What is driving growth in Google searches post Chat-Gpt adoption in 2023?
"Generative AI creates entirely new kinds of queries that never existed before"
Some of GenAI usage is related to this, but not all of it. Anecdotally I've definitely moved some of the searches I used to do on Google to ChatGPT, probably ~30% which is material at least for myself.
"AI Integration: Google’s rapidly rolling out generative AI features (e.g., Gemini, Search Generative Experience), aiming to remain the go-to interface. So far, advertisers continue seeing ROI on the platform. No exodus."
Again, from my own experience, I've found Google's integration of AI to be clumsy and unwieldy, and Gemini's AI results tend to hallucinate a lot more than ChatGPT or Claude.
Google Cloud is growing faster than AWS right now, and 35–40% margins are standard for scaled cloud players — we’re just applying peer comps. The $10B EBIT for YouTube is based on comps like Netflix, adjusted for its ad-heavy model and minimal content spend. Play Store’s $25B EBIT isn’t a guess — it’s a conservative extrapolation from 2019 court docs showing 62% EBIT margins, discounted to ~53–55% on $47B in revenue today.
As for Search — a 10–12x multiple is way too low for something this dominant, high-margin, adaptable and still growing. We’re thinking closer to 20x — and if it’s currently being valued at 5x, that’s a 4x re-rate on the core business.
Hope you don't mind the pushback by the way. A couple counterpoints:
What if lower margins (ie lower prices) are the reason that Google cloud is growing faster than AWS though? Seems a little optimistic to stick a higher-than-market multiple on profits that it isn't even making yet.
Regarding the play store - I had a look at those court documents, and they list the revenue in 2019 as $11.2b (and operating profit as $7b), to get that 62% figure. However if I look at the same sources that give $47b for 2024, they give about $30b back in 2019. So I think there's a problem here between gross and net revenue (since most of the gross revenue goes to the developer) and I wonder if this is causing you to significantly overestimate the play store's EBIT?
On Play store - The $47B could be gross, and if that’s the case, EBIT would be lower than we initially estimated (if we say net revenue is 14-15B, then EBIT may be closer to 8-9B for that segment. Since Google doesn’t disclose the breakdowns, we are indeed estimating.
That said, total EBIT stays the same — it just shifts where the earnings are showing up. And if more of it’s coming from Search, a re-rate there would have an even bigger impact - i.e moving 16B of EBIT from Play to Search.
On Cloud — fair point on margins, but we’re not too concerned. AWS followed a similar path early on, and Google just posted strong growth despite capacity constraints.
I'm trying to work out how you've gotten to the implied EBIT in the post. I'm seeing 112 for 2024. If I add the numbers in your post (and knock off half from google cloud to get down to the actual number) I still end up at about 120 - where am I going wrong?
I think what my gripe with this post comes down to is the way it makes some less-than-conservative (at least in my view) assumptions about all the other segments in order to back out a valuation that is clearly absurd for the search business, then go "look how cheap the market is valuing search". But I think it's pretty obvious that the market isn't valuing search at 5x EBIT - it just doesn't make as optimistic assumptions as you do on the other businesses.
I think what's really happening is the market thinks cloud is worth say 40x current EBIT versus your 60x; market thinks youtube is worth 20x versus your 30x; and search is worth say 10x instead of your 20x. Which is completely fine, I don't think your valuations are wrong (I don't have an opinion either way) but I think it makes your disagreement with the market slightly more marginal and less obvious who is right.
I hope this didn't come across as lecturey, I don't intend it to (but aware sometimes when I read things back they do)
Really appreciate the thoughtful pushback, Matt — this is why we post.
Re: total EBIT — 2024A was $112B, and the 2025 consensus estimate is $127B. We should’ve been clearer in the table that we’re looking at forward guided EBIT of $127B.
The market may well be valuing those segments differently — so your point is heard.
For us, this is a long-term hold, and the goal is to be directionally right. As Buffett put it, you don’t need to know a man’s exact weight to see he’s fat— and to us, Search looks like a business trading well below what it’s worth.
Even if it’s at 10x today (on $96B of EBIT), a re-rate closer to 20x would imply ~50% upside on that piece alone.
I think you underestimate the second level effects of gen-ai.
What is driving growth in Google searches post Chat-Gpt adoption in 2023?
"Generative AI creates entirely new kinds of queries that never existed before"
Some of GenAI usage is related to this, but not all of it. Anecdotally I've definitely moved some of the searches I used to do on Google to ChatGPT, probably ~30% which is material at least for myself.
"AI Integration: Google’s rapidly rolling out generative AI features (e.g., Gemini, Search Generative Experience), aiming to remain the go-to interface. So far, advertisers continue seeing ROI on the platform. No exodus."
Again, from my own experience, I've found Google's integration of AI to be clumsy and unwieldy, and Gemini's AI results tend to hallucinate a lot more than ChatGPT or Claude.
Why should google cloud be worth 30x+ when AWS is only woth 20-30x by your estimation?
Where does the 35% margin come from?
Where does $10b EBIT for youtube come from? I thought they didn't disclose these numbers.
Is $25b EBIT for play store a guess or a reported number?
Seems like a pretty optimistic set of assumptions you have to make to get search to a 5-6x EBIT implied valuation.
Then even if that valuation for search doubles, you only add about 25% onto the total valuation.
Google Cloud is growing faster than AWS right now, and 35–40% margins are standard for scaled cloud players — we’re just applying peer comps. The $10B EBIT for YouTube is based on comps like Netflix, adjusted for its ad-heavy model and minimal content spend. Play Store’s $25B EBIT isn’t a guess — it’s a conservative extrapolation from 2019 court docs showing 62% EBIT margins, discounted to ~53–55% on $47B in revenue today.
As for Search — a 10–12x multiple is way too low for something this dominant, high-margin, adaptable and still growing. We’re thinking closer to 20x — and if it’s currently being valued at 5x, that’s a 4x re-rate on the core business.
Hope you don't mind the pushback by the way. A couple counterpoints:
What if lower margins (ie lower prices) are the reason that Google cloud is growing faster than AWS though? Seems a little optimistic to stick a higher-than-market multiple on profits that it isn't even making yet.
Regarding the play store - I had a look at those court documents, and they list the revenue in 2019 as $11.2b (and operating profit as $7b), to get that 62% figure. However if I look at the same sources that give $47b for 2024, they give about $30b back in 2019. So I think there's a problem here between gross and net revenue (since most of the gross revenue goes to the developer) and I wonder if this is causing you to significantly overestimate the play store's EBIT?
On Play store - The $47B could be gross, and if that’s the case, EBIT would be lower than we initially estimated (if we say net revenue is 14-15B, then EBIT may be closer to 8-9B for that segment. Since Google doesn’t disclose the breakdowns, we are indeed estimating.
That said, total EBIT stays the same — it just shifts where the earnings are showing up. And if more of it’s coming from Search, a re-rate there would have an even bigger impact - i.e moving 16B of EBIT from Play to Search.
On Cloud — fair point on margins, but we’re not too concerned. AWS followed a similar path early on, and Google just posted strong growth despite capacity constraints.
We welcome the thoughts and feedback!
I'm trying to work out how you've gotten to the implied EBIT in the post. I'm seeing 112 for 2024. If I add the numbers in your post (and knock off half from google cloud to get down to the actual number) I still end up at about 120 - where am I going wrong?
I think what my gripe with this post comes down to is the way it makes some less-than-conservative (at least in my view) assumptions about all the other segments in order to back out a valuation that is clearly absurd for the search business, then go "look how cheap the market is valuing search". But I think it's pretty obvious that the market isn't valuing search at 5x EBIT - it just doesn't make as optimistic assumptions as you do on the other businesses.
I think what's really happening is the market thinks cloud is worth say 40x current EBIT versus your 60x; market thinks youtube is worth 20x versus your 30x; and search is worth say 10x instead of your 20x. Which is completely fine, I don't think your valuations are wrong (I don't have an opinion either way) but I think it makes your disagreement with the market slightly more marginal and less obvious who is right.
I hope this didn't come across as lecturey, I don't intend it to (but aware sometimes when I read things back they do)
Appreciate the discussion.
Really appreciate the thoughtful pushback, Matt — this is why we post.
Re: total EBIT — 2024A was $112B, and the 2025 consensus estimate is $127B. We should’ve been clearer in the table that we’re looking at forward guided EBIT of $127B.
The market may well be valuing those segments differently — so your point is heard.
For us, this is a long-term hold, and the goal is to be directionally right. As Buffett put it, you don’t need to know a man’s exact weight to see he’s fat— and to us, Search looks like a business trading well below what it’s worth.
Even if it’s at 10x today (on $96B of EBIT), a re-rate closer to 20x would imply ~50% upside on that piece alone.
Appreciate the back and forth - thanks again.