Alphabet is Literally Worth Less Than the Sum of Its Parts

Is Google going to break itself up? David Streitfeld posed the question earlier this week in a column for The New York Times, riffing on a research report by Gil Luria for the investment firm D.A. Davidson that made the case. To answer the question: no, Google is not going to break itself up. But that doesn't mean there aren't a few compelling ideas buried in the provocation...

Actually, Streitfeld and Luria are sort of arguing for different things – or at least, different reasons for the same proposed outcome. Streitfeld wonders if Google might be wise to preemptively break themselves up before the government does it for them. Sort of like AT&T back in the day. Luria thinks Google should break itself up because it's become a conglomerate, in the worst sense of the word. That they're big and bloated and would execute better, or unlock more value, at least for shareholders, as a more nimble company – or series of companies, as it were.

Again, it's not going to happen. And that's because what Luria views as a weakness and the government views as a problem, Google's size and scale and scope, the company views as a strength. Are there downsides to it? Sure. But right now, as we're entering the age of AI, Google will need to leverage every asset at their disposal to ensure the company can live well beyond the days of web search. And it's starting to work.

Further, Google already sort of did what Luria is suggesting when they formed Alphabet. We're coming up on the 10 year anniversary of that move (I happened to be there when it happened, working at one of the subsidiaries, GV, which became an "Other Bet", though I had zero vision into the rationale or decision making.) Granted, it perhaps hasn't fully unlocked the full value of all the pieces (as outlined below), but it at least makes them somewhat easier to see as separate entities. Well, except for YouTube, which should probably be its own "Bet" but Google has clearly avoided that, even though they now have to directly report numbers around the property given its size.

Speaking of, the world is now fully awake to what a behemoth YouTube is – not just in user generated content, in terms of all time spent on anything. UGC, music, Hollywood content, podcasting, etc. Meanwhile, Waymo – one of the "other bets" spun out from "X", yet another "other bet" – is clearly hitting an inflection point in some of its early markets, rising further, faster than yesterday's ride-sharing services did.

Other parts of Google are not getting their due. If Waymo were publicly traded, Mr. Luria argued, investors might give it something closer to Tesla’s $1 trillion valuation, especially since Tesla’s self-driving cab ambitions are little more than a concept at this point. The same goes for YouTube when compared with its rival Netflix, a Wall Street darling.

Mr. Luria estimated that all the parts of Google could separately be worth more than $3.7 trillion, or nearly double the company’s valuation now. “Investors want a big-bang breakup, not isolated spinoffs,” he wrote.

But we don't really need to speculate on the valuation Waymo would command because they were valued by investors just about nine months ago. It wasn't $1T – or anything close to that. It was $45B. Granted, the round was led by Alphabet, but again, plenty of other savvy investors were involved. And yes, it does seem like they got a hell of a deal. Given where AI startups are currently being valued – and yes, where Tesla is being valued on the promise of much of what Waymo is already delivering, it's fair to think they could get a much higher valuation. But let's just say $500B, just to be safe.

YouTube is tricky versus Netflix because they have almost the opposite models. Yes, they're converging with Netflix increasingly doing advertising and YouTube selling subscriptions, but they're not really the same businesses right now other than they're both going after your eyeballs. Still, if we gave it a similar revenue multiple – and by "we", I mean ChatGPT – we get roughly a $300B valuation.

So just with those two parts of Alphabet/Google, we have $800B+ in value.

But what about Gemini? You could certainly argue that it is now hitting its stride at just the right time. And its already so woven in between all the various parts of the company that you couldn't really break it off/spin it out. But let's just say, for the sake of argument, that you could have DeepMind running as a separate company again. How much might that be worth? If OpenAI stands at $300B with a product in ChatGPT with more consumer users than anyone else in AI, it might be reasonable to value DeepMind just below that line, maybe $250B?

Honestly, probably more. Potentially a lot more. But I'm being conservative here.

And in my conservatism, we've already shot past $1T.

Google Cloud is also not a separate entity, but it's thought to be around 15% of the parent company's overall revenue as it approaches a $50B business. Given the market comps, would probably have an outside valuation of around $250B.

I've previously tried to run some numbers on what Chrome is worth internally to Google (even though it doesn't directly make revenue, it powers a lot of Search and, importantly, makes it so Google doesn't have to pay as much to other browser makers in terms of revenue share). And, of course, it's what the government is going directly after. It's probably worth another $100B.

Android is even harder to measure given the way Google monetizes it. But stand-alone, maybe with Play, it would probably be at least another $100B business. Probably more, but again, I want to be conservative.

So now we're cleanly at $1.5T. And that doesn't include many other "Bets" and businesses – which are smaller and bucketed together as yes, the "Other Bets". Alphabet itself is valued at almost exactly $2T right now. So if you want to consider the above estimates to be at all accurate, that means the market is valuing Google Search (and Advertising) – the crown jewel – at just $500B. Now obviously that's not true. What the market is really doing is valuing Search & Advertising as most of the $2T, and selling nearly everything else short.

Even if you only valued Search and Advertising as a $1T business (which would be around 4x revenue), we still have roughly $500B in value that could be unlocked if the pieces were being "fairly" valued by the market – and thus, the argument to separate them from the whole. There are only around 30 companies in the world worth that amount in terms of market cap. Entire companies. Google seemingly has that hidden in their couch cushions.

So that's probably the strongest case to be made for a break-up. But again, they wouldn't do that. Certainly not to get the DoJ off their backs and certainly not right now, with everything going on in AI. Instead, perhaps there's an argument to be made that they should be more cleanly separated under the Alphabet umbrella to better unlock at least some of that extra value.

There too, there are undoubtedly good reasons to like that there are more levels of obfuscation when an entity isn't operating wholly separate, namely on the reporting front. So instead, Google – Alphabet – just has to suck it up and bear it with a valuation that should probably be a lot closer to the "Big 3" – Microsoft, NVIDIA, and Apple – than to Meta, a company which is still something like 97% driven by their feed-based advertising (albeit now split across Facebook and Instagram) and is desperately searching for that thing after the thing.

Google used to get the same bad rap – that it was a one-trick pony. It's now a many-trick pony, albeit one that is still sort of valued like a one-trick pony.