Hacker News Reader

Comment by jairuhme | Hacker News Reader

> “Some of the facilities figured out way too late, ‘oh we should probably be checking how much we’re paying for electricity per month.’ That’s pretty basic when 30-40% of the cost of the crop is electricity,” he says.

> “Those things were just overlooked because investors will pay for it, it doesn’t matter,”

I'm always amazed when startups operate like this. I'm equally in awe when they get capital and VC's don't notice this either. I guess the later could be naive and just assume that the tech required to make it works requires this, but it is still baffling that nobody thinks its a good idea to get one of your bigger costs down

View on HN

I always found it extremely weird that no one runs even the most basic back of envelope calculations on things like this. Like how much power and at what cost we need to input or what is the throughput of some system and then how much we need to charge just to cover the building costs...

Most simple multiplication and division... Just to get some idea of unit economics. Thus do things make sense even in your own fantasy world of ultra-rich tech workers.

Very simple reasons actually. VCs dont fund people based on logic, but on ability to BS.

Go hang out with some bootstrapped founders. They will have great common sense. If not they are dead.

What's stopping VCs or their staffs from double checking or keeping track after the money has been disbursed?

The same reason I'm slow to check my lottery tickets. Pre-check I can imagine the ticket has won. Post-check I know it hasn't.

I've always assumed VCs take gambles like this because they know it just takes a single success to cover their lost bets.

So being first to bet on a possible winner is probably statistically better than taking the time to sift through the actual details of these startup ideas.

It just needs to last long enough to sell it forward to more dumb money (e.g., Saudis).

Because in order to do this math you actually have to understand the business/product- which should tell you something - or there are is some hand waving bet that like “costs will come down with scale”

People do. They get attacked for it actually. There was this blog post about an indoor wheat farm exhibition piece and it criticized how it actually consumes more land when you add in the land requirements for renewable energy. The comment section consisted mostly of people attacking the author.

There are plenty of people on HN who will tell you how growing LEDs are more efficient than sunlight and other nonsense to justify vertical farming.

> Most simple multiplication and division... Just to get some idea of unit economics.

I always considered this to be the fundamental promise of vertical farming. Maximum efficiency and easy forecasting of outcomes. Basically, reducing farming to a function with several very controllable inputs and an easy computation yielding the outcome.

> but it is still baffling that nobody thinks its a good idea to get one of your bigger costs down

Bean counters can rarely distinguish between different types of cost. Labor costs can scale with automation and tech, but energy input can typically only make incremental efficiency gains, which decrease over time. So you really need to look at energy input vs your competition.

Even those willing to pay a premium for “local farming” are doing so because of the supposed environmental benefits. But if energy to grow indoors exceeds energy for regular farming + transportation then you’re bad for the environment in addition to wasting dense city space. Unless you’re in a highly unusual electric mix region.

> Right now, the economics for vertical farming make the most sense in the Middle East, where extreme heat prevents the growing of crops in fields, food security is a pressing issue and energy in the region is cheaper, says Rabobank’s van Rijswickl, as well as former Infarm employees.

LOL, case in point. Wonder where that “cheap energy” is coming from. This article is waaay to soft in their assessment.

In general, it’s not that hard to do back-of-the-napkin energy and CO2 calculations. No amount of buzzwords and green-tinted concept art changes that there are real physical quantities at play.

There's probably some well established hand waving routine that's played whenever energy comes up. Perhaps there's some paper that starts with a ridiculously high first guess for energy demand and then goes through a list of reasons why it would not be quite as bad, concluding with some multiplied percentage of how it would be so much better than expected.

Certainly including the old "photovoltaics use a wider spectrum than plants, so if we take the best panels, re-emit the energy at a wavelength that matches the plant's chemistry we could get more from a given square unit of sunlight if everything else was free". Last five words conveniently omitted.

I believe VC at this stage are willing to accept wasting money for the chance of finding that one solution/technology that will make it all worth it. They aren't paying for you to run a business, they're paying 100 crazy engineers/scientists like Sam Altman to get one Open AI for agriculture. They know you're being wasteful. They can afford it and can't be bothered to help you run your business by checking your bills.

> I'm equally in awe when they get capital and VC's don't notice this either

VCs invest other people's money (Limited partners) and get paid fees based on the size of their fund. VCs also get paid when they hit a homerun, but it's much easier for a VC to double how much they invest (and double their fees in the process) than to double the ROI by making more intelligent bets. The bigger the checks they write the more they make.

VSs don't want to to get blamed for making dumb investments, as that would interfere with their ability to raise money from LPs in the future. That's why they want another firm to take the lead in any investment round.

The incentive structure guarantees that VCs make terrible investments and have very little accountability.

> I'm always amazed when startups operate like this.

I'm not. This is just people trying to acquire other people's money as efficiently as possible. Siphon off as much as you can and then disappear.

I'd say that's how farmers operate if they're new farmers. They will screw up their first year and start looking for help, asking other farmers for advice, etc - but these companies aren't a part of a community, they don't want to help each other, and when they run out of the investment money the first year, they just give up.

Colossal due diligence failure; I'd say the investors in the VC funds that financed Infarm have a lawsuit.

Typical of a lot of VCs they have absolutely 0 idea how to judge a business plan. So they just gamble on people and companies that look interesting to them. This is one of the least efficient if you look at one company, but it injects so much money around that it helps people bootstrap their own ventures with the money that leaked. So indirectly it may be a not too bad model that requires pure incompetence. (last sentence is almost /s)

Which startup has everything figured out in the start? If investors are this picky, they are not investing in startup and won't be getting a deal which could grow 10x.

If Tesla failed due to high cost to manufacture or OpenAI failed because it couldn't earn revenue or Uber failed because of doing illegal stuff or overpaying drivers you would have said the same thing.

So invest a tiny amount to figure that out before investing a lot of money. There is a reason funding goes in rounds - round one we have a few smart people who want to work together, fund us for 6 months (maybe a year) to figure out where there is a market. The output of that round is those calculations, then you get more money for round two to create tiny prototypes/models to verify the problems you identified in round one are solvable, then round three to perfect the prototypes, and round four to scale it. You can add or remove rounds as needed, the important point here is that as rounds more $$$ the risk needs to go down.

No, my point was that tiny bit wouldn't have worked with this. Some startup get benefit of scale like Tesla or OpenAI or Uber. If they are constrained in how much money they could burn they wouldn't have gotten anywhere. They all chose to not do calculation on RoI and won(but most would lose). To make vertical farming successful we need new discoveries and inventions that requires money and has uncertain success probability.

Startup investors are wary of the fact that they are more likely going to loose all of the money.

Tiny would have been enough to run the math.

Some things do need a large scale to work. However you should do whatever prework is needed to run that math before investing in scale.

I don't think you are getting my point. The math of RoI can't be run as the probability is uncertain.

One example is OpenAI invested multiple billions in research and just out of pure luck found PMF for ChatGPT which even they said they didn't expected. Vertical farming is like that. You need miracle to make it better than normal and miracle requires huge chunck of money to even have non zero probability.

Reply on news.ycombinator.com