Year: 2018

Amidst vague and uninformed questions during today’s House Judiciary hearing with Facebook, Google, and Twitter on social media filtering practices, Representative Steve King (R-Iowa) dropped a bombshell. “What about converting the large behemoth organizations that we’re talking about here into public utilities?”

King’s suggestion followed his inquiries about right-wing outlet Gateway Pundit losing reach on social media and how Facebook’s algorithm worked. The insinuation was that these companies cannot properly maintain fair platforms for discourse.

The Representative also suggested that there may be need for “review” of Section 230 of the Communications Decency Act that protects interactive computer services from being treated as the publisher of content users post on their platforms. If that rule was changed, social media companies could be held responsible for illegal content from copyright infringement or child pornography appearing on their platform. That would potentially cripple the social media industry, requiring extensive pre-vetting of any content they display.

The share prices of the tech giants did not see significant declines upon the Representative’s comments, indicating the markets don’t necessarily fear that overbearing regulation of this nature is likely.

Here’s the exchange between King and Google’s Global Head of Public Policy and Government Relations for YouTube Juniper Downs:

King: “Ms Downs, I think you have a sense of my concern about where this is going. I’m all for freedom of speech, and free enterprise, and for competition and finding a way that competition itself does its own regulation so government doesn’t have to. But if this gets further out of hand, it appears to me that Section 230 needs to be reviewed.

And one of the discussions that I’m hearing is ‘what about converting the large behemoth organizations that we’re talking about here into public utilities?’ How do you respond to that inquiry?”

Downs: “As I said previously, we operate in a highly competitive environment , the tech  industry is incredibly dynamic, we see new entratnts all the time. We see competitorsacross  all of our products at google, and we believe that the framework that governs our services is an appropriate way to continue to support innovation.”

Unfortunately, many of the Representatives frittered away their five minutes each asking questions that companies had already answered in previous congressional hearings or public announcements, allowing them to burn the time without providing much new information. Republican reps focused many questions on whether social media platforms are biased against conservatives. Democrats cited studies saying metrics do not show this bias, and concentrated their questions on how the platforms could protect elections from disinformation.

Protestors during the hearing held up signs behind Facebook’s Head of Global Policy Management Monica Bickert showing Facebook CEO Mark Zuckerberg and COO Sheryl Sandberg as heads of an octopous sitting upon a globe, but the protestors were later removed.

One surprise was when Representative Jerrold Nadler (D-New York) motioned to cut the hearing for an executive session to discuss President Trump’s comments at the Helsinki press conference yesterday that he said were submissive to Russian president Vladimir Putin. However, the motion was defeated 12-10.

Serial entrepreneur and angel investor Elad Gil has become renowned in Silicon Valley startup circles, largely because he not only funds startups but he advises many, too. Some of the tech companies he has logged time with include Airbnb, Twitter, Google, Instacart, Coinbase, Stripe, and Square.

All have evolved into massive growth companies, and because Gil says he has seen common challenges emerge as each has taken off, he decided to write about how to address some of these challenges in a book that’s being released today called The High Growth Handbook.

Interestingly, Stripe, the online payments company, is the publisher, having recently added book publishing to the list of services it offers.

We talked with Gil about the book recently to learn more.

TC: What was the impetus for this project? You were tired of repeating yourself when talking with founders?

EG: I decided to write it after seeing a lot of the common patterns that late-stage companies follow. A lot of them wind up having exactly the same issues.

TC: Including . . .

EG: Some of the biggest issues center on what a founder’s role as CEO should be as a company scales, how to hire a board, how to hire executives, how to do re-orgs, how to buy a company for the first time, how to build a a  real product organization, how to do late-stage financings, how to think about secondaries offerings . . .

TC: That’s a lot of issues! But aren’t companies getting advice from their investors along the way? Isn’t that partly the point of working with venture capitalists?

EG: The truth is founders sometimes don’t want to raise these questions out of fear that they’ll  seem naïve or not fully up to the role. You also see some VCs who understand early-stage really well but who haven’t themselves operated at scale and maybe don’t have a lot of experience with re-orgs and M&A.

TC: Going back to the many challenges you raise, which do founders most often fumble?

EG: This biggest mistakes you see tend to be around hiring and leadership. That’s probably the most likely reason for outright failure.

TC: Can you elaborate?

EG: Say you’re an engineer who started a company. You don’t know the role of VP of sales or CFO or COO. Executives are very tricky versus individual contributors and these are people you’re going to spend a lot of time with and who may define the role differently than you might assume they would. Some of the questions you need to be asking is can they hire great people and manage them productively? Are they strategic? Are they collaborative? Are they ethical people who will fundamentally do the right thing?

TC: You also mentioned M&A, which seems like a minefield for any executive team, given that acquisitions so often don’t work out. What do founders misunderstand about M&A?

EG: I think that’s partly the misunderstanding, that M&A can be a mistake. You can think of M&A two ways: from a  portfolio perspective, meaning you are overseeing strategic assets, or that you’re buying something that you’re willing to experiment with. Either way, throughout the history of tech, most of the great companies have gotten where they are through acquisitions. Cisco, Microsoft, Google, Facebook. You have to be willing to have things not work, but it’s a fallacy to think it doesn’t work generally, which you sometimes see at companies with less experienced employees. They might think, “Why buy this company when we can hire 10 engineers instead and build the same thing?” What that misses is that companies can be under-resourced and constrained but they could be working on something good. Companies also sometimes think they have to integrate tech stacks when they should let something run independently and just redistribute the product.

TC: You’ve been in the startup world a while. What are some of the biggest shifts you’ve seen, aside from maybe all the late-stage capital we’ve seen flood into Silicon Valley?

EG: One of definitely the rise of the COO, a change ushered in by Facebook’s Sheryl Sandberg. Before her, founders were often relegated to secondary roles as we saw when Eric Schmidt took over for Larry and Sergey at Google. What used to happen: you’d hire a CEO and the you’ve move into a kind of VP role. Now, founders are able to stay with the company with the support of a COO and not give up control.

Obviously, too, the time to IPO has lengthened considerably, driven by all the late-stage capital that’s available. There’s a generation of entrepreneurs that’s been scared to go public out of fear of the enormous overhead associated with it, when the reality is you do not need to hire 100 people to deal with a public offering. They also worry that Wall Street is too quarter-to-quarter focused, but if you look at companies like Google and Facebook and Amazon [and their ability to continue to innovate], that’s probably not true. And you see arguments around hiring, but when you go public, you can pay people less because they know their stock is real and liquid.

Thankfully, companies like Spotify and Dropbox and Square have started setting the tone that it’s okay to get out there.

Tinder will begin testing Bitmoji in its app in Canada and Mexico, the company announced today. The new integration comes courtesy of the recently launched Snap Kit, which allows third-party apps to take advantage of Snap’s login for sign-up and features, like its Snap Map and Bitmoji, among other things. In Tinder, users in the test regions will be able to send the Bitmoji to their matches within their chat conversations.

Tinder was one of Snap’s debut partners for Snap Kit, along with Patreon and Postmates. However, it hadn’t yet launched the integrations until today.

“With our Bitmoji integration, we’re giving users a playful new way to engage with matches. This is just one way we work with partners to add features that encourage users to experiment with more personalized ways of chatting; in this case, it’s the freedom to get creative with avatars,” said Tinder Chief Product Officer Brian Norgard, in an announcement.

This isn’t the first time Tinder has added functionality to enhance its chat experience. It already offers the ability to use emoji, and, thanks to a 2016 partnership with Giphy, users can send GIFs to matches to help break the ice and have more playful conversations.

For those who have access to Bitmoji, the option will appear in place of the emoji button in the chat interface. When you press the green Bitmoji icon (next to the GIF button) for the first time, you’ll have to tap “Connect to Snapchat” to authenticate.

From then on, you can search across your Bitmoji collection using keywords in the search box, or you can tap on the color-coded bubbles that aggregate commonly used Bitmoji expressions like “good morning,” “coffee,” “busy,” and others. You can also tap on “recents” to find those you have used in the past.

Tinder has been running a growing number of experiments as of late, having launched tests of things like Tinder Places for finding matches you may cross paths with, A.I. suggestions on who to “Super Like,” a real-time feed of social updates, curated selections called Tinder Picks, a video feature called Tinder Loops, and more, over the past six months or so. The video feature rolled out globally earlier this month, which indicates that at least some tests will turn into product features for all to use.

Tinder says the Bitmoji feature is in testing in Canada and Mexico, but didn’t confirm if or when it would roll out to other markets, like the U.S.

 

Ubisoft has implemented a new system in Rainbow Six Siege that bans players for using toxic language in the text chat, according to PC Gamer.

Yesterday, a number of players started whining about being temporarily banned from the game after using a racist or homophobic slur in the text chat. The first offense results in a 27 minute ban. Second and third offenses will cost players 2 hours of game time, and any following toxic language will result in an official investigation and potentially a permanent ban from the game.

Like most games, Ubisoft’s Rainbow Six Siege has a Code of Conduct that forbids “any language or content deemed illegal, dangerous, threatening, abusive, obscene, vulgar, defamatory, hateful, racist, sexist, ethically offensive or constituting harassment.” However, most games do close to nothing to enforce these rules, which has led to an overwhelmingly toxic gaming community overall.

This new banning system doesn’t come out of the blue. Ubisoft has been talking about removing toxic language from the platform for a while. In fact, the dev team wrote in April that it has plans to add new features to limit the use of offensive language on the game.

From the April post:

Our team is working on the creation of an automated system that will censor text chat in game based on a chat filter list. This will replace words that have been identified as offensive and provide players with a notification that their language was found to be unacceptable. We will also be tracking the number of times players trigger this filter and will take action as necessary for players that are intentionally having a negative impact on other player’s gaming experience.

The current iteration of the banning system doesn’t censor the offensive words in text chat but rather goes straight to the 27-minute ban.

The gaming world has grown into an increasingly toxic environment, as there has been little to no policing of users’ behavior and communication. But eSports and Twitch are putting gamers in the spotlight, and brand endorsements are a huge piece of the upward trajectory of esports. But no brand is going to get behind a gamer that uses homophobic or racial slurs.

Ubisoft says on Twitter that the ban system is still being updated and the team is taking feedback from the community to continue refining it, as well as punishing players who are cheating.

Despite the massive glitches that prevented online shoppers from being able to browse and buy from Amazon’s Prime Day sale yesterday, the retailer is this morning claiming its Prime Day sales in the U.S. are “bigger than ever,” and grew faster than last year’s Prime Day within the first ten hours of the annual sales event. In addition, customers bought “millions” of Amazon devices, including Amazon’s top sellers Fire TV Stick, Echo Dot, Fire 7 tablet, as well as the Instant Pot 6 Qt, and the Lifestraw Personal Water Filter for Hiking, Amazon said.

While Amazon doesn’t tend to reveal hard numbers, a third-party report from Feedvisor backs up Amazon’s statement to some extent, saying that online shoppers spent 54 percent more in the first three hours of Prime Day this year, than the first three hours last year, according to a report from Bloomberg.

Amazon acknowledged the website’s struggles in a statement published on Twitter yesterday, saying it was “working to resolve this issue quickly.” But it also noted that customers had ordered more items in the first hour of Prime Day compared with the first hour the year prior.

The retailer this morning again referenced the site issues which prevented shopping (potentially to the tune of millions in lost sales) but has not offered details as to what, specifically, went wrong.

“It wasn’t all a walk in the (dog) park, we had a ruff start – we know some customers were temporarily unable to make purchases,” the company said in a statement released today. (The dog puns are references to the 404 pages featuring the dogs of Amazon that appeared when the site was non-functional.)

Though the Prime Day sale continues today through 11:39 PM PT, not everyone may be willing to give Amazon a second shot –  some angry customers were already threatening to cancel their Prime memberships and shop elsewhere, according to reports.

While Amazon’s site has experienced glitches on previous Prime Days due to the massive influx of traffic, yesterday’s downtime was unprecedented.

The problems with Amazon’s site began shortly after Prime Day kicked off at 3 PM ET on Monday, June 16.  The landing page for Prime Day didn’t work, and when links were clicked, visitors were often sent to error pages. The deal navigation page was also down for an extended period of time, which impacted shoppers’ ability to browse the items on sale. And when shoppers did manage to load items into a cart and checkout, they would sometimes get errors there, too.

Web performance monitoring firm Catchpoint began running detailed tests at 4 PM ET, which confirmed the site load times, on average, were far slower than usual.

Typically, users may wait 2 to 3 seconds for a website home page to load, but during the Prime Day outage, the home page was taking 6, 7 and even as high as 8 or 9 seconds to load on the desktop. And on mobile, webpage load times were even slower – reaching as high as 30 seconds at times across the U.S.

Error pages (the ones with the Amazon dogs) were also appearing more than usual, as indicated in the chart above.

The issues improved around 4:50 PM ET, but customers still experienced spotty problems with Amazon’s app as late as 8:17 PM ET, Catchpoint said.

In addition, even when the Amazon.com homepage became accessible to users, product search was not working and shoppers would see an error message (below) in some instances. This continued from 4:15 PM to 10 PM ET nationwide.

The situation this morning is resolved, according to Catchpoint’s tests, but if an influx of shoppers return to get the sales they couldn’t manage yesterday, it could impact the site again.

Prime Day was estimated to bring in $3.4 billion in sales, up over 40 percent from last year, Bloomberg had reported. It remains to be seen if Amazon will still meet those estimates, and what – if anything – it may do to placate brands whose limited-time deals passed by during the outage.

 

 

 

Let’s be real: you are most certainly never going to be as good as Steve Nash, Chris Paul, James Harden — or really any professional NBA player. But it probably won’t stop you from trying to practice or model your game around your favorite players, and spend hours upon hours figuring out how to get better.

And while there are going to be plenty of attempts to smash image recognition and AI into that problem, a company called NEX Team is hoping to soften the blow a bit by helping casual players figure out their game, rather than trying to be as good as a professional NBA player. Using phone cameras and image recognition on the back end, its primary app HomeCourt will measure a variety of variables like shot trajectory, jump height, and body position, and help understand how to improve a player’s shooting form. It’s not designed to help that player shoot like Ray Allen, but at least start hitting those mid-range jumpers. The company said it’s raised $4 million from Charmides Capital and Mandra Capital, as well as Steve Nash, Jeremy Lin, Sam “Trust The Process” Hinkie (sigh), Mark Cuban and Dani Reiss.

“We don’t call ourselves a basketball company, we think of ourselves as a mobile AI company,” CEO and co-founder David Lee said. “It happens that basketball is the first sport where we’re applying our tech. When you think about digitizing sports, as a runner or cyclist, you’ve had access to a feedback loop for a while [on treadmills and other tools]. But for basketball and other sports like basketball, that loop didn’t exist. We believed with computer vision, you can digitize a lot of different sports, one of which is basketball. We’re not just building an app for the professional basketball athletes, we’re focused on building an app where value can be generated across the basketball community.”

The app starts off with an iPhone. Players can boot up their camera and begin recording their shots, and the app will go back and track what worked and what didn’t work with that shot, as well as where the player is making and missing those shots.It’s not tracking every single motion of the player, but once a player makes a shot, it will track that trajectory and shooting form, like where his or her feet are planted. That kind of feedback can help players understand the kinds of small tweaks they can make to improve their shooting percentage over time, such as release speed or jump hight. And while it’s not designed to be hugely robust like the kinds of advanced tracking technology that show up in advanced training facilities at some larger sports franchises, it aims to be a plug-and-play way of getting feedback on a player’s game right away.

Still, that doesn’t necessarily stop the app from showing up in slightly more professional situations, like recruiting or in athletic centers on college campuses, Lee said. Each college is looking for the next DeAndre Ayton or Ben Simmons, as well as new ways to try to find those recruits. While not every college will end up with the top recruits in the country and get bounced in the second round of the NCAA Men’s Basketball tournament, it offers an additional way for younger players to refine their game to the point they potentially get the attention of those universities — or the NBA, should the one-and-done rule that requires athletes to play a year in college end up disappearing.

“A lot of these coaches are looking at a lot of evaluation tools,” Lee said. “If Alex is waking up at 5 a.m. to put in work, it’s not just about makes and misses, it’s about work ethic. It’s harder to evaluate and digitize a sport. Only [a fraction] of the basketball happens in their practice facilities. How do they help their players evaluate their workout sessions when they’re in those situations? That opens up the doors to do that as well.”

In order to appeal to those broader audiences, the startup is rolling out bite-sized challenges as a way to try to attract the more casual consumers that want to dip their toes into HomeCourt. You see these kinds of challenge-based activities in apps like Strava as a way to try to attract users or keep them engaged in a lighter and more competitive way without having to go into a full-on event like a race or a tournament. It’s one way to try to wrangle the competitive elements of sports like basketball without a ton of competitive pressure as users get more and more comfortable with the way they play and their shooting style.

That bite-sized style of activity also serves pretty well when it comes to creating content, as has been proven popular by apps like Overtime that specialize in highlights of certain players. HomeCourt hopes to add a social layer on top of that to, once again, increase that kind of stickiness and build a community around what would otherwise be a purely technical tool — and one that might scare off more casual players with a very sabermetrics-feeling approach.

Lee also said he hopes the app will eventually broaden into other sports, like Golf or Tennis, where tracking the ball might be more complicated or the motions considerably different from basketball. That’s based on building technology that tracks the movement of the player, and not just the ball, in order to determine the trajectory or success of that specific shots. The hope is that basketball is a first step in terms of achieving that.

“For golf, seeing your whole form as going into your swing is more important — that’s the input in terms of getting where the ball goes,” Lee said. “We’re trying to think about how to reduce as much friction as possible. Imagine being able to use the app to track makes or misses, but also tracking your player movement and form, measuring it, and comparing it to another player’s backswing. We’re hoping to do that in basketball [first].”

Investor Connie Chan has made such an impact on her colleagues at Andreessen Horowitz (a16z) since joining the firm in 2011 that today, they’re announcing Chan’s appointment to general partner.

It’s a big deal for the Sand Hill Road venture firm for numerous reasons. First, Chan becomes just the second woman in the firm’s nine-year history to be appointed to the role of general partner. As readers may recall, a16z announced the first female general partner in its history late last month, bringing aboard former federal prosecutor Katie Haun to help lead its new cryptocurrency fund.

According to Jeff Jordan — himself an a16z general partner and the former CEO of OpenTable — the nine-year-old firm has also never before promoted someone from within to the post of general partner, instead pulling in people like Jordan himself with senior operating experience. (Jordan also distantly held posts as the president of PayPal, the CFO of Hollywood Entertainment, and the CEO of Reel.com.)

The “policy was not to promote internally,” says Jordan. Yet such thinking has changed recently as 16z has grown and the operating functions that it uses to support its portfolio companies have matured. Indeed, when it came to Chan, a Stanford alum who logged four years with the private equity firm Elevation Partners and another two years in product management roles at HP and Palm, a lack of direct experience in scaling a business was eventually outweighed by her ability to identify and support talented founders, Jordan says.

He credits Chan, for example, with the firm’s initial investment in the digital scrapbook site Pinterest. “It wasn’t a contentious discussion, but Connie really pounding the table is what led to our investment in the company,” he says.

Pinterest was most recently valued at more than $12 billion. When a16z led a $27 million round in the company back in 2011, it was reportedly valued at $200 million.

Chan also brings to the table two other things desperately needed at Andreessen Horowitz: an understanding of China’s market and key contacts there. From her earliest days with the firm, Chan has spearheaded its Asia network, helping the firm’s portfolio companies navigate regional opportunities through quarterly visits to the country, as well as gaining a “one- to four-year advantage in [understanding] consumer mobile” trends in the U.S., which often start first in China, she notes.

Consider that Chan pushed a16z to invest in the electric bike and scooter company Lime last year after studying the unit economics of Lime’s China-based predecessors Ofo and Mobike. (She calls Lime’s very recent funding, which was led by Uber and GV and will see Lime more closely integrated with Uber, a “big, big positive.”)

Chan, who’d worked for HP in Beijing, also introduced the a16z portfolio company, Lyft, to the China-based ride-hailing giant Didi back in 2015. Didi wound up investing $100 million investment in Lyft as a result and announcing a ride-sharing, cross-country collaboration as part of the tie-up.

If you’re wondering if Chan’s new role suggests in any way that Andreessen Horowitz might begin making direct investments in China, she says it will not, that the “focus will still be on the U.S., with most of our deals in Silicon Valley.”

We separately ask if the firm might be gearing up to announce a new fund, given the recent appointments of both Chan and Haun — who bring the number of general partners at a16z to 12 — and the fact that the firm announced its most recent, flagship fund a little more than two years ago. The answer we are given: “There’s nothing to announce at this time.”

Congress may soon vote on a new bill that would set net neutrality down as a matter of law rather than a set of rules to be changed every few years by the FCC. The “21st Century Internet Act,” introduced by Rep. Mike Coffman (R-CO), would ban blocking, throttling, paid prioritization, and eliminates all questions of jurisdiction.

The bill, announced online and at an event in Washington, DC today, would modify the Communications Act of 1934 (greatly built upon by the 1996 Telecommunications Act) and add a new “Title VIII” full of stipulations specific to internet providers.

This would settle the decades-long dispute over whether internet access is an “information service” or a “telecommunications service,” a legal distinction that either reins in (the former) or unleashes (the latter) the FCC on ISPs.

Instead of quibbling over whether the FCC has authority to write the rules or not, and then quibbling over the rules themselves, the act just codifies the rules as law and sets the FCC as the official watchdog.

The Commission shall have the authority to initiate investigations, bring enforcement actions, issue declaratory rulings, conduct rulemakings, and take other such actions… necessary to implement the requirements of this title.

It would no longer be a question of whether the FCC wants to have net neutrality rules or not — net neutrality would be the law and it would unequivocally be the Commission’s job to enforce it.

The basic blocking, throttling, and paid prioritization bans are very similar to the 2015 rule’s, and the law even institutes the “general conduct” rule that many complained was too vague. This catch-all rule says an ISP “may not unreasonably interfere with or disadvantage” users or edge providers from accessing or providing lawful content and services.

Because it isn’t specific, it means practices that may or may not be legal, such as zero rating, have to be evaluated case by case. That can be a lot of work — but it’s hard to think of a better way to provide against the shifting tactics of crafty ISPs.

Interestingly the act would require the FCC to investigate “unfair or deceptive acts or practices,” something that is frequently on the FTC’s plate — false advertising, misrepresenting the product, that kind of thing. Presumably this is to settle any jurisdictional dispute there, though the FTC may still come into play here and there.

Broadband providers would be eligible to receive money from the FCC’s Universal Service Fund, which it uses to help fulfill its mainline duty of making sure communications infrastructure is up to snuff. And while it could ask ISPs to contribute to the fund, the FCC is barred from rate regulation — telling providers what they can and can’t charge for their services. It was a worry that under the 2015 rules, the only thing stopping the Commission from doing so was voluntary forbearance (the technical term for opting out of statutory authority) from the capability to do so.

They also wouldn’t count as “common carriers,” a designation that comes with other responsibilities and oversight. This choice is practical if not, some may argue, completely correct: internet providers really do seem to qualify as common carriers as they are generally defined. But before the main reason for designating them as such was to justify the application of Title II, which granted the FCC authority to enforce the 2015 rules.

It’s all very confusing, right? This law, however, really cuts through a lot of the cruft of the past few decades and clearly establishes the “bright line” rules consumers know and understand.

Who will oppose it and why? Broadband providers will of course say first that the act is unnecessary because they’ve pledged to follow the rules voluntarily, and second that it will prevent them from innovating with services that technically break rules but are in fact beneficial to consumers. Don’t be fooled — these services don’t exist and never did. The only one that comes close is zero rating, and it’s a sham.

Remember, broadband providers also loudly called for “regulatory certainty” instead of the seesawing rules of the last five years. Be careful what you wish for!

Conservatives may oppose it because it expands regulations rather than reduces them, which is at least a valid position to take, as little as consumers may agree with it. The rules also have their origins in the Obama administration, which makes them burn partisan politicians at the touch.

It’s always hard to say whether a bill will be a success, and the processes are so slow anyway that it might be a year or more before we see this on the floor of the House. As for the President, it’s hard to say what he’ll do, as with so many other issues. That it would be difficult to cast the bill in partisan terms, practical as it is and introduced by a Republican Congressman, but that doesn’t mean it won’t be vetoed anyway.

Notably Rep. Coffman is lighting the candle under the FCC at both ends (if you’ll permit the confused metaphor) by supporting the House’s Congressional Review Act petition to undo the current administration’s rules: “While my bill moves through the Congress, I am taking an ‘all of the above’ approach by simultaneously signing the discharge petition on the CRA, and introducing my bill.”

At the very least the bill seems a good example of a law that takes the short path to providing consumers with net neutrality protections: direct and to the point, with no sweetheart clauses or funny business as far as I can tell. That said, it may yet be butchered in committee. We’ll follow its progress carefully.

As Amazon looks to increasingly expand its cashier-less grocery stories — called Amazon Go – across different regions, there’s at least one startup hoping to end up everywhere else beyond Amazon’s empire.

Standard Cognition aims to help businesses create that kind of checkout experience based on machine vision, using image recognition to figure out that a specific person is picking up and walking out the door with a bag of Cheetos. The company said it’s raised an additional $5.5 million in a round in what the company is calling a seed round extension from CRV. The play here is, like many startups, to create something that a massive company is going after — like image recognition for cashier-less checkouts — for the long tail businesses rather than locking them into a single ecosystem.

Standard Cognition works with security cameras that have a bit more power than typical cameras to identify people that walk into a store. Those customers use an app, and the camera identifies everything they are carrying and bills them as they exit the store. The company has said it works to anonymize that data, so there isn’t any kind of product tracking that might chase you around the Internet that you might find on other platforms.

“The platform is built at this point – we are now focused on releasing the platform to each retail partner that signs on with us,” Michael Suswal, Co-founder and COO said. “Most of the surprises coming our way come from learning about how each retailer prefers to run their operations and store experiences. They are all a little different and require us to be flexible with how we deploy.”

It’s a toolkit that makes sense for both larger and smaller retailers, especially as the actual technology to install cameras or other devices that can get high-quality video or have more processing power goes down over time. Baking that into smaller retailers or mom-and-pop stores could help them get more foot traffic or make it easier to keep tabs on what kind of inventory is most popular or selling out more quickly. It offers an opportunity to have an added layer of data about how their store works, which could be increasingly important over time as something like Amazon looks to start taking over the grocery experience with stores like Amazon Go or its massive acquisition of Whole Foods.

“While we save no personal data in the cloud, and the system is built for privacy (no facial recognition among other safety features that come with being a non-cloud solution), we do use the internet for a couple of things,” Suswal said. “One of those things is to update our models and push them fleet wide. This is not a data push. It is light and allows us to make updates to models and add new features. We refer to it as the Tesla model, inspired by the way a driver can have a new feature when they wake up in the morning. We are also able to offer cross-store analytics to the retailer using the cloud, but no personal data is ever stored there.”

It’s thanks to advances in machine learning — and the frameworks and hardware that support it — that have made this kind of technology easier to build for smaller companies. Already there are other companies that look to be third-party providers for popular applications like voice recognition (think SoundHound) or machine vision (think Clarifai). All of those aim to be an option outside of whatever options larger companies might have like Alexa. It also means there is probably going to be a land grab and that there will be other interpretations of what the cashier-less checkout experience looks like, but Standard Cognition is hoping it’ll be able to get into enough stores to be an actual challenger to Amazon Go.

Expense management software provider Certify is beefing up its artillery against rival Concur with the acquisition of Abacus, which enables companies to deal with expenses in real time. The deal’s financial terms were not disclosed. The addition of Abacus will help Certify, which includes other expense management solutions like Nexonia and ExpenseWatch under one umbrella, become a stronger rival to SAP-owned Concur by reaching new customer segments.

Founded by Omar Qari, Josh Halickman and Ted Power, Abacus says it was the first real-time expense reporting solution on the market when it launched in 2013. The Y Combinator alum, whose investors included General Catalyst, Bessemer Venture Partners, Google Ventures and Salesforce Ventures, currently counts 1,000 customers. Its team will join Certify and the Abacus product will continue to be independent.

Expenses are a bane for everyone involved: the employees who need to turn in receipts, the managers who have to approve them and everyone in the finance department who needs to reconcile corporate credit cards and make sure company policy is followed. Abacus eases their pain with features like automatic expense suggestions and Slack integration for employees.

For companies, it lets them set prompts to enforce spending limits and make sure details, like client names, are filled in correctly. If expenses need to be approved by specific managers or departments, Abacus routes them to the right person. Real-time analytics also help companies make quick budget decisions.

Abacus’ clients include Betterment, Dropbox, GLG and North American Substation Services. In an email, Qari, the CEO of Abacus, told TechCrunch that Abacus is a good fit for “companies that need out-of-the-box flexibility in approval flows, spending controls and ERP sync.”

For example, he said North American Substation Services, which provides installation, repair and maintenance work for high-voltage stations, uses Abacus to speed up its account receivables by billing back expenses closer to when they actually happened, while Dropbox chose Abacus to reimburse interview candidates more quickly.

Certify was acquired by K1 Investment Management last year and combined with expense management software providers Nexonia, ExpenseWatch and Tallie to serve a total of 10,000 businesses. Certify says this makes it the largest competitor to Concur.

Each brand operates independently, serving its own niche, like Abacus will. Qari said that “in a prospect overlap analysis, we found hardly any opportunities are common across the portfolio, highlighting how unique each brand’s segment is. The expense management industry’s typical customer profile is fairly fragmented, so it’s going to take multiple solutions approaching the space from multiple angles to fully satisfy market demand.”