wsetchell

https://businessquant.com/intuit-revenue-breakdown-by-segmen... https://investors.intuit.com/news/news-details/2022/Intuit-R...

The vast majority of Intuit's revenue comes from products for small businesses (e.g. Quickbooks).

It'd also be really interesting to a similar view on cost per square foot. The US is probably expensive from high labour + land costs, but it might more even expensive than those would suggest.

I wonder if there's a way to factor in planning time needed. My perception is that the US requires years of (e.g. environmental) reviews before construction can begin which countries like China do not have.

TSMC has a great business today, but world governments are dumping around a few hundred billion dollars in subsidies towards local TSMC competitors (e.g. ~$25B-$75B from the CHIPS act alone depending on how you count)

TSMC's moat isn't enough to withstand that much subsidized competition. All of those new competitors will chase market share and destroy the profitability of the market.

Solid move.

What is this fraternal order and how did you find it?

Average life expectancy is down. Life expectancy for rich people is up, so non-rich life expectancy is dropping faster than the average.

Conquistador: Hernan Cortes by Levy - How Cortes conquered the Aztecs with 400 men and 16 horses.

Judgement in Managerial Decision Making by Bazerman - Things which bias your thinking and how to avoid them.

My Life and Work by Henry Ford - How Henry Ford thinks. Stories about engineering and building a company.

Homage to Catalonia by Orwell - The famous author's experience in the Spanish civil war which colored much of his work.

I would think of this like a normal customer development / lean startup problem.

Is there anyone other than your company using this API? If not, it might make sense to find a single "customer" and make sure the API solves tons of problems for them. If you cannot get first adopter, iterate on the product/API until you can create something that a customer loves.

It sounds like people don't see strong enough value in being the early adopters of these APIs. You might need to iterate on API/product until they do.

Increased signal (e.g. more or different data) is one of the many levers to increase ad value.

Without changing signals you can still make meaningful long term progress through ranking, delivery, and format improvements.

I wonder how they're thinking about this versus the steady march of solar + battery performance.

I would think that in the 5-10 years it'll take to get this safe enough for a commercial product, solar + battery would be scaling and they'd never be able to catch up on price due to the efficiency ceiling.

Maybe there's something possible in aviation for this. There might be electric drones/planes applications which would be way better if they didn't have to carry batteries.

The article does not point out how the system can go astray.

Design docs are great when they can in 1-2 pages describe a how an important/large/complex system will work and the tradeoffs made when designing it.

Some groups in Google confuse design docs conflated with the promotion process, and thus engineers start create long/time consuming docs for every little thing as part of building a case for promotion.

To justify spending that money, they'd have to think they'd get better returns on their R&D than shareholders could get elsewhere (e.g. by just buying the S&P 500).

It is quite difficult to beat the S&P 500, and they're saying they don't have any ideas which can.

I don't get people's aversion to stock buybacks.

If a company has money it cannot productively spend, it should return that money to shareholders. Stock buybacks are a tax efficient way to do that.

The company believes the asset wasn't very profitable for them to own, so they sold it and gave the shareholders the money back.

That packaging looks incredibly expensive. Even if the other components/service was free, that alone would make this price uncompetitive.

That makes me think this is a long long way from being a self sustaining business.

For the most part, the standard strategy still works fine; spend less than you earn, keep an emergency fund, invest in low-fee stock/bond funds like the ones recommended here https://www.bogleheads.org/wiki/Three-fund_portfolio.

As real interest rates go lower, the cost of capital goes down. That would make some capital intensive businesses possible/profitable that wouldn't be otherwise.

I don't think deflation is the natural state of things. Even before fiat currency the money supply grew (e.g. gold mines) and we had inflation.

Productivity growth is normal-ish, but the amount you prices have been all over the place throughout history.

Waymo has the best technology in self driving.

They have not deployed their tech at scale, when competitors (Tesla) has widely deployed worse technology.

Even if they deploy their tech widely, it isn't clear it will generate significant revenue/profit. It might not be that expensive (think less than 1B) to build a good enough self driving car in 10 years. That'd create competition and drive down prices.

Someone else might figure out how to capture the value of self driving cars too. Maybe the profitable parts of self driving cars are the "apps" you can build once self driving cars are cheap.

I wouldn't invest in Waymo at 100B until they have a real business with real revenue and a real moat to protect that business.

Google takes large long-term bets that are closely related to their core business / competency. Some of those seem to be clearly working (see Cloud, Deepmind). As a shareholder, I want to see more of those.

Back when I was at Google, X (which many Other Bets came from) had a goal of all their projects having meaningful impact in 10 years. They've been working on Wing and Waymo for close to 10 years now. Those projects are not yet meaningfully impacting many people.

As an armchair CEO, I have doubts about the compensation (more salary / less equity vs startups) and funding (fewer choices funding sources for the companies, weird incentives for the investors vs VC funds) model for "Other Bets". Based off of that and the lack of results, I think they should force the Other Bets to stand on their own vs handing them more cash to burn.

Overall revenue going up 20% YoY is amazing. The other bets going up 20% YoY is disappointing.

It is interesting to see the low revenue, slow revenue growth (~20% YoY), and quickly growing expenses from "Other Bets".

If the Alphabet model made sense, I'd expect to see a company starting to clearly take off after 5 years. The data shows the opposite.

Yeah I was there during that time too. Chasing down launchcal bits was really expensive.

Requiring every product team to go through a central human rights review seems like a bad idea.

Not only would it be expensive (new overhead for lots of product teams), it would be counter productive.

Today the product leaders are accountable for the impact of their decisions, since they are the only ones who are making them.

If there's a central decision committee, the product leaders would be able abdicate their responsibility since they didn't fully make the decision.

Over time incentives will cause the the product leaders to push for more risky decisions, committee to get less risky, and the whole company to slow down.

I think it might actually be the opposite.

Anecdata:

- <50% of the people who joined my team recently were referrals.

- >50% of my friends in "serious" relationships met offline.

I googled this too. The second definition fits better here

'reduced in force, effect, or physical thickness.'

Re 2) Biology and Physics are both really expensive (think particle accelerators and human genome). We could dramatically increasing academic computer science funding to let tackle those large interesting problems.

Or 29 if you include WhatsApp.

Here's some problems YC might be able to fix.

Equity Liquidity -- I can't wait 10 years to get liquidity on options. YC could offer to purchase some employee equity when the company raises additional rounds.

Employee Branding -- FANG on a resume looks better than working at a startup. That really helps when it is time to look for a new job.

Thanks for the explanation!

Short of turning off ads or ad targeting, do you have any ideas about how to make this better?

I hear this point frequently, but never really understood it. Can you explain? Here's the flow as I understand it.

1) Advertiser tells tech company "please show this ad to people you think are interested in X"

2) Tech company uses its private data to figure out who to show the ad to.

3) If you don't click the ad, end of story. No data leaves the tech companies servers to tell anyone anything about you.

4) If you click the ad, the place that ad directs to will know the ad you came from. Åt most, they can use referral source to infer some data about you.

To me, data is sold means that large amounts of personal data are shared about me without my consent. What am I misunderstanding?