Category: UNCATEGORIZED

Meet Dejbox, a French food delivery startup that tries to avoid busy cities in order to accommodate people who really need a new lunch option. The company is a full-stack food delivery startup that designs its own meals, cook them, sell them and deliver them.

“Corporate headquarters are more and more often far from city centers. But what about lunch options for those areas?” co-founder and co-CEO Vincent Dupied told me.

Answering this question creates logistical challenges more than anything else. It’s hard to cover wide areas that are spread out all around busy cities, such as Paris, Lille and Lyon. And Dejbox has made some radical decisions that set them apart from well-known companies, such as Deliveroo, Uber Eats or even Frichti.

Each delivery person drives a truck with 100 to 150 meals. This way, they can deliver multiple offices during one run. It means that customers can’t just order something and get it 30 minutes later.

They need to complete their order before 10:30am or 11am to get it for lunch time. And of course, you can also order multiple days in advance in case you don’t want to think about lunch for the rest of the week.

When it comes to sales, Dejbox tries to spot the most promising companies to pitch them the service. After that, multiple employees usually order from Dejbox every day. It means that delivery persons carry multiple meals and leave them in the kitchen or at the reception desk.

“Our delivery persons are a bit like mail carriers, they have the same itinerary every day and their own clients,” Dupied said.

That’s why Dejbox wants to empower its delivery staff as much as possible. They’re all full-time employees and they get monthly reports telling them how much revenue they’ve generated for the company.

This combination of low customer acquisition cost, low unit economics and high lifetime value has been working well. Partech first spotted them at the end of 2015 and invested a tiny $560,000 seed round (€500,000). Dejbox was only delivering 80 meals per day in the Lille area back then.

The startup quickly expanded to Paris and Lyon with the exact same focus on corporate headquarters in boring areas. In March 2017, Dejbox raised a $2.3 million Series A round (€2 million) from Partech and Leap Ventures. The company launched in Bordeaux a few months later.

And the company is quite transparent when it comes to metrics. During the first ~18 months, the startup generated $1.4 million in revenue, $4.5 million in 2017 and $11.3 million in 2018 (€1.2 million, €4 million and €10 million respectively).

This year, the company plans to generate $22.5 million in revenue (€20 million) and open two new cities — Nantes and Grenoble. Dejbox now delivers 10,000 people every day.

Elon Musk’s brain computer interface (BCI) venture Neuralink will provide some more insight into what they’ve been working on for the past two years, during which time we’ve heard very little in the way of updates on their progress. In 2017, we learned that Neuralink’s overall driving mission was to help humans keep pace with rapid advancements in AI, ensuring that we can continue to work with ever-more advanced technology, by closing the input and output gap between ourselves and computers.

Musk has famously forewarned of the potential dangers of artificial intelligence, and what happens when it becomes more powerful relative to out own ability to control and understand it. He also founded OpenAI alongside Sam Altman and others as a research organization hoping to collaborate on the development of AI specifically designed to benefit, not harm humanity.

At Recode’s Code conference in 2016, and again in 2017 at an event in Dubai, Musk discussed how BCI could help people communicate with computers with much higher bandwidth and lower latency than is possible now, using our relatively primitive input methods (keyboard, mice and touch all introduce a surprising amount of lag and fidelity loss if you think about it).

Neuralink seems to be Musk’s long-term play to help bridge this gap, which basically amounts to an effort to accelerate the kind of ‘singularity’ moment advocated by Ray Kurzweil and others wherein humanity effectively merges with technology as a new stage of engineered evolution. In the interim, however, it originally set its sights on addressing medical conditions including epilepsy and clinical, chemical depression via implanted brain interfaces, as a stepping stone to its more ambitious and more broadly applicable goals for BCI.

This update from Neuralink should shed light on where they are with that approach (if that’s even still the approach), and who knows: we might even get a clue as to how and when we can expect to truly become one with our computers.

In a long awaited decision, the Federal Elections Commission will now allow political campaigns to appoint cybersecurity helpers to protect political campaigns from cyberthreats and malicious attackers.

The FEC, which regulates political campaigns and contributions, was initially poised to block the effort under existing rules that disallow campaigns to receive discounted services for federal candidates because it’s treated as an “in kind donation.”

For now the ruling allows just one firm, Area 1 Security, which brought the case to the FEC, to assist federal campaigns to fight disinformation campaigns and hacking efforts, both of which were prevalent during the 2016 presidential election.

Campaigns had fought in favor of the proposal, fearing a re-run of 2016 in the upcoming presidential and lawmaker elections in 2020.

FBI director Christopher Wray said last in April that the recent disinformation efforts were “a dress rehearsal for the big show in 2020.”

In an opinion published Thursday, the FEC said the rules would be relaxed because Area 1 “would offer these services in the ordinary course of business and on the same terms and conditions as offered to similarly situated non-political clients.” In other words, political campaigns are not given a special deal but are offered the same price as others on its lowest tier of service.

Several other companies, like Facebook and Google-owned Jigsaw, have already offered free services to campaigns to fight disinformation and foreign hacking efforts.

However many political campaigns still are not taking basic security precautions, researchers found.

A spokesperson for Area 1 did not return a request for comment.

Bloomberg reported last April that Amazon was working on a home robot codenamed ‘Vesta’ (after the Roman goddess of the hearth and home) last year, and now the publication says that development on the project continues. Plus, the report includes new details about the specifics of the robot, including that it will indeed support Alexa and have wheels to help it move around. My terrible artist’s rendering of what that could look like is above.

The plan for Vesta was apparently to release it this year, but it’s not yet quite ready for mass production, according to Bloomberg’s sources. And while it could end up mothballed and never see the light of day, as with any project being developed ahead of launch, the company is said to be putting more engineering and development resources into the team working on its release.

Current prototypes of the robot are said to be about waist-high, per the report, and make their way through the world aided by sensor-fed computer-vision. It’ll come when you call thanks to the Alexa integration, per an internal demo described by Bloomberg, and should ostensibly offer all the same kind of functionality you’d get with an Echo device, including calling, timers and music playback.

For other clues as to what Vesta could look like, if and when it ever launches, a good model might be Kuri, the robot developed by Bosch internal startup Mayfield Robotics which was shuttered a year ago and never made it to market. Kuri could also record video and take photos, play games and generally interact with the household.

Meanwhile, Amazon is also apparently readying a Sonos-competing high-quality Echo speaker to debut next year.

Hello and welcome back to Equity, TechCrunch’s venture capital-focused podcast, where we unpack the numbers behind the headlines.

This week was a bit of a turn-about. Kate was off this week and Alex was back, so we brought back a few favorites to tide us over until our regular leader returns. For this IPO-themed episode, we had TechCrunch and Extra Crunch’s Danny Crichton in the studio along with Deloitte’s Barrett Daniels, a prior guest on the show and one of our favorite humans.

It being roughly the middle of the year, we decided to do a bit of run through the first two-quarters’ worth of tech IPOs. There was, as you can expect, a lot to get chat about.

We started with notes on how the Chinese venture capital market is changing, most notably in terms of its share of the world’s largest venture rounds. After leading the world for what felt like years, venture investment into China-based companies is slipping. And the declines are picking up attention (here, here).

But not all the news was gloomy on the show this week. Indeed, while some global data relating to the global IPO market wasn’t exactly sparkling, the US-listed tech IPO market is doing really well.

After we went over a number of the companies that went out and did well post-IPO (nearly every company aside from the ride-hailing players), we conceded that things are pretty damn warm for companies going public. At the same time, we couldn’t agree on how long the IPO market would remain so welcoming.

If it stays open, more unicorns will make it out. If the IPO window closes soon, we’ll see hundreds of unicorns trapped on the wrong side of the glass.

And we wrapped with notes on everyone’s favorite space-faring SPAC. We’ll see you all really soon.

Amazon is reportedly looking to offer an Echo that more directly competes with high-end speakers like the Sonos line of device of Apple’s HomePod, according to a new report from Bloomberg. The speaker should be released sometime next year, according to the sources cited in the report, and will be somewhat wider than the existing Echo models (perhaps more akin to the Echo Sub, pictured above), packing in four separate tweeters to help boost the song quality.

It will, of course, also offer access to the company’s Alexa voice assistant, which is what has propelled Echo to its current level of success. Bloomberg notes that it’s also likely to work better for the high-fidelity audio version of Amazon’s music streaming service that has previously been reported to be in the works.

This could make for an interesting working relationship with some of Amazon’s existing partners, including Sonos, since it sounds like this will be a direct competitor. Newer Sonos speakers, including the Sonos One and Sonos Beam, support Alexa voice commands out of the box. While both Echo devices and Sonos support multi-room streaming and speaker grouping, Sonos has always had far superior audio quality when compared to the Echo hardware – albeit at a premium price.

Sonos, meanwhile, is gearing up to launch speakers powered by its technology with Ikea, with the Symfonisk line that is set for release in August. Smart speakers are a busy space with a lot of money and interest from many companies big and small, but Amazon has a lot working in its favor if it can also produce something that wins on high-quality audio at a reasonable price.

SpaceX is preparing for a key test in development of its next-generation Starship spacecraft, with a target of performing a ‘hover’ demo of its ‘Starhopper’ prototype rocket on Tuesday next week. The test will be the first time the rocket attempts liftoff with the new Raptor engine SpaceX is designing to succeed the Merlins used in its current Falcon 9 and Falcon Heavy rockets.

The hover test will involve getting the Starhopper off the ground to a height of about 20 meters, which is nowhere near high enough to prove its ability to launch things into space, but which is a key step in the development of both the Raptor engine and the reusable Starship spacecraft, which is the cornerstone of SpaceX’s ambitious plans to develop human presence in space, and eventually establish a colony on Mars.

SpaceX CEO Elon Musk noted that this first untethered test flight will also involve moving the Starhopper sideways, and that the company will hopefully also be able to launch the first version of its Starship spacecraft in just “a few months,” with a target altitude of around 20 km (around 65,600 feet). Note that Musk’s timelines tend to be highly optimistic, so adjust your expectations accordingly.

In April, Starhopper performed a tethered test, which took place near its Texas development facility, lasted all of two seconds, and followed quickly on the heels of an earlier quick ignition test of the Raptor engine firing on Starhopper for them first time.

Musk also previously said that the company is planning a more formal presentation of Starship, and that we could see this as early as later this month. This would be a chance for SpaceX (and presumably its CEO) to more fully articulate its plans for getting Starship operational, and what it hopes to accomplish once it does.

Some M&A is afoot in the world of market research and analysis: WPP today announced that it would sell 60% of Kantar — the firm that provides stats and insights on how consumers buy and think of products in services in areas like technology, media, health and more (we’ve written many a story on TechCrunch citing Kantar figures) — to Bain Capital, the private equity firm. The all-cash transaction is expected to net Kantar $3.1B — minus tax and continuing investments that it will make in Kantar after the deal — and it values Kantar at $4 billion (or £3.2 billion), London-based WPP said.

The deal is a biggie that caps off months of speculation, after WPP announced in October 2018 that it planned to look for an outside investor to take a stake in Kantar, in part to raise some revenue from the transaction, and in part to have fresh investment in the operation. The plan had always been for WPP to keep a stake, since there are a lot of areas where Kantar works with other parts of WPP, one of the world’s biggest advertising agencies.

Others who had been interested in buying the stake reportedly also included CVC, Apollo and Platinum.

The partial divestment underscores both how WPP has been reorganising and redefining itself in the wake of the departure of its longtime CEO and figurehead Martin Sorrell last year, who resigned under a cloud of controversy.

Mark Read, who took over as CEO in the wake of that, has taken a different approach when it comes to M&A, in part to offset sluggish growth, and this is one product of that.

“Kantar is a great business and we look forward to working with Bain Capital to unlock its full potential. As a strategic partner and shareholder in Kantar, WPP will continue to benefit from its future growth while our clients continue to benefit from its services and capabilities,” Read said in a statement. “I would like to thank [Kantar CEO] Eric Salama, his team and everyone at Kantar for their tremendous contribution to WPP – a contribution that will continue as we develop the business together. This transaction creates value for WPP shareholders and further simplifies our company. With a much stronger balance sheet and a return of approximately 8% of our current market value to shareholders planned, we are making good progress with our transformation.”

As ever more of our media consumption moves to digital platforms, companies like Kantar that have been built to track that activity have had an opportunity to growth their positioning and relevance to the bigger picture of how media is used.

Others that compete against it in this area include Nielsen and comScore. The latter has had a more challenging time of it, however, with a sizeable amount of corporate upheaval and a tumbling stock price: it recently announced that it would be raising up to $50 million to rebuild and recapitalise its business.

“Our new ownership structure presents a great opportunity for Kantar, our employees and our clients. In Bain Capital we have a partner who shares our ambition, brings relevant expertise and – with WPP – can help us accelerate our growth and impact for clients,” said Salama, in . “We are focused on delivering ‘human understanding at scale and speed’ and the ‘best of Kantar’ more consistently. We will do so by investing more in talent and by becoming a more technology-driven solutions provider.”

It’s not clear whether Bain was chosen as the highest bidder, or because it looked like the best partner for the deal in terms of common strategic goals, or a combination of both.

In any case, the plan will be to expand the business through more investments and acquisitions.

“Kantar is a market leader in many areas and we are excited to be partnering with its management team and WPP to build on this remarkable platform for growth,” said Luca Bassi, an MD at Bain Capital Private Equity, in a statement. “We see many opportunities for expansion and will invest in technology to expand the company’s capabilities and reinforce its global leading position.”

“We believe that we are well-positioned to support Kantar, alongside WPP, in driving forward the business in a rapidly changing industry,” added Christophe Jacobs van Merlen, another MD. “Our deep sector knowledge, operational expertise and strong track record of partnering with management teams to accelerate growth gives us confidence that we can help Kantar grow both organically and by acquisition.”

Other tech/media holdings in the Bain Capital portfolio include I Heart Media and ADK, an Asian ad agency.

VW Group is investing $2.6 billion in capital and assets into Argo AI, the Pittsburgh-based autonomous vehicle startup that burst onto the scene two years ago with $1 billion in backing from Ford.

The deal, which has been rumored for months, is part of a broader alliance between VW Group and Ford that covers autonomous and electric vehicles.

The Argo piece of this tie-up involves more than just an injection of capital, in return for a stake in the startup and board seats. It turns Argo into a global company, or at least one with operations in U.S. and Europe. And it instantly boosts its staff by 40%.

The deal

VW has committed $1 billion in capital into the startup and also will purchase Argo AI shares from Ford for $500 million over three years. Ford will invest the remaining $600 million of its previously announced $1 billion cash commitment in Argo AI.

VW is also handing over Autonomous Intelligent Driving, the self-driving subsidiary that was launched just two years ago to develop autonomous vehicle technology for the Volkswagen Group. AID is valued at $1.6 billion.

The Munich-based AID team will become Argo’s European headquarters, a move that will expand its staff 40% to more than 700 employees.

The deal raises Argo’s valuation to more than $7 billion. Despite the extra contribution of AID, Ford and VW Group will hold equal stakes in Argo. The remaining equity has been set aside for employees, the companies said Friday. The deal is still subject to the approval of regulators.

Argo AI is developing the virtual driver system and high-definition maps designed for Ford’s self-driving vehicles.

Argo will treat VW and Ford as separate customers. This means Volkswagen and Ford will independently integrate Argo AI’s self-driving systems into its own purpose-built vehicles. Argo AI’s focus remains on delivering a SAE Level 4-capable SDS to be applied for ride sharing and goods delivery services in dense urban areas.

“Our agreement with Volkswagen positions us as a technology platform company, expands the potential
geography for deployment and will further fuel our product development,” Argo AI CEO and co-founder Bryan Salesky said in a blog post making the announcement.

Automakers Ford and Volkswagen have announced a partnership today that covers a number of areas, including autonomy (via a new investment by VW in Argo AI) and collaboration on development of electric vehicles. This EV tie-up will see Ford use Volkswagen’s MEB platform, which it’s using as the core of its forthcoming line of consumer electric vehicles, to develop “at least one” fully electric car for the European market that’s designed to be produced and sold at scale.

Volkswagen’s MEB is a big bet by the German automaker, meant to provide for all-electric models what the MQB platform before it did for the automaker’s internal combustion engine cars. The idea behind these platforms is that they are modular and flexible enough to cover a range of different vehicle types, while ensuring that there’s enough of a repeatable core that the cost of redevelopment from model to model is greatly decreased.

The MEB platform is already planned for use across a number of announced vehicles to be released by VW and VW-group automakers, including Audi, SEAT, Skoda and more between 2019 and 2023. Ford will be the first announced automaker outside of the Volkswagen Group to make use of the MEB. The automakers specify that their are only concrete plans for one model at the moment, but the option to expand that to more pending how that initial collaboration goes is baked into the deal.

Ford also plans to deliver over 600,000 cars for the European market based on the MEB architecture over the course of six years for the first model alone, and the automakers note in a press release that there are considerations for a second Ford model based on the platform to be developed. Ford says in the release that this is just part of its overall, ongoing commitment to EVs, and that its work on crossover and other imported U.S. market models including Mustang and Exploer for Europe will continue. Both automakers will also remain independently and separately owned.

The general details of a partnership between the two automakers was reported as in the works by Reuters earlier this month, but is now confirmed and official by both automakers.