Felix Salmon

Jon Stokes has a fascinating column making a credible case that the VC and tech bubble is hampering development of the cloud.

I recently had a sit-down chat with Ping Li, a venture capitalist at Accel Partners who does investments across the layers of the cloud stack… he explained that the talent shortage is stifling fundamental innovation in the cloud space.

To do really fundamental engineering innovation of the kind that was done, say, in the early days of Google and VMware, you need to hire and retain teams of talented engineers. But in today’s go-go funding environment, top engineers are being enticed with truckloads of money to break off and form two- and three-person startups. This phenomenon, explains Li, is why “many of the really big innovations happen in less frothy times.” He did go on to clarify that “some great companies do get created in these times (like Amazon in the last bubble). It’s just harder given talent shortage.”

I asked Jon if this meant that real innovation in the cloud requires pretty big teams, and can’t be done by smaller startups. And the answer there is absolutely yes — if you’re looking for something huge like Amazon’s AWS, which required the full focus of Amazon’s large technical staff over a multi-year timeframe. Scalable websites can, thanks to Amazon’s cloud, now be launched with a handful of employees. But to develop the cloud itself takes serious resources — to the point at which it’s now conventional wisdom that you need to be Amazon, Apple, Google, or Microsoft to even play in that game. And even they’re starving for talent.

There’s the email I got a few months ago from a friend of mine and product manager at Apple, who was wondering if I knew any cloud computing hackers that they could hire. When we get to the point where Apple product managers on the client side are reaching out to their personal networks in search of cloud coding talent for the world’s largest tech company, you know it’s bad out there.

Jon frames the problem as one of supply and demand:

The current crisis in the cloud is the product of too many dollars and transistors chasing too few coders and sysadmins. It will take a while for the latter to catch up with the former… unless, of course, another major downturn strikes. It seems ironic that less money could equal more innovation, but it wouldn’t be the first time that a wave of downsizing and tight money boosted productivity.

I asked him whether looser skilled-immigration policies might help, and he said probably not:

I think that the root problem isn’t one of geography–it’s that this stuff is happening so fast (i.e. Moore’s Law and my cheap transistors argument) that the hardware build-out is outstripping programmer education. And by “programmer education” I mean not only the number of programmers being trained in aggregate across the world, but also programming as a discipline’s ability to empower ordinary mortal to develop and deploy software on these massively parallel systems. The cloud has to be “de-ninjafied”, so to speak. Getting max productivity out of the cloud has to be brought within the grasp of non-ninjas, the way that, say, VisualBasic from MSFT brought building a relatively complex custom relational database application within the grasp of the average local technical college graduate.

This rings true to me. The cloud is not located in any particular country, and if there were great engineers who could be hired to work on it from Beijing or Bangalore, I’m sure that Apple and Amazon are more than capable of doing that. What needs to be done here is basically cloud-development grunt work: taking a young and complex technology, and building the tools which can bring it to the masses. There’s not a lot of glory in that — while companies which live in the cloud, like Airbnb, can find themselves worth billions, the engineers who work on the cloud are more like the utility workers of the internet. And it’s easy to see why they might be finding more attractive opportunities, right now, elsewhere.

Here’s how Jon puts it:

In order to move the cloud itself forward in a major way by solving large batches of related fundamental technical problems you need longer timeframes. You can fiddle around in the guts of the cloud, smoothing out this and optimizing that, and adding features and bells and whistles. But to do the big projects, you need time.

Now, there are shorter-term innovations that can and will get done in the cloud, so VCs have plenty to fund. But to shift the tectonic plates, you need time and resources.

This isn’t like sustainable energy — it’s not something that the government can or should be stepping in to fund. More money pouring into the tech space would only exacerbate the current problems.

There’s a case to be made that AWS is the result of what happened when Amazon, after the dot-com bust, found itself with an unusual degree of access to the time and talent of a large number of engineers. The cloud is young; it could do with a lot more development along those lines. But as Jon says, we’re unlikely to see such fundamental evolution in cloud architecture for a while. Because for the time being, smaller, lighter, and riskier projects look much more attractive.

Nick Denton lost his bet with Rex Sorgatz by the narrowest of margins — just 10 million pageviews, or alternatively just four days. But he handed over a check for $100 with a smile, and even threw us a huge party into the bargain! Hence, obvs, my not-entirely-sober status by the time we’d waited for Lockhart Steele to finish his eleven-course dinner and make his way up to the Gawker Media rooftop.

Here’s to next year — when Denton will have to achieve 700 million pageviews in order to avoid writing a second check. I promise to film the outcome, whoever the winner might be.

I’m in Sofia today, where I gave a talk on Groupon at the DigitalK conference. This post isn’t the speech that I gave, which was much shorter and more conversational; the slides I used are here.pdf. There’s not much new in this post, for those who have been following what I’ve written on Groupon over the past few months; I basically wrote it to get a feel for how I wanted my speech to flow. But here you go anyway.

It’s almost universally known, among people who live or work anywhere near the intersection of technology and finance, that Groupon is the fastest-growing company the world has ever seen. Technology companies are often fast-growing, of course, but Groupon’s growth rate is astonishing even by tech standards. Check out this chart:

Starting at zero, Groupon got within shouting distance of $1 billion in revenues within a single year. It took Zynga two years to get to that point, it took Amazon three years, and it took Facebook four years. eBay hadn’t even got there after five years. This isn’t entirely or even mostly a function of Groupon’s business model; much more important is the massively increased willingness of people to buy things online now than when the likes of eBay, Yahoo, and Amazon launched in the 1990s.

And it’s possible to quibble over terminology here, too: in its latest filing, Groupon now calls this number “billings”, with “revenues” being about half of what we see here. But whatever you call it, it’s a monster stream of cash which is flowing into the company, and you can add to these revenues some $1.1 billion in new equity capital, which is also helping to fuel expansion. Groupon isn’t just growing fast: it’s also raising money at a rate that no other company has ever dreamed of.

Importantly, the stream of cash flowing out of the company is even bigger. Half of Groupon’s billings go to merchants, usually small local businesses. Much of the rest goes towards Groupon’s rapidly-growing payroll, and to fund expansion into new cities and countries. And then there’s more than $900 million which has been used to cash out early investors in the company, including CEO Andrew Mason — a man who is now extremely wealthy even if Groupon stock goes to zero tomorrow.

Groupon is a very innovative company, and this is one of its most important innovations — the idea that the founder can and even should be able to cash out to the tune of millions of dollars very early on in the company’s lifecycle, while it is still raising new VC funds.

The argument here makes a certain amount of theoretical sense. VC investors are looking for home runs, and they’re willing to see a reasonably large percentage of their portfolio investments fail to achieve that end. Essentially, they want the CEOs they’re backing to take on as much risk as possible.

But there’s a problem with this model: CEOs are human, and humans are naturally risk-averse. When Andrew Mason first saw that he’d built Groupon into an inherently highly-profitable Chicago company, he could have decided to fund further expansion only out of the company’s profits, while keeping some portion of those profits for himself and his investors. Groupon would have grown at a much more normal pace, and would certainly never have generated eye-popping charts like this one.

Over the course of one year, from the first quarter of 2010 to the first quarter of 2011, Groupon’s subscriber base increased 24-fold; its revenue increased 14-fold; its sales rate increased 15-fold; and it swung from a profit of $8 million to a loss of $146 million.

These are the kind of figures which make eyes go wide — with greed, if you’re a VC, and with fear, if you’re an businessman trying to build a company which can deliver a reliable long-term profit stream. By cashing out a significant portion of the CEO’s stock, his backers essentially turned him from businessman to VC — they aligned his incentives with theirs. He won’t want for money ever again, so he’s no longer the type of person who would look at a company making $8 million a quarter and think that was pretty good. Instead, he wants to take risks, grow at a breakneck pace, and create a company which is likely to go public, later this year, at a valuation somewhere in the neighborhood of $15 billion. He wants to change the world.

That’s the idea, anyway; we’ll see how it works out. Historically, VC rounds have been about providing capital to companies which need it; in Groupon’s case, they’re more about finding a way to cash out early investors. And so a lot of people who own Groupon stock today didn’t really put money into the company, so much as they simply bought pre-IPO stock on the secondary market. If they end up making a fortune in the IPO, then other companies will certainly start looking at the Groupon model as something maybe worth emulating.

What’s sure, however, is that the kind of growth and ambition exhibited by Groupon is catnip to journalists looking to puncture something which looks very much like a bubble. Going public before you’ve achieved sustainable profitability? Using seemingly made-up measures like Adjusted Consolidated Segment Operating Income instead of generally-accepted accounting principles? Becoming a billionaire before the age of 30, while refusing to play according to the spoken and unspoken rules of both Wall Street and Fleet Street? It’s a recipe for getting the press to turn on you.

But in fact the conventional wisdom on Groupon is narrow-minded, a little bit silly, and largely based on journalists kidding themselves that “everybody” thinks Groupon is a huge success, and that therefore it falls to them to debunk the myth. In reality, the huge-success meme was extremely short-lived, and stems largely from the fact that Google attempted to buy Groupon for $6 billion at the end of 2010. Ever since Groupon turned Google down, there’s been a steady drip of stories saying that they were idiotic to do so and that valuations of Groupon in the $15 billion to $25 billion range are utterly ridiculous.

Now, I’m not going to take a position on how much Groupon is worth; I’m neither an investment banker nor an equity analyst. But what I’d like to do is run down a few reasons why a stratospheric valuation could conceivably be justified, and then look at a few of the potential potholes which face Groupon in its attempt to justify that valuation.

First of all, Groupon has cracked local, in a way that pretty much nobody else has been able to do. We spend most of our disposable income at merchants located within easy striking distance of where we live — but until Groupon came along, those merchants had no good way to reach us online. Everybody’s interested in what’s going on locally, and Groupon worked out that a steady stream of daily emails, each one touting a great local deal, would be hugely attractive to millions of people. This is advertising you want to get.

Second, Groupon has created advertising that is guaranteed to work. By setting a minimum number of people who need to sign up for a deal before it’s activated, merchants can be sure that the needle will be moved and their effort won’t be wasted. This is something of a holy grail in advertising and marketing circles, and it absolutely helps to explain Groupon’s spectacular growth. Merchants hate to spend money on marketing because they fear they’re being swindled by fast-talking sales reps. With Groupon, they know exactly what they’re signing up for, and they won’t end up spending huge amounts of money on nothing.

Indeed, those merchants are not spending any money at all: they’re being paid. This is another great Groupon innovation: create a form of advertising which merchants not only pay nothing for up front, but which they actually get paid. Yes, there’s a cost to providing their goods or services, and in many cases that cost is greater than the amount of money they’re getting from Groupon. Merchants who get too greedy for a big up-front paycheck can end up ruining themselves when those coupons get redeemed. But anybody who’s ever run a small business knows that the promise of money in hand is always going to be incredibly attractive when compared to advertising or marketing which has to be paid for.

In fact, Groupon gives advertising away for free. It has an astonishingly valuable email list, and many merchants would pay good money to be able to send out wittily-written ads to local Groupon subscribers. But they don’t need to do that. Groupon makes its money from the tiny minority of customers who actually pay for a deal. But that leaves millions of people every day who read ad copy which is targeted directly at them. That targeted advertising is extremely valuable, and Groupon isn’t charging a penny for it. Because it has an alternative source of income, Groupon doesn’t need to charge merchants for the privilege of being included in its emails. And so a merchant who values that exposure is well ahead of the game as soon as the email goes out.

But another Groupon innovation goes one further than that — it’s the Groupon commitment device. A commitment device is the way the people force themselves to do something which they know they want to do, for fear that for some reason or other human weakness might otherwise mean they wouldn’t do it. The classic commitment device is marriage: it helps people stay together when otherwise they might drift apart. A mortgage is also a commitment device, which forces you to spend a large sum of money every month slowly building equity in your home, until after 30 years you own it outright. A Groupon, of course, is nowhere near as important as marriage or a mortgage. But it has a similar effect. I see a Groupon in my email — let’s say it gives me $50 off a meal at a restaurant I’ve been meaning to try down the street. By buying the Groupon with a click of my mouse, I force myself to go to that restaurant — something I might well never have got around to, otherwise.

Groupon forces its customers to buy its products using something which isn’t very innovative at all — the hurry-it-won’t-last-long sales pitch. By making sure that offers can only be bought for a day or two, Groupon forces people to make a decision now as to whether they want to do this thing. And that non-innovative part of the Groupon model is one of its big potential weaknesses, as I’ll come to in a minute. But first of all there are some potential strengths to Groupon.

What we’ve seen up until now is the way that Groupon has worked and grown to date. But looking forwards, optimists see lots of other great promise in the company. I don’t want to dwell on these, because forecasting the future of any tech company is always a mug’s game. Some of them are likely to work, others will probably fail. There’s Groupon Now, and the mobile applications which are nascent but growing fast. There’s the move from services into products. There’s the Getaways travel product. And there’s targeting — the crucial way in which Groupon promises to be able to target customers according to their purchasing preferences and a myriad of other factors, rather than just going on what city they live in. It hasn’t happened yet, but I suspect that over the medium term, Groupon will succeed or fail based on whether it manages to crack the targeting nut. Having a huge subscriber base is a necessary condition for targeting, but it’s far from sufficient.

I have no idea what any of these businesses might be worth, or what kind of probability to apply to them succeeding. But looking down this list, and looking at the kind of money which Groupon is bringing in without these businesses, it’s definitely possible to see how this could be a $20 billion company, potentially.

But.

There are also risks to the Groupon model, and we’re already seeing some of them materialize.

One risk is that merchants stop wanting to play ball. Maybe they move their business to to a Groupon clone which offers them a bigger cut of the proceeds. Maybe they don’t return to Groupon because it turns out that too many Groupon buyers are only coming because they bought the Groupon, and don’t become valuable repeat customers. Or, conversely, maybe it turns out that too many Groupon buyers are people who would have come anyway, and so the merchant is simply taking a big haircut on their normal revenues. Or perhaps — as we’ve seen happen a few times, according to press anecdote — merchants simply get overwhelmed by Groupon traffic, and thereby alienate their existing customers.

If you look at the established market of Boston, the trend here is not good. As markets mature, they won’t be as white-hot as they were in their youth. But one big problem for investors is that none of them really have a clue what kind of revenue per merchant is necessary for profitability.

The other big risk is that consumers stop wanting to play ball. The novelty wears off, they find too many Groupons sitting unused in their desk drawer, they get burned one too many times by a deal which seemed really good in the email but which turned out in real life to be disappointing.

Already we’re seeing signs that this is happening: according to a guy called Sam Hamadeh, Groupon’s revenue per customer has fallen from $15 per month to $3 per month. Now the number of customers is still growing fast, but clearly profits are going to be hurt if those customers don’t spend nearly as much as they used to. Here are the numbers for Boston, again:

And there are other risks, too, including big possible legal risks. There are lots of laws governing coupons, in all 50 states and around the world, and Groupon seems to be happily violating dozens of them, while doing its utmost to fob legal responsibility off onto merchants who can’t possibly know what the law says.

But the biggest risk of all, which pretty much encompasses all of the other ones, is simply that Groupon will develop a bad reputation. If people don’t trust Groupon, then it’s all over.

In the beginning, Groupon got away with a lot, thanks partly to how innovative it was and partly because of the jocular tone to its emails. But at this point everybody knows what the model is, and the humor is hardy surprising any more. And increasingly consumers and merchants are asking just how good Groupon’s deals are, really. Not in terms of save $X if you spend $Y, but in terms of the intrinsic quality of the merchants being featured.

It seems to me that if Groupon wants the Yipit charts to start going up and to the right, if it wants to delight consumers, and if it doesn’t want to become shorthand for desperate-and-crappy merchants resorting to a last-ditch effort to get people into their otherwise-empty stores, then it’s going to have to start imposing more editorial control over its sales team.

In the long run, people will buy coupons only from those sites they trust to send them to great merchants. Groupon has first-mover advantage, but consumers are fickle, and will happily switch their allegiance to smaller companies they think are cooler, if Groupon makes them feel a bit unwashed every time they buy into an offer.

How can Groupon ensure that it features only merchants its email list will love? I haven’t a clue. But that’s the single biggest task facing the company. If it wins at that, it’ll be fine. If it fails, I fear it will slowly wither away.

BankSimple came out with a preview of their service today and it’s very cool.

A couple of hours after watching this video, I got an email from Citibank — addressed to “Dear Feliz Salmon” — with the subject line “Announcing the new Citibank Online”.

Citibank has a video too, but it’s not embeddable. After looking at it, I have to say that the new Citibank Online looks much like the old Citibank Online: I don’t think there’s anything for BankSimple to worry about here.

The most interesting bit, for me, was the huge difference in the way in which the two banks put their videos together. BankSimple got its CEO to walk us through his bank account, as it appears on his website. Citi, by contrast, showed us an impossible account which looks like this:

“Jim Smith”, here, has $54,662.00 in his checking account. Of that, just $23,612 is “Available Now”. Maybe one of those 25 unread messages might tell him what seems to be the problem with the other $31,050. But he’s unlikely to read them: after all, his last login was on May 14 and it’s now sometime in the fall: his credit-card payment is due November 10, even though his Expense Analysis stops at June 5.

But my favorite bit of this screenshot is the breakdown in the expense analysis. Apparently Mr Smith withdrew $68,040 from ATMs in the past five months — that’s $13,608 per month, or about $450 per day, every day. All that cash is 57% of Smith’s total expenditures, which means that those expenditures add up to $119,368 in all, or about $286,484 per year.

Except: we’re also told that travel expenses of $9,948 are 13% of the total, which implies an expenditure rate of about $183,655 per year.

And then there are the wonderful “Other” and “Uncategorized” expenses. (The difference between the two, of course, is unexplained.) “Other” is $3,000 and 14% of the total, implying $51,428 of expenses per year.

But “Uncategorized”, at 16% of the total, is — get this — $299 million. Which means that Mr Smith seems to be on track to spend roughly $4.485 billion this year, in total.

Citi obviously poured much more money into its video than BankSimple did into theirs. It has high production values, even unto graphs which literally come out of your computer:

Clever trick, that — although customers expecting to see it in real life are probably risking disappointment. Again, the numbers aren’t internally consistent: the top expenditure category on the left is Groceries, which is just $55.50, or 2.3%, when broken down in the list on the right. Other big expenses on the left, like Business and General Merchandise, don’t even appear in that list. And Mr Smith seems to have gone on a serious diet: his total expenses from January 1 through July 20 now total just $2,411.22.

It’s literally inconceivable to me that anybody, watching these two videos, would get remotely excited about the Citibank product. Meanwhile, BankSimple already has a long waiting list of people desperate to switch over — a list which is likely only to get even longer now that glimpses of the product are being made public.

The first BankSimple accounts will start being opened in “a few weeks”, according to today’s post. They will be glitchy, as any 1.0 product always is. But I’m quite certain that BankSimple’s early customers will be patient and forgiving and that BankSimple’s executives and employees will be accessible, helpful and responsive whenever there’s a problem.

BankSimple will never release a screenshot or video of one of its products where there’s a line item showing $299 million in personal “Uncategorized” expenses. I know this despite the fact that they haven’t even launched yet, because that simply isn’t who they are. Citi, by contrast, is rudderless, spending huge sums of money putting together silly videos advertising its website and sending out 346 million card offers to North American customers per quarter. It won’t change the way it does banking. But I can.

Christopher Mims makes a really good point:

It makes no sense that writers like Felix Salmon, who is generally excellent on just about everything, describe Netfilx, even pre-split Netflix, as an inexpensive alternative to cable. It’s not. It’s only inexpensive if you take fast broadband at home for granted — you know, like every tech pundit and journalist on the planet.

To be fair, it’s a mistake all of those pundits makes regularly — the conflation of their own situation with that of the wider public. But only one in three Americans pays for broadband, which means that something like two-thirds of the population has access to it. That’s not bad (it’s not great either – it puts us something like 27th in world broadband penetration) and it leaves out precisely the people who are being left behind by both our economy and the digital divide.

I moved to the US before the rollout of the cable modem, and for me it was a game-changer: within a few months of its arrival, sometime in the late 90s, I switched from cable-and-no-broadband to broadband-and-no-cable. I was one of the earliest cord-cutters, long before YouTube or Netflix or any real video content on the web which I had any desire to watch. I didn’t want to watch TV on my computer: I just preferred content online to the content on the TV.

Now, over a decade later, it’s possible to look at the population more broadly, and see how their preferences have revealed themselves. And Mims is right: if you have a cable line coming into your home, you’re much more likely to have cable-and-no-broadband than you are to have broadband-and-no-cable. Cord-cutting was a privileged, yuppie behavior when I did it in the 90s, and it remains a privileged yuppie behavior today.* Sure, I like having an extra $100 in my wallet every month due to the fact that I don’t have cable. But I could easily afford it if I wanted it — the fact is that I stopped watching cable long before I cut that cord.

For the time being, the price of broadband — largely set by cable companies — is being set high enough that cable-but-no-broadband subscribers are not switching to broadband-but-no-cable. In order to cut the cord, it seems, you need broadband first: you need cable and broadband, and then you need to come to the decision that you can do without the cable bit.

So, yes, let’s slow down on visions of free or cheap online services supplanting cable for America’s poor. Because Mims is right: broadband is not free. And the cost of Netflix is therefore comparable to the cost of cable — with no live-TV services at all, and in general a much narrower selection of things to watch. At some point, I’m convinced that IP-based video will indeed replace cable. But in order for it to do so, the cost of broadband is going to have to come down. And that doesn’t look as though it’s going to happen any time soon.

*Update: What I mean here is that the behavior is displayed by privileged yuppies, not that it’s inherently yuppie. The kind of people I’m talking about are people who, given the choice between a lean-forward activity and a lean-back activity, tend to choose the former. Either because they prefer surfing the internet to channel surfing, or else because they have work to do online. Those people tend to be part of the employed-and-educated middle classes.

Our fabulous social media guru, Anthony DeRosa, doesn’t use Groupon, and neither do I, and neither do any of the people in our social circles, that we know of. Now we’re guys, while Groupon skews female. And most likely we do know people who use Groupon; we just don’t know who they are. But the fact is that at heart it’s pretty uncool. That’s fine — many hugely successful companies are uncool and based on saving people money, up to and including Walmart. But here’s the problem: Groupon can’t afford to be uncool just yet, because it needs to do one last big capital-raising round at a high valuation in order to get the cash it’s going to burn through in the coming year or two.

Henry Blodget has some smart analysis today, concluding that “if Groupon raises a boatload of money in an IPO, the company will be able to keep spending aggressively on marketing and not have to worry about running out of money or dealing with slower growth for a while.” So it’s important that Groupon is able to tell its high-growth, high-intrinsic-profitability story at least through its IPO.

But Groupon’s web traffic looks like it might be falling, and Connie Loizos has been talking to analysts, including PrivCo’s Sam Hamadeh, who increasingly, don’t buy it:

Groupon’s model simply doesn’t make sense, say the number crunchers. While the company’s early success was premised on customers spending an average of $15 per month — and being affordable to acquire — these days customers cost Groupon in the double-digit dollars to acquire, says Hamadeh, and they’re spending closer to $3 a month, with “every indication” that even that figure is declining, says Hamadeh.

The monthly spend per customer is a key number to look at. There doesn’t seem to be any doubt that it’s going down; the big question is whether it’s going to level off with Groupon becoming a big and sustainably profitable business, or whether it’s just going to approach zero.

Or, to put it another way, can Groupon make the transition from being a fun fad to being a basic part of the way people spend money on a monthly basis? I think it can. But in order to do that, it’s going to have to concentrate increasingly on targeting and on the quality of the merchants it chooses to feature.