Big Tech Incremental ROIC Analysis - GOOGL
I'm bullish, but concerned about cloud commoditization in an AI pullback.
Disclaimer: This post is not financial advice, and is not a solicitation to purchase or sell any securities. For more info, please review the disclaimer.
If you missed it, here’s Part I in this series analyzing META :
Big Tech Incremental ROIC Analysis - META
Disclaimer: This post is not financial advice, and is not a solicitation to purchase or sell any securities. For more info, please review the disclaimer.
As you’ve no doubt seen big tech is spending hundreds of billions of dollars a year on AI. This is the fundamental driver for NVIDIA’s dominance and ascension to the most valuable company on the planet. While these capital expenditures are widely talked about, there are many concerns that the transition from relatively capital-light to capital-intensive business models will cause a huge bubble and shareholders will lose lots of money.
Alphabet (I’ll call them Google from here to match the stock ticker) is doubling their capital expenditures this year and forecasting a significant increase next year as well. They’re raising cash from equity and debt sales to invest even more aggressively in AI. $300B of Capex in one year seems possible. While no one knows the return on those future outlays, Google’s financials give us a clue as to how these investments are likely to perform. I think Google is likely to continue to earn strong returns, but given the amount of capital being deployed, particularly in Google Cloud, the hyperscalers like Google are more exposed to a pullback in AI demand than other companies like META.
Google has increased their invested capital by ~60% in the past 12 months. Google highlights that their 2026 expected capital expenditures are six times greater than 2022 Capex. 6x in 4 years. wow. Estimates are for Google Cloud to grow 60+% to $96B in Revenue for 2026 and then more than 50% to $145B in 2027. I estimate Google Cloud growth is responsible for around half of Google’s 2025 Capex and more than 1/3rd of the massive $180B+ in Capex they’re forecasting for this year.
Before we speculate about the future, let’s take a step back and look at Google’s historical financials in the table below. We see a spike in operating income particularly in year two of Covid (2021) without a corresponding increase in invested capital, leading to great ROIC for Google. Afterwards, there was a digestion period before Operating Income and Invested Capital began to ramp in 2024 with AI, Gemini was officially released in December 2023. In recent years, the ROIC has fallen but still remains excellent and above 2019 levels.
In 2025, incremental ROIC was a relatively anemic 22.8%. To be clear, this is still a pretty phenomenal result. If you can do better than 22.8% before taxes on your corporate project, you should probably do as many of those as possible! However, this is a dramatic decline from 75% incremental ROIC in 2024.
Part of the reason for this decline in ROIIC (Return on Incremental Invested Capital) is that it’s taking longer for Google to build new data centers and other construction projects. From 2011-2019 Google’s construction in progress was 70-75% of that year’s Capex suggesting it took about 9 months to get capital projects operational. In the last 3 years, the average construction in progress as a percent of capital expenditures was around 100%. This shows delays or longer capital project timelines as well as the continued ramp up in capital expenditures.
To adjust for this we can attribute this year’s increase in operating income to last year’s investments as they mature and begin to generate returns. Google’s lagged incremental ROIC increases to 45%. This is still below the historical average for Google. The true incremental ROIC likely lies somewhere in the middle between 22% and 45%. In Q1 2026, incremental (non-lagged) ROIC accelerated back to almost 40%, suggesting that Google is still seeing excellent returns on their capital projects.
As mentioned earlier, the big news in the past couple of weeks is that Google made a huge departure from their historical capital allocation and issued $80 Billion. I recommend reading this post for some more context and analysis:
With Google’s capital expenditures ramping up, no one knows what the incremental ROIC will be, as we are 12-24 months out from those expenditures beginning to generate revenue. Google has been very clear that they’re striving to maintain ROIC discipline, and keeping that in mind along with the potential for them to use all these capital investments for their own internal products like YouTube search as well as DTC and B2B AI offerings if third-party demand falls through.
Unfortunately, Google doesn’t really break out which segments they’re investing substantial incremental Capex in and what those returns are. YouTube remains very capital light, and the improved recommendation engine is probably extremely high (100%+) ROIC. In contrast, Google Cloud ROIC may be in the teens or lower if we see a slowdown in third-party demand for cloud compute.
If you estimate a 30% ROIC for Google Cloud and a one-year lag between expenditures and generating revenue then roughly $50B of 2025’s 91B in Capex was for Google Cloud and another ~$65B of this year’s 180B is for Google Cloud.
Overall, Google has one of the best businesses in the world, and they have pivoted aggressively to include AI overviews and summaries in their searches, allowing them to claim “Google Search is delivering AI to more people than any other product in the world.” I’m optimistic that Google will see strong returns on their incremental capital expenditures here, and I think they were smart to proactively raise capital and get the public blessing of Berkshire Hathaway.
However, I’m less confident in the future returns and ability to repurpose this hardware if third-party demand fails, then I am from META. At the same time Google trades at a much richer valuation than META. I still like Google, but in a pullback the hyperscaler business model is at much greater risk of commoditization. Consequently, I like Google but have a heavier weighting in META at this time.
Appendix:
Google call explaining the equity raise and AI Capex opportunity:
GOOGL June Investor Presentation Transcript
GOOGL June Investor Presentation






I wonder how Google balances the high capital spending with maintaining strong ROIC, especially as construction timelines extend and AI demand remains uncertain. This could shape the long-term viability of their strategy.