysis for proprietary code and open source use. They will evaluate potential IP infringement. No point buying you if you end up costing more in lawsuits!
For your team skills: Sitting down with your engineers will tell them plenty enough without understanding the details of this or that algorithm. The last thing a corporate wants is to be accused of stealing!
Lawyers engaged early can help. The later the clean-up, the more costly and painful.
Develop a feel for your “market”
Develop relationships and create champions within corporates. It will help promote your deal when the time comes, and will let you keep your finger on the pulse of corporate strategy to time your moves.
Do you read the earning calls of Cisco or IBM (or others relevant to you)? This is where strategies are presented. Are your keywords coming up there or in their press releases?
Chris Gilbert, former CEO of Ubiquisys (sold to Cisco for more than $300 million) was very deliberate in planning his exit.
“Selling starts on day one and is a leadership-only function — work out who will be your buyer. Only the CEO can do this. Constantly articulate why a company should buy you,” Gilbert said. Bring clear messages into the acquiring company so it can be presented upwards: give them the presentation you would like them to show their boss! When the time is right, force decisions through competition. If you know they have to buy you, your starting position is strong.”
The dark art of price discovery
There are dozens of formulas (from DCF to comparables) to evaluate a deal — which also means none is “correct.” What matters is: How much would you sell for, and how much is the buyer ready to pay?
Gilbert, at Ubiquisys, described how close interactions with his banker helped drive the price up among the bidders assembled.
Just like buyers, we meet bankers and lawyers too rarely at startup events, but there is much to learn with them. They make deals happen, avoid value erosion and optimize price. They often also make introductions before you engage them, to build goodwill and earn your business.
And if you worry about fees, the right banker handsomely pays for itself by finding more bidders and playing “bad cop” for you, avoiding direct confrontation with your future employer. Do you want a slice of the watermelon or the whole grape?
Final twist: Exits are not exits
When asked about what happens after an M&A or IPO, buyers said they generally hoped the founders would stay with them for many years. Often using re-vesting, earn-outs or shares of the acquiring company to incentivize them. Neville, from IBM, mentioned a security company they acquired whose founder is now the head of one of the largest IBM divisions.
In the case of IPOs, supposedly the ultimate “exit,” any block of shares sold by founders would face extreme scrutiny and might cause a price drop.
So who’s exiting during those deals? Investors (and not always).
Eventually, if the average age of a startup at exit is 8-10 years, the active duty period of founders (if not replaced in the meantime) extends even more. Better love the problem you’re solving, and your customers!
Thanks to speakers, participants and supporters of this Master Class series:
London: Frederic Rombaut (Seraphim Capital), Joe Tabberer (FirstBank), Chris Gilbert (Ubiquisys), Jonathan Keeling (Crowdcube), Fred Destin, Tony Fish (AMF Ventures, James Clark (London Stock Exchange), Denise Law (SGCIB).
Paris: Frederic Rombaut (Seraphim Capital), Manuel Gruson (Dassault Systemes), Pierre-Henri Chappaz (Rothschild Global Advisory), Christine Lambert-Goue (All Invest), Olivier Younes (EXPEN), Eric Carreel (Withings), Fabien Bardinet (Balyo), Xavier Lazarus (Elaia Partners), Pierre-Eric Leibovici(Daphni). Jean de La Rochebrochard (Kima Ventures), Jeremy Sartre (SmartAngels), Gwen Regina Tan (Entrepreneur First).
San Francisco: Natasha Ligai (Logitech), Matt Cutler (Cisco),Will Hawthorne, (CODE Advisors), Ryan Rzepecki (JUMP Bikes), Charles Huang (Guitar Hero), Jeff Thomas (Nasdaq), Shahin Farshchi (Lux Capital), Ammar Hanafi (Moment Ventures), Adam J. Epstein (Third Creek Advisors), Nathan Harding (EKSO Bionics), Kate Whitcomb, Anthony Marino and Ethan Haigh (SOSV).
New York: Todd Neville (IBM), George Patterson (HSBC), Ryan Rzepecki (JUMP Bikes), Aaron Kellner (SeedInvest), Jeremy Levine (Bessemer Venture Partners), Taylor Greene (Collaborative Fund), Adam Rothenberg (BoxGroup), Eli Curi (Fenwick & West), Ian Engstrand and Salil Gandhi (Goodwin), Warren Spar(Sparring Partners Capital), Duncan Turner, Vivian Law and Sheng Ge (SOSV).
Powered by WPeMatico
Knotel acquires 42Floors in order to build the blockchain of property
Another day, another blockchain project. This time sources are reporting that Knotel – an office space rental service in Manhattan – has acquired 42Floors, a commercial real estate search engine in order to, according to founder Amol Sarva, get “access to data and technology on over 10 billion square feet of office space, driving further liquidity to Knotel’s marketplace while also accelerating its plans for a blockchain platform.”
The deal is not yet complete.
Knotel is building the Agile HQ platform, a way to rent office space for a few hours or a few months without getting stuck in a lease. The company has 1 million square feet of space in New York, San Francisco, London, and Berlin and it raised $100 million in funding. The company claims it has more has more buildings in New York than WeWork.
“42Floors built a powerful tool to organize a dark market that hasn’t changed in a hundred years,” said Amol Sarva, CEO of Knotel. “It’s still backroom and bilateral while the rest of the world is becoming digital and standardized. This is what leads to transactions that take months to close with a dozen middlemen – no reliable information. You can buy a house faster than you can rent a floor. Partnering together will help give owners and customers what they both want: truth.”
The reported 42Floors acquisition enables the company to bring new properties onto its platform and could let non-blockchain-based contracts move to the blockchain.
UPDATE – Text changed to reflect the type of business and ICO plans.
Powered by WPeMatico
Are scooter startups really worth billions?
It’s been hard to miss the scooter startup wars opening fresh, techno-fueled rifts in Valley society in recent months. Another flavor of ride-sharing steed which sprouted seemingly overnight to clutter up sidewalks — drawing rapid-fire ire from city regulators apparently far more forgiving of traffic congestion if it’s delivered in the traditional, car-shaped capsule.
Even in their best, most-groomed PR shots, the dockless carelessness of these slimline electrified scooters hums with an air of insouciance and privilege. As if to say: Why yes, we turned a kids’ toy into a battery-powered kidult transporter — what u gonna do about it?
An earlier batch of electric scooter sharing startups — offering full-fat, on-road mopeds that most definitely do need a license to ride (and, unless you’re crazy, a helmet for your head) — just can’t compete with that. Last mile does not haul.
But a short-walk replacement tool that’s so seamlessly manhandled is also of course easily vandalized. Or misappropriated. Or both. And there have been a plethora of scooter dismemberment/kidnap horror stories coming out of California, judging by reports from the scooter wars front line. Hanging scooters in trees is presumably a protest thing.
Scooter brand Lime struck an especially tone-deaf tech note trying to fix this problem after an update added a security alarm that bellowed robotic threats to call the cops on anyone who fumbled to unlock them. Safe to say, littering abusive scooters in public spaces isn’t a way to win friends and influence people.
Even when functioning ‘correctly’, i.e. as intended, scooter rides can ooze a kind of brash entitlement. The sweatless convenience looks like it might be mostly enabling another advance in tech-fueled douche behavior as a t-shirt wearing alpha nerd zips past barking into AirPods and inhaling a takeaway latte while cutting up the patience of pedestrians.
None of this fast-seeded societal friction has put the brakes on e-scooter startup momentum, though. Au contraire. They’ve been raising massive amounts of investment on rapidly inflating valuations ($2BN is the latest valuation for Bird).
But buying lots of e-scooters and leaving them at the mercy of human whim is an expensive business to try scaling. Hence big funding rounds are necessary if you’re going to replace all the canal-dunked duds and keep scooting fast enough for the competition.
At the same time, there isn’t a great deal to differentiate one e-scooter experience over another — beyond price and proximity. Branding might do it but then you have to scramble even harder and faster to create a slick experience and inflate a brand that sticks. (And it goes without saying that a scooter sticky with fecal-matter is absolutely not that.)
The still fledgling startups are certainly scrambling to scale, with some also already pushing into international markets. Lime just scattered ~200 e-scooters in Paris, for example. It’s also been testing the waters more quietly in Zurich. While Bird has its beady eye on European territory too.
The idea underpinning some very obese valuations for these fledgling startups is that scooters will be a key piece of a reworked, multi-modal transport mix for urban mobility, fueled by app-based convenience and city buy-in to greener transport options with emissions-free benefits. (Albeit scooters’ greenness depends on what they’re displacing; Great if it’s gas-guzzling cars, less compelling if it’s people walking or peddling.)
And while investors are buying in to the vision that lots of city dwellers are going to be scooting the last mile in future, and betting big on sizable value being captured by a few plucky scooter startups — more than half a billion dollars has been funneled into just two of these slimline scooter brands, Bird and Lime, since February — there are skeptical notes being sounded too.
Asking whether the scooter model really justifies such huge raises and heady valuations. Wondering if it isn’t a bit crazy for a fledgling Bird to be 2x a unicorn already.
Shared bike and scooter fleets are paving the way to a revolution in urban mobility but will only capture little value in the long term. Investors are highly overestimating the virtue of these businesses.
— Thibaud Elziere (@tiboel) June 18, 2018
The bear case for these slimline e-scooters says they’re really only fixing a pretty limited urban mobility problem. Too spindly and unsafe to go the distance, too sedate of pace (and challenged for sidewalk space) to feel worthwhile if you don’t have far to go anyway. And of course you’re not going to be able to cart your kids and/or much baggage on a stand-up two wheeler. So they’re useless for families.
Meanwhile scooter invasions are illegal in some places and, where they are possible, are fast inviting public and regulatory frisson and friction — by contributing to congestion and peril on already crowded pavements.
After taking one of Lime’s just-landed e-scooters for a spin in Paris this week, Willy Braun, VC at early stage European fund Daphni, came away unimpressed. “I didn’t feel I was really saving time in a short distance, since there is always many people in our narrow sidewalks,” he tells us. “And it isn’t comfortable enough for me to imagine a longer distance. Also it’s quite expensive ($1 per use and $.15/min).
“Lastly: Before renting it I read two news media that told me I had to use it only on the sidewalks and they tell us that we should only use it on the road during the onboarding — and that wearing an helmet is mandatory without providing it). As a comparison, I’d rather use e-bikes (or emoto-bikes) for longer journey without hesitation.”
“Give us Jump instead of Lime!” he adds, namechecking the electric bike startup that’s been lodged under Uber’s umbrella since April, adding a greener string to its urban mobility bow — and which is also heading over to Europe as part of the ride-hailing giant’s ongoing efforts to revitalize its regionally battered brand.
“Uber stands ready to help address some of the biggest challenges facing German cities: tackling air pollution, reducing congestion and increasing access to cleaner transportation solutions,” said CEO Dara Khosrowshahi wheeling a bright red Jump bike on stage at the Noah conference in Berlin earlier this month. Uber’s Jump e-bikes will launch in Germany this summer.
E-bikes do seem to offer more urban mobility versatility than e-scooters. Though a scooter is arguably a more accessible type of wheeled steed vs a bike, given you can just stand on it and be moved.
But in Europe’s dense and dynamic urban environments — which, unlike the US, tend to be replete with public transit options (typically at a spectrum of price-points) — individual transport choices tend to be based firstly on economics. After which it’s essentially a matter of personal taste and/or the weather.
Urban transport horses for courses — depending on your risk, convenience and comfort thresholds, thanks to a publicly funded luxury of choice. So scooters have loads of already embedded competition.
TechCrunch’s resident Parisienne, Romain Dillet — a regular user of on-demand bike services in the city (of which there are many), and prior to that the city’s own dock-based bike rental scheme — also went for a test spin on a Lime scooter this week. And also came away feeling underwhelmed.
“This is bad,” he said after his ride. “It’s slow and you need to brake constantly. BUT the worst part is that it feels waaaaaay more dangerous than a bike. Basically you can’t brake abruptly because you’re just standing there.”
Index Venture’s Martin Mignot was also in Paris this week and he took the chance to take a Lime scooter for a spin too — checking out the competition in his case, given the European VC firm is a Bird backer. So what did he think?
“The experience is pretty cool. It’s slightly faster than a bike, there’s no sweating. The weather was just amazing and very hot in Paris so it was pretty amazing in terms of speed and lack of effort,” he says, rolling out the positively spun, vested view on scooter sharing. “Especially going up hill to go to Gare du Nord.
“And the lack of friction — just to get on board and get started. So in general I think it’s a great experience and I think it feels a really interesting niche between walking and on-demand bikes… In Paris you’ve also got the mopeds. So that kind of ‘in between offering’. I think there’s a big market there. I think it’s going to work pretty well in Paris.”
Mignot is a tad disparaging about the quality of Lime’s scooters vs the model being deployed by Bird — a scooter model he also personally owns. But again, as you’d expect given his vested interests.
“Obviously I’m biased but I would say that the Xiaomi scooter/Ninebot scooter is higher quality than the one that Lime are using,” he tells us. “I thought that the Lime one, the handlebar is a little bit too high. The braking is a little bit too soft. Maybe it was the one I used, I don’t know.”
Talking generally about scooter startups, he says investors’ excitement boils down to trip frequency — thanks exactly to journeys being these itty-bitty last mile links.
But it’s also then about the potential for all that last mile hopping to be a shortcut for winning a prized slot on smartphone users’ homescreens — and thus the underlying game being played looks like a jockeying for prime position in the urban mobility race.
Lime, for example, started out with bike rentals before jumping into scooters and going multi-modal. So scooter sharing starts to look like a strategy for mobility startups to scoot to the top of the attention foodchain — where they’re then positioned to offer a full mix and capture more value.
So really scooters might mostly be a tool for catching people’s app attention. Think of that next time you see one lying on a sidewalk.
“What’s very interesting if you look at the trip distribution, most of the trips are short. So the vast majority of trips if you’re walking, obviously, are less than three miles. So that’s actually where the bulk of the mobility happens. And scooters play really well in that field. So in terms of sheer number of trips I think it’s going to dwarf any other type of transportation. And especially ride-hailing,” says Mignot.
“If you look at how often do people use Uber or Lyft or Taxify… it’s going to be much less frequent than the scooter users. And I think that’s what makes it such an interesting asset… The frequency will be much higher — and so the apps that power the scooters will tend to be on the homescreen. And kind of on top of the foodchain, so to speak. So I think that’s what makes it super interesting.”
Scooters also get a big investor tick on merit of the lack of friction standing in the way of riding vs other available urban options such as bikes (or, well, non-electric scooters, skateboards, roller blades, public transport, and so on and on) — in both onboarding (getting going) and propulsion (i.e. the lack of sweat required to ride) terms.
“That’s what’s so brilliant with these devices, you just snap the QR code and off you go,” he says. “The difference with bikes is that you don’t have to produce any effort. I think there are cases where obviously bikes are better. But I think there are a lot of cases where people will want something where you don’t sweat.
“Where you don’t wrinkle your clothes. Which goes a little bit faster. Without going all the way to the moped experience where you need to put the helmet, which is a bit more dangerous, which a lot of people, especially women, are not super familiar with. So I think what’s exciting with scooters as a form factor is it’s actually very mainstream.
“Anyone can ride them. It’s very simple to manoeuvre. It’s not super fast, it’s not too dangerous. It doesn’t require any muscular effort — so for older people or for people who just don’t want to sweat because they’re going to a meeting or something. It’s just a fantastic option.”
Index has also invested in an e-bike startup (Cowboy) and the firm is fully signed up to the notion that urban mobility will be multimodal. So if e-scooters valuations are a bit overcooked Index is not going to be too concerned. People in cities are clearly going to be riding something. And backing a mix is a smart way to hedge the risk of any one option ending up more passing fad than staple urban steed.
Mostly Index is betting that people will keep on riding robotic horses for urban courses. And whatever they ride it’s a fairly safe bet that an app is going to be involved in the process of finding (docklessness is therefore another attention play) or unlocking (scan that QR code!) the mobility device — opening up the possibility that a single app could house multiple mobility options and thus capture more overall value.
“It’s not a one-size fits all. They’re all complementing each other,” says Mignot of the urban mobility options in play. “I would say e-bikes are probably a little bit more great for little bit longer trips because you’re sitting down. But again it takes a little bit longer, because you have to adjust the saddle, you need to start peddling. There’s a bit more friction both on the onboading and on the riding. But they’re a bit better for slightly longer distances. I would say for shorter distances there’s nothing better than the scooter.”
He also points out that scooters are both cheaper and less bulky than e-bikes. And because they take up less street space they can — at least in theory — be more densely stacked, thereby generating the claimed convenience by having them sitting near enough to convince someone not to bother walking 10 minutes to the café or gym — and just scoot instead. So scooters’ slimline physique is also especially exciting to investors. (Even if, ironically, it’s being deployed to urge people to walk less.)
“I think we will end up with more density of scooters. Which is super important,” he continues. “People will, in the end, tend to take the vehicle that they can find where they are. And I think it’s more likely, eventually, that they will get a scooter than an e-bike. Just simply because they take less space and they are less expensive.”
But why wouldn’t people who do get won over to the sweatless perks of last mile scooting just buy and own their own ride — rather than shelling out on an ongoing basis to share?
Unlike bikes, scooters are mobile enough to be picked up and moved around fairly easily. Which means they can go with you into your home, office, even a restaurant — disruptively reducing theft risk. Whereas talk to any bike owner and they’ll almost invariably have at least one tale of theft woe, which is a key part of what makes bike sharing so attractive: It erases theft worry.
Add to that, you can find e-scooters on sale in European electronics shops for as little as €140. So if you’re going to be a regular scooterer, the purely economic argument to just own your own looks pretty compelling.
And people zipping around on e-scooters is a pretty common sight in another dense European city, Barcelona, which has very scooter-friendly weather but no scooter startups (yet). But unless it’s a tourist weaving along the seafront most of these riders are not shared: People just popped into their local electronics shop and walked out with a scooter in a box.
So the rides aren’t generating repeat revenue for anyone except the electricity companies.
Asked why people who do want to scoot won’t just buy, rather than rent Mignot talks up the hassle of ownership — undermined slightly by the fact he is also a scooter owner (despite the claimed faff from problems such as frequent flat tires and the chore of the nightly charge).
“The thing you notice very rapidly: There are two things, one is the maintenance,” he says. “The models that exist today are not super robust. Maybe in a very flat, very smooth roads, maybe Santa Monica, maybe it’s a little bit less true but I would say in Europe the maintenance that is required is fairly high… I have to do something on mine every week.
“The other thing is it takes a little bit of space. If you have to bring it to a restaurant or whatever type of crowded place, a movie theatre or wherever you’re going, to an office, to a meeting room, it’s a little bit on the heavy side, and it’s a little bit inconvenient. So certainly some people will buy them… But I also think that there are a lot of cases where you’d rather have it just on-demand.”
Unlike Mignot and Index, Tom Bradley, of UK focused VC firm Oxford Capital, is not so convinced by the on-demand scooter craze.
The firm has not made any e-scooter investments itself, though mobility is a “core theme”, with the portfolio including an on-demand coach travel startup (Sn-ap), and technology plays such as Morpheus Labs (machine learning for driverless cars) and UltraSoc (complex circuits for automotive parts, which sells to the likes of Tesla).
But it’s just not been sold on scooter startups. Bradley describes it as an “open question” whether scooters end up being “an important part of how people move around the cities of the future”. He also points to theft problems with dockless bike share schemes that have not played out well in the UK.
“We’re not convinced that this is a fundamental part of the picture,” he says of scooter sharing. “It may be a part of the picture but I personally am not yet convinced that it’s as big a part of the picture that people seem to be prepared to pay for.”
“I keep thinking of the Segway example,” he adds. “It’s an absolutely delightful product. It’s brilliant. It’s absolutely brilliant. In a way that these electric scooters are not. But obviously it was much more expensive. And it made people feel a bit weird. But it was supposed to be the answer — and it’s not the answer. Before its time, perhaps.”
Of course he also accepts that capital is “being used as a weapon”, as he puts it, to scoot full-pelt towards a future where shared electric scooters are the norm on city streets by waging a “marketing war” to get there.
“Venture capital valuations are what someone is prepared to pay. And in this case people are valuing potential rather than valuing the business… so the valuations [of Bird and Lime] are being driven more than anything by the amount of money being raised,” he says. “So you decide a rule of thumb about what is acceptable dilution, and if you’re going to raise $400M or whatever then the valuation’s got to be somewhere between $1.6BN and $2BN to make that sort of raise make sense — and leave enough equity for the previous investors and founders. So there’s an element of this where the valuations are being driven by the amount of capital being raised.”
Oxford Capital’s bearish view on scooter sharing is also bounded by the fund only investing in UK-based startups. And while Bradley says it sees lots of local mobility strengths — especially in the automotive market — he admits it’s more of a mental leap to imagine a world leading scooter startup sprouting from the country’s green and pleasant lands. Not least because it’s not legal to use them on UK public roads or pavements.
“If you look at places like Amsterdam, Berlin, they’re sort of built for bikes. London’s getting towards being built for bikes… Cycling’s been one of the big success stories in London. Is [scooter sharing] going to replace cycling? I don’t know. Not so convinced… It’s obviously easy for anyone to get on and off these things, young and old. So that’s good, it’s inclusive. But it feels a little bit like a solution looking for a problem, the sorts of journeys people talk about for these things — on campus, short urban journeys. A lot of these are walkable or cycle journeys in a lot of cities. So is there a mass need?
“Is this Segway 2 or is this bike hire 2… it’s hard to tell. And we’re coming down on the former. We’re not convinced this is going to be a fundamental part of the transport space. It will be a feature but not a huge part.”
But for Mignot the early days of the urban mobility attention wars mean there’s much to play for — and much that can be favorably reshaped to fit scooters into the mix.
“The whole thing, even on-demand bikes, it’s a two year old phenomenon really,” he says. “So I think everyone is just trying to learn and figure out and adapt to this new reality, whether it’s users or companies or cities. I think it’s very similar to when cars were first introduced. There were no parking spaces at the time and there were no rules on the road. And fast forward 100 years and it looks very different.
“If you look at the amount of infrastructure and effort and spend that has been put into making — and I would argue way more than should have — into making a city car-friendly, if you only do a 100th of the same amount of effort and spend into making some space for bicycles and light two-wheel vehicles I think we’ll be fine.
“That’s the beauty of this model. If you compare the space of the tech and if you look at the efficiency of moving people around vs the space, the scooters are simply the most efficient because their footprint on the ground is just so small.”
He even makes the case for scooters working well in London — arguing the sprawl of the city amps up the utility because there are so many tedious last mile trips that people have to make.
Even more so than in denser European cities like Paris, where he admits that hopping on a scooter might just be more of a “nice to have”, given shorter distances and all the other available options. So, really, where urban mobility is concerned, it can actually be courses for horses.
Yet, the reality is London is off-limits to the likes of Bird and Lime for now — thanks to UK laws barring this type of unlicensed personal electric vehicle from public roads and spaces.
You can buy e-scooters for use on private land in the UK but any scooter startups that tried their usual playbook in London would be scooting straight for legal hot water.
It’s not just the British weather that’s inclement.
“I’m really hoping that TfL [Transport for London] and the Department for Transport are going to make it possible,” says Mignot on that. “I think any city should welcome this with open arms. Some cities are, by the way. And I think over time once they see the success stories in other parts of the world I think they all will. But I wish London was one of those cutting edge cities that would welcome new innovation with open arms. I think right now, unfortunately, it’s not there.
“There’s a lot of talk about air quality, and so on, but actually, when push comes to shove… you have a lot of resistance and a lot of pushback… So it’s a little bit disappointing. But, you know, we’ll get there eventually.”
Powered by WPeMatico