Great post. I think the problem with most is they use the "a stock is cheap" argument without taking into account the timeframe of ownership. Google I believe is a buy, but that's from me using a DCF for the next 5 years. The longer you project, the less attractive it is, of course. But time matters.
Hey Jacob, thanks. So the longer your forecast period, the more attractive an investment looks, not less. The explicit forecast period captures all cash flows where ROIIC is above WACC.
A shorter timeframe of just 5 years should make Google look even less attractive, because a lot of Google's value is captured beyond that period. What are your assumptions for your 5-year DCF?
It can go both ways. A longer forecast can artificially make a stock look less attractive, especially if most of the growth is near-term and it slows with time. Like with Google and cloud now. (Like most mature businesses)
I agree with you though. You’re absolutely right, a lot of Google’s growth will be for the future because of their many investments today. But that also comes with a lot more assumption risk the longer you forecast.
(My assumptions are growth of 15% in OCF (plus cash and equivalents) while maintaining a healthy multiple of 25. Includes a 30% margin of safety.)
Quick thoughts: we all know that none of the current AI models make money, and they are all burning money, and it's not going to stop simply b/c of this reason: the subscription model doesn't nearly cover their costs on compute and inference. So, while everyone is talking about how Google "could" be disrupted, no one is talking about how long it would take for the so-called "disruptors" to be profitable and how many more rounds of funds they need to raise.
Who will be the last man standing from this bunch, do we even know? And don't forget what DeepSeek (and there will be many more coming out of China) has done is lowering the costs and entry barriers for more AI startups to compete which will bring down the revenue of these AI companies (Good luck Sam Altman) Google, on the other hand, has a massive war chest at its disposal and is positioned to disrupt itself IF it chooses to, the question is when and how. Google knows the dilemma it currently faces, so it's not like they are waiting to be disrupted
Thanks for your input, really well said. Of course Google has the potential and ability to disrupt itself, but it's an incumbent's dilemma. Google Search generates almost $200 billion in revenue, why would you want to disrupt that?
It'd be an extremely tough decision to make with a lot of short-term pain.
And the fact that LLMs are getting commoditized doesn't mean they won't pose a threat to Google Search anymore...
I guess what I am trying to say is that if time is the factor here since it's about who will be the last man standing (burn rate is not helping any of the current AI companies), the only ones that stand a chance are the big techs, so I won't be betting on OpenAI at all. Also, their current valuation has already dropped according to Scott Galloway
you would choose to disrupt yourself if you can find an alternative that won't disrupt your revenue and profits in the long run. There are plenty of examples of such in the past
Very fair writeup and I agree. The ad business is facing an existential threat and it’s hard to handicap. I’m sure some people thought Kodak was cheap in the 1990s when the digital camera started taking off or Blueberry when the iPhone was introduced. I don’t know what the future looks like but I’d rather invest where I have more certainty.
Nice analysis, once again. It really highlights why search is so important to them and the potential future expectations for the stock. However, I still see search as an opportunity. Here’s why: right now, they still hold by far the largest market share when it comes to search. But if they want to maintain search’s power as it stands, they’ll need to leverage their current position more effectively. What I mean is, they can improve search beyond its current state and make it more competitive with LLMs. Given the massive data they have, they should be able to reimagine search in a way that makes it not only more efficient but also more compelling. I hope you can see the bigger picture I’m trying to paint here. I believe this approach is key to sustaining their dominant market position.
Hey Felix, thanks for your message. I completely agree with you and wrote a similar thesis to yours in the article:
"If Google Search evolves to offer both conversational, answer-focused AI responses and traditional search results, LLMs from other players could become obsolete for answering queries. By then, LLMs will obviously still have its place, but less so for searching answers."
However, my valuation model assumes that Search will continue to be successful so you could consider my valuation a bullish scenario. I'm not saying Search is doomed; far from it. But even so, Google isn't drastically undervalued...
Great post. I think the problem with most is they use the "a stock is cheap" argument without taking into account the timeframe of ownership. Google I believe is a buy, but that's from me using a DCF for the next 5 years. The longer you project, the less attractive it is, of course. But time matters.
Hey Jacob, thanks. So the longer your forecast period, the more attractive an investment looks, not less. The explicit forecast period captures all cash flows where ROIIC is above WACC.
A shorter timeframe of just 5 years should make Google look even less attractive, because a lot of Google's value is captured beyond that period. What are your assumptions for your 5-year DCF?
It can go both ways. A longer forecast can artificially make a stock look less attractive, especially if most of the growth is near-term and it slows with time. Like with Google and cloud now. (Like most mature businesses)
I agree with you though. You’re absolutely right, a lot of Google’s growth will be for the future because of their many investments today. But that also comes with a lot more assumption risk the longer you forecast.
(My assumptions are growth of 15% in OCF (plus cash and equivalents) while maintaining a healthy multiple of 25. Includes a 30% margin of safety.)
Quick thoughts: we all know that none of the current AI models make money, and they are all burning money, and it's not going to stop simply b/c of this reason: the subscription model doesn't nearly cover their costs on compute and inference. So, while everyone is talking about how Google "could" be disrupted, no one is talking about how long it would take for the so-called "disruptors" to be profitable and how many more rounds of funds they need to raise.
Who will be the last man standing from this bunch, do we even know? And don't forget what DeepSeek (and there will be many more coming out of China) has done is lowering the costs and entry barriers for more AI startups to compete which will bring down the revenue of these AI companies (Good luck Sam Altman) Google, on the other hand, has a massive war chest at its disposal and is positioned to disrupt itself IF it chooses to, the question is when and how. Google knows the dilemma it currently faces, so it's not like they are waiting to be disrupted
Thanks for your input, really well said. Of course Google has the potential and ability to disrupt itself, but it's an incumbent's dilemma. Google Search generates almost $200 billion in revenue, why would you want to disrupt that?
It'd be an extremely tough decision to make with a lot of short-term pain.
And the fact that LLMs are getting commoditized doesn't mean they won't pose a threat to Google Search anymore...
I guess what I am trying to say is that if time is the factor here since it's about who will be the last man standing (burn rate is not helping any of the current AI companies), the only ones that stand a chance are the big techs, so I won't be betting on OpenAI at all. Also, their current valuation has already dropped according to Scott Galloway
you would choose to disrupt yourself if you can find an alternative that won't disrupt your revenue and profits in the long run. There are plenty of examples of such in the past
Helpful and timely post. In-depth but not too much. Thanks!
Glad you found it helpful!
Very fair writeup and I agree. The ad business is facing an existential threat and it’s hard to handicap. I’m sure some people thought Kodak was cheap in the 1990s when the digital camera started taking off or Blueberry when the iPhone was introduced. I don’t know what the future looks like but I’d rather invest where I have more certainty.
Uncertainty is what the market really dislikes
Nice analysis, once again. It really highlights why search is so important to them and the potential future expectations for the stock. However, I still see search as an opportunity. Here’s why: right now, they still hold by far the largest market share when it comes to search. But if they want to maintain search’s power as it stands, they’ll need to leverage their current position more effectively. What I mean is, they can improve search beyond its current state and make it more competitive with LLMs. Given the massive data they have, they should be able to reimagine search in a way that makes it not only more efficient but also more compelling. I hope you can see the bigger picture I’m trying to paint here. I believe this approach is key to sustaining their dominant market position.
Hey Felix, thanks for your message. I completely agree with you and wrote a similar thesis to yours in the article:
"If Google Search evolves to offer both conversational, answer-focused AI responses and traditional search results, LLMs from other players could become obsolete for answering queries. By then, LLMs will obviously still have its place, but less so for searching answers."
However, my valuation model assumes that Search will continue to be successful so you could consider my valuation a bullish scenario. I'm not saying Search is doomed; far from it. But even so, Google isn't drastically undervalued...
I see, strong thesis indeed!