OK Google, what’s your next move?

In-depth

Facing growing worker unrest, serious competition from Amazon in digital advertising, and several missed opportunities to pivot, is Google doomed to decline and irrelevance, or does it still have some cards to play? The tech giant’s hardware may provide a clue.

Google’s problems are multiplying. Earlier this month, thousands of Google employees around the world staged a mass walkout to protest the company’s handling of sexual harassment complaints, and to demand worker representation on the Alphabet board. This industrial action is just the most recent of a series of high-profile incidents at Google which suggest a fractious internal culture is developing.

To compound woes at the company’s Mountain View campus, Google fell foul of EU antitrust regulations earlier this year. Having incurred a fine of $5 billion for forcing manufacturers to pre-install a suite of apps on Android phones and tablets, Google now has to license its search and productivity software to hardware companies, opening it up to competition.

But occasional antitrust fines and unrest among the workforce are to be expected at a near-monopolist that employs upwards of 70,000 people around the world. They might not even be a serious concern for a company that was performing well commercially. Unfortunately, Alphabet isn’t. Stock prices tumbled last month in response to a third quarter earnings report that showed the company had missed its revenue projections.

Since it’s fighting heavyweight competition for market share in several verticals where it’s spent billions in R&D, it could be tempting for the search giant to retreat into its core activity: serving ads. But that business is under threat too, as both consumers and lawmakers grow increasingly privacy-conscious, and shoppers look to other platforms for product searches.

Page, Brin and Pichai will surely be familiar with the fate of former seemingly invulnerable technology behemoths, like AT&T and General Electric, that lost the ability to innovate and suffered a slow decline into irrelevance. Can the company that put ‘algorithm’ into the common lexicon avoid a similar destiny, and what pivots are still available to Alphabet that would keep it in the race to be one of the first $1 trillion companies?

Mutiny at Mountain View

The global walkout by an estimated 20,000 Google employees on November 1st was primarily about protesting the company’s history of releasing executives facing sexual harassment allegations with large severance packages. But it was also a culmination of growing unrest among Google staff over the values of a company which recently removed the famous “don’t be evil” directive from its official code of conduct. In June the company announced that it would not renew a Google Cloud contract with the Pentagon after 4,000 employees signed a petition demanding that Google stop building “warfare technology”. Then in August, hundreds of Google staffers signed a letter demanding that the company’s secret plans to build a censored search engine for the Chinese market be made transparent.

The huge demand in Silicon Valley for talented software engineers has long acted to drive up salaries. But collusion between the industry’s top employers on not hiring each other’s staff (Google, Apple, Adobe and Intel settled an antitrust lawsuit for $415 million in 2015) and the dominance of an individualist, libertarian ideology among coders has worked to keep employee power in check. Until now.

The visceral reaction of Google employees to the James Damore memo, which attempted to make the case that women were underrepresented in technology jobs due to biological differences, demonstrated that while a thriving culture exists at Google, it’s one that is deeply divided and not under the control of the company’s leadership. Staff are realising that, acting collectively, they have power, as the walkout organisers indicated in The Cut:

“A company is nothing without its workers. From the moment we start at Google we’re told that we aren’t just employees; we’re owners. Every person who walked out today is an owner, and the owners say: Time’s up.”

Their demands include an end to forced arbitration and adding worker representation to Alphabet’s board. While Google could spin the whole episode for positive press by quickly acceding to at least some of the requests, the underlying battle for the soul of the company is likely to continue, eating up the bandwidth of the leadership team to the detriment of long-term commercial strategy.

Monopolies ain’t what they used to be

On the same day as the Google walkout, web inventor Tim Berners-Lee added his voice to the growing chorus calling for the big tech companies to be broken up. Disappointed at the current state of the internet, he told Reuters that government intervention to curtail the monopoly power of Silicon Valley was inevitable:

“What naturally happens is you end up with one company dominating the field so through history there is no alternative to really coming in and breaking things up. There is a danger of concentration.”

All of the consumer-facing firms that NYU professor Scott Galloway refers to as The Four (Google, Facebook, Amazon and Apple) are increasingly coming under scrutiny from lawmakers, academia and the commentariat due to the massive amounts of personal data they collect and process, and the commercial power that they wield. Galloway told Recode why checking monopoly power is important:

“If the four largest retailers — Walmart, Kroger’s, Home Depot and CVS — got together every morning and said, ‘We’re going to focus all our resources on putting No. 5, Target, out of business,’ Target would go out of business. That’s what’s going on with Snap. The four biggest players — Instagram, Facebook, WhatsApp and [Facebook] Messenger — every morning meet and say, ‘How do we put these guys out of business?’”

Google’s EU fine for abusing the market power it wields through the Android operating system and the Google Play store isn’t the first time it’s received a rap on the knuckles from the european competition commissioner. It follows a similar €2.1bn penalty levied on the search giant in 2017 for “[abusing] its market dominance as a search engine by promoting its own comparison shopping service in its search results, and demoting those of competitors.”

While Google isn’t the only tech giant under the scrutiny of the EU’s competition authority – preliminary investigations have been launched into Amazon’s use of third-party data and Apple’s disregard of efforts to standardise phone charger designs – the successful antitrust actions against Google provide plenty of clues for similar probes by US authorities, who have the power to do more than impose punitive fines.

The sacking of US Attorney General Jeff Sessions after the US midterm elections might delay any nascent antitrust investigations by that office, but a marked antipathy towards Google from the White House signals danger. US president Donald Trump has repeatedly accused Google’s search results of being “RIGGED” in favour of liberal publishers and, in August, suggested that Google, Facebook and Amazon could present “a very antitrust situation, the three of them”.

This marks a step change in political relations for Google, which spends millions more than its nearest rivals on lobbying US lawmakers ($18 million in 2017), and whose top brass were notoriously close to the Obama administration. The recent resignation of Susan Molinari, the company’s most experienced lobbyist, can only compound Google’s political problems.

Neal Gandhi, CEO of The Panoply, points to Microsoft’s antitrust battles in the 90s and 00s to demonstrate how a similar episode could open up Google to competition on a variety of fronts, and pave the way for a new wave of innovation:

“Alphabet and Facebook have been too busy growing like crazy to be aware of the challenges that meet companies that become dominant in their marketplace. Microsoft went through their antitrust several years ago. That episode dragged on for twenty one years and ended Microsoft’s dominance of the computing industry. It also made the spread of computing to other devices possible and opened up new possibilities like smartphones, tablets, watches and, now, the Internet of Things.”

However, he also pointed to Microsoft’s recent resurgence as evidence that tech giants can emerge from their antitrust battles as stronger, more resilient companies.

Google misfiring in ecommerce…

As a result of the EU’s ruling about Google shopping ads, the company split off its comparison service into a separate business which would need to compete against similar services in Google search results. Whether or not this ‘chinese wall’ is effective at curtailing Google’s monopoly power within its own search engine, when it comes to growing revenue from digital advertising, the intervention of competition authorities must be one of Google’s lesser concerns.

Amazon’s share of worldwide digital ad spend is rapidly eating into Google and Facebook’s duopoly. While the commerce giant still has a long way to go before catching up with the other two in absolute terms (eMarketer predicts that, by 2020, Amazon will have captured 7.0% of U.S. digital advertising, compared with Facebook’s 20.8% and Google’s 35.1%), it’s well on the way to edging them out when it comes to product searches. As early as 2016, some studies were already showing that Amazon was the first choice of consumers for shopping and product discovery, with one report putting Amazon’s share at 55% and Google’s at 28%. This is troubling for Google, since retail marketers spend more on digital advertising than those in any other industry, while consumer packaged goods is one of the fastest growth areas.

In a recent interview with The Sun Magazine, monopolies expert Stacey Mitchell explains Amazon’s ambitions:

“Amazon wants to control the underlying infrastructure of commerce. It’s becoming the place where many online shoppers go first. Even just a couple of years ago, most of us, when we wanted to buy something online, would type the desired product into a search engine. We might search for New Balance sneakers, for example, and get multiple results: sporting-goods stores, shoe stores, and, of course, Amazon. Today more than half of shoppers are skipping Google and going directly to Amazon to search for a product. This means that other companies, if they want access to those consumers, have to become sellers on Amazon. We’re moving toward a future in which buyers and sellers no longer meet in an open public market, but rather in a private arena that Amazon controls.”

Google launched Express, its own shopping fulfillment service, five years ago, in an attempt to curtail Amazon’s dominance of commerce. But having abandoned experiments with a subscription model, perishable goods and its own delivery hubs, the service has struggled to attract as much as 9% of the online traffic of its largest retail partner, Walmart. As a service operated almost entirely by contractors making use of existing merchant supply chains, Express looks unlikely to shake up logistics on anywhere near the scale of Amazon.

…and cloud computing…

There’s still room for a huge amount of growth in cloud computing: providing infrastructure as a service, platform as a service and serverless computing environments. Gartner predicted that the industry would grow 21% year on year to hit $186bn in 2018. But despite having been in the business since 2008, when it launched App Engine, the first precursor of Google Cloud Platform, Google’s business lags way behind the market leaders Amazon Web Services (with a 52% market share) and Microsoft Azure (with 13% of the market).

Given Google’s technical capabilities and already existing infrastructure investment, it’s surprising that the company not only allowed Amazon to get a four-year head start in the cloud computing business, but also let Microsoft, who entered the market two years later than Google, grow a cloud business four times as big.

At this point, cloud infrastructure is an arms race, and Amazon’s successful incursion into digital advertising just augments their already sizeable war chest, as Matt Asay of TechRepublic explains:

“As much as AWS is booming, Amazon’s advertising business grew 139% last year, helping to push Amazon’s profit up 221% year over year. In other words, for anyone hoping that AWS would somehow feel constrained to rein in spending in order to maintain profitability, it doesn’t look like it will need to. AWS can run profitably in its own right, but bolstered by a massively profitable advertising business? That’s enough to push Microsoft and Google to dig even deeper into their CapEx budgets, and likely enough to push IBM and Oracle out of the market entirely. They simply can’t keep the pace.”

Despite a $380m buyout of development platform Bebop in 2015, installing former VMWare chief exec Diane Greene as head of cloud businesses, adding many more salespeople, and ploughing 200-400% of revenues back into infrastructure, at the current rate of growth it would take Google’s cloud offering until 2030 to catch up with AWS revenues. (Microsoft Azure, at its current rate of growth, could be level with Amazon in just two years.)

And, while Google is backing out of cloud contracts with the Pentagon after employee protests, a recent request for proposals to move all of the US Department of Defense data onto the cloud (estimated contract value: $10bn), seems to have been written with Amazon in mind.

…and entertainment

Another puzzler is how Google, which acquired YouTube for $1.65 billion in 2006, has failed to make a sizeable profit from the world’s most popular video streaming platform. While Alphabet execs have been silent on the issue, the rumours from Mountain View over the past few years have consistently painted a picture of YouTube as a “roughly break-even” business. Competitors like Amazon and Netflix have been pouring billions into developing original content for their on-demand streaming services, whereas Google has largely stuck with an ad-supported, user generated content model, where content moderation and copyright enforcement consume resources at an exponentially growing rate. YouTube has attempted to launch ad-free subscription products several times, but relying on existing YouTube stars to generate premium content, instead of taking Amazon’s approach of investing in full-scale production, and internal confusion over whether YouTube should be aiming towards music or TV and movies, has resulted in meagre uptake for the latest incarnation, YouTube Premium.

Google Play Store revenues, from mobile app, music and book downloads, continue to grow by around 20% per annum. But antitrust actions, like the EU’s, which aim at ending Google’s status as the default Android marketplace provider, will create an opportunity for competitors, like Amazon, to capture a bigger share of the market.

Meanwhile, on Alphabet Street

What of the non-Google companies within Alphabet? Will any of them step in with strong revenue performance in the near future? Unlikely. They’re mostly long-term R&D efforts which have been spun out of Google to allow for more autonomy.

Calico is working on technologies to extend human lifespan. While the commercial potential of developing new drugs, genetic therapies and computational biology is huge, the company hasn’t made any headlines in its first five years. Chronicle is a fairly unremarkable cybersecurity company. Fiber, the broadband company, has proven to be incredibly capital intensive and hugely loss making. Loon’s high-altitude balloons will circumvent the need to go to the expense of laying fiber optic cables, but won’t be generating revenue anytime soon. Verily, formerly known as Google Life Sciences, has the potential to exploit huge inefficiencies in the US healthcare system, but doesn’t turn a profit yet. DeepMind, the artificial intelligence research subsidiary, generates fantastic PR through initiatives like AlphaGo, but its losses mount.

In surprise news this week, and way ahead of most people’s expectations, Waymo, Alphabet’s autonomous vehicles company, will launch the world’s first driverless car service in December. Operating under a new brand, it will theoretically compete directly with Lyft and Uber, both of which have never turned a profit due to having to pay human drivers. But getting consumers comfortable with driverless cars will require a very careful, gradual rollout (initially the service will cover about 100 square miles around Phoenix) and many years before the capital investment in Waymo is repaid.

The pivot to… AI solutions?

So what’s left for Google, given that its digital ads business is under threat from both ad blockers and competitors, and that it’s trailing behind competitors in most of the verticals to which it could pivot? The answer, surprisingly, might lie with hardware.

Google’s Home range of smart speakers may not have infiltrated as many homes as Amazon’s Echo range but, after a slow start, they’re now flying off the shelves quicker. That’s largely down to a marketing push by Google, but it’s also because Google Assistant is both smarter than Amazon’s equivalent voice assistant, Alexa, and more familiar to users due to its inclusion in the latest versions of Android. A cursory look at Google’s hardware range is enough to see that Google Assistant is built into all of them – Pixel, Home, Laptops, Tablets – and is a major selling point of these premium consumer electronics devices.

Google isn’t spending anywhere near enough on marketing and distribution to compete with Apple in the premium device market. Instead, these devices are intended to send a message – that Google’s tech, and more importantly, its AI, is the best.

Google announced its shift to becoming an “AI first” company in 2017. Google Research was subsequently renamed Google AI, and earlier this year CEO Sundar Pichai said publicly that AI is a more important development than electricity and fire. The company’s commitment to machine learning has seen them develop TensorFlow, an open-source code library to help developers create neural networks and machine learning algorithms. It’s also designed and produced two generations of Tensor Processing Units (TPUs) – chips optimised for carrying out the kinds of operations required by machine learning tasks – and is selling access to them via Google Cloud. A third generation is in development.

“For a very long time, we held back and said Intel and Nvidia are really great at building high-performance systems,” said Cliff Young, a Google engineer speaking of the company’s chip manufacturing efforts at a recent conference. “We crossed that threshold five years ago.”

Writing in ZDNet, Ovum’s Tony Baer points to Google’s best bet at remaining an indispensible company in the decades ahead:

“[A] subtle pivot could be found with Google Cloud’s initial foray into what it terms “solutions.” That’s Google speak, not for competing in the applications market or selling direct to enterprise customers, but for leveraging its AI to add smarts to partner applications. So, the first of the solutions to roll out is Contact Center AI which provides intelligence to hybrid chatbot/human call center processes by using natural conversation to streamline the “phone tree” process and then prompt live representatives with pertinent information — all based on machine learning. Google has signed up a mix of call center and systems integrators as partners for its first solution venture. There will be more.”

If Google can position itself as the foremost authority in artificial intelligence, while using its AI-driven consumer- and enterprise-facing products to bolster its ‘halo’ effect; and if it can guide and shape the development of machine learning applications, while providing a platform on which enterprises and third parties rely to deliver AI innovations; then we might be hearing the phrase ‘OK, Google’ for a long time to come.