Hacker News Reader

Conflating WeWork, Uber, and SaaS businesses is inaccurate.

With a SaaS business - most of which are b2b - there are tremendous costs to switching providers and going without is typically not an option. For example, if company uses Salesforce, they must have a CRM and it will likely save them little to no money to switch to a competitor (ex. Hubspot) but it will create enormous organizational disruption as people and processes are built around the existing platform. So customer life-cycle is likely 10 years or more for the customers that generate the majority of their revenue. And if we do have a recession and they bleed customers, Salesforce can always reduce costs by cutting sales and support staff. The existing capital investments in the software are already paid-off.

In comparison, WeWork's model is predicated on being easier to join and leave than a typical office space. As a result, they attract businesses with shallower capital reservoirs. Meanwhile, WeWork is holding inventory risk in that they hold the long-term lease or the property itself. So in the event of a recession their customer base will be the first to die out or cut costs via less desks and/or working from home.

(WeWork may or may not be off-setting this risk via financial engineering but the question is at what cost as someone has to take the other side of the bet and the downside risk here is pretty obvious. And, oh look, the yield curve just inverted.)

Finally, 10 years in, Uber is still a commodity business. I generally think Uber is better than Lyft and worse than Via, but none of their advantages are anywhere near as durable as a SaaS business. Don't like Ubers new prices? Don't like the the CEOs new haircut? You can use one of their other competitors starting tomorrow. Even if you use Uber today there is zero reason you can't switch to Waymo/Apple/whatever in the future. And in the worst case scenario, Uber paid CAC to educate consumers who will use something else in the future.

Agreed. The business models and risks between these companies are completely different.

The main commonality I was highlighting was that these business have high upfront costs and make the money back over a long period of time. Because of this, profits are a not very good way to measure the health of these businesses.

Uber has network effects. It's definitely possible and inevitable even that they won't be on top anymore at some point, but what they are doing is not exactly a commodity if it relies on a dense two-sided marketplace.

While it does have network effects, I'd argue it's not recession proof (mildly cyclical, even), and there's low switching costs to using other forms of transport (for example, I haven't used Uber in lieu of Lyft / driving my own car and had no issues)

Uber has shown great growth in markets during recessions: ex Brazil. The theory is that ridesharing is a driver constrained market, and during recessions, more people drive for uber to make ends meet. Also, people invest less in large purchases (cars) and instead spend more on services with a higher marginal cost (taxi, ride share.

WeWork is aggressively buying out the good spots. They're going for a limited resource before competitors can, and unlike software, the entry barrier is just money, brand, and interior decoration skills.

Monopolizing available office space in a large city is a tall order though. But they might be able to be the coworking Walmart and drive the Mum and Pop stores out of business, through sheer economies of scale.

Oh right, I was comparing to my own city, where is a fight for the best office spaces. It might not be the case in New York City, but would make a huge difference in, say, Sydney or Kuala Lumpur.

I believe once you start hitting the 8 to 10 people, WeWork makes you sign up for a 1-year agreement since it's more difficult for them to find new occupants for those spaces. Not sure about the exact number as it may differ per location.

As for signing/leaving leases, the experience is comparable. You can leave any lease if you're willing to pay the penalties defined without much of a hassle. Usually, the hassle has to do with the renegotiation of the terms. But I do agree, WeWork has been more flexible about breaking the initial agreement.

Investments are not part of your operating income. Your income only consists of flows that accrue over time. If you make a $1200 investment, the expense associated with the investment is accounted for by spreading that expense over the lifetime of the asset. For a 20 year asset, this would become a $60 expense per year.

It's not like it's a new thing that businesses need to invest to to acquire customers. The entire science/discipline of accounting exists to provide accurate metrics of business performance. Its ridiculous to say that "profits are a bad metric" because of this.

Also, WeWork is not a software business, since they do not sell software. Why would those metrics apply here when all their money is made through leasing office space?

That’s the purpose of accounting in principle, but in practice GAAP financials are an incomplete measure of financial performance. In the instant that WeWork acquires a new customer, it realizes no additional income and no change in shareholders’ equity. But assuming its marginal revenue is higher than its marginal cost, the expected present value of its future profits is certainly higher than it was the instant before it acquired that customer.

In the immediate instance, no, WeWork does not recognize anything. But in its first month, rental income should exceed the month's depreciation for the same real estate. The fact that income has not kept up implies that occupancy is low. WeWork is fundamentally a real estate/leasing company, not a tech company.

> In the immediate instance, no, WeWork does not recognize anything. But in its first month, rental income should exceed the month's depreciation for the same real estate.

This is accurate.

> The fact that income has not kept up implies that occupancy is low.

This is inaccurate. (To be clear, occupancy may be low, but WeWork's reported losses don't imply that that's the case.) If WeWork turns around and spends that rental income plus some of its investors' capital on customer acquisition expenses, it will report losses, even if those expenses bring about a net increase in the expected present value of future profits. This is just a result of the timing difference between the recognition of acquisition expenses and the revenue that they produce.

I may be wrong but I don't think that last statement is true if the lifetime of the customer/asset is short--especially shorter than expected. The business model seems too new and for me it hasn't proven itself viable.

You’re right. Replace “marginal revenue/cost” with “marginal expected present value of lifetime revenue/cost” and that statement holds.

I’m not an accountant so I’m not sure if there are specific rules that would make this not applicable to SaaS, but the thing you describe is precisely what proper accounting is designed to solve. If you buy a factory this month and then use it to make widgets for the next 10 years you don’t show a huge loss when you buy the factory, but rather capitalize the cost and then depreciate it over the useful life of the factory so that in every period you can match the revenue with the associated costs. Shouldn’t customer acquisition costs be similarly capitalized over the expected life? If the company is actually profitable then it should show a profit not a loss. Any reason why this isn’t more common among SaaS businesses?

The big difference between a factory and customer acquisition costs is that the factory is an asset. It has enduring value. Customer acquisition costs do not. When you pay the customer acquisition costs, the money you spent is gone and non-recoverable. But you can recover some of the value of a factory by selling it.

Why isn't a customer who pays $X/month an asset? If you purchase a bond that paid the same X/month in interest, that would certainly be considered an asset on the books.

As an accountant, I can tell you that the accounting standards do require certain things to be recognised once they become more likely than not (e.g. provision for expected losses) but this is not extended to e.g. recognising customer relationships as an asset, mostly because these things are fuzzy and hard to measure and easy to game.

If the company buys another company then they can recognise the difference between the purchase price and the company's assets as a new asset vaguely named 'goodwill' which covers things like customer relationships and everything else the accounting standards otherwise miss, but this isn't allowed otherwise.

Plenty of things you can’t sell get capitalized, recoverability isn’t a particularly important aspect for capitalizing something.

At least under international GAAP, assets can only be capitalised if they are expected to deliver future economic benefits. This doesn't mean that they have to be sold directly - they can be internal benefits such as a building or a factory machine that makes it cheaper to produce goods.

How would you deal with spending a bunch of money on redesigning a company's website (in practice, as opposed to in theory)? On the one hand it has a lot of similarities to making improvements to a retail location (which would typically be amortized). On the other hand, my experience (not as an accountant, but as someone who has been exposed to accounting in small business contexts) is that that would typically get booked as an expense.

It would be capitalised as an intangible asset. In practice I assume a small company might not bother, especially if the website was built by their own staff, but the clients I worked on seem to get charged quite a lot for outsourced IT work so they would certainly want to capitalise it.

Indeed, that’s what I was trying to say as well. The asset needs to have value to the firm in terms of future benefits but doesn’t need to be sellable on its own.

I'm curious if you have any examples off the top of your head. Not trying to be combative, but I'm genuinely drawing a blank here.

It is actually more common among SaaS businesses than you think. If you look at balance sheet from Salesforce etc, they tend to have items called "deferred commission" or something similar which are captialized marketing and sales expense.

How does it factor in that they are leasing their buildings? Won't their rents eventually go up, and won't they have to remodel all the time to keep up their hip/modern work spaces? It seems like funky math that I'm not understanding.

Also, when a recession hits, it seems like the first offices a company closes has are the ones on a short-term lease at a co-working space....

[Edit] My second comment is because I work for a large company who opened an office in a WeWork space to be more hip and relevant....and I see a lot of other "big" companies with offices in the building.

Definitely not a WeWork booster, but I think I can see the other side of these questions.

> How does it factor in that they are leasing their buildings? Won't their rents eventually go up, and won't they have to remodel all the time to keep up their hip/modern work spaces?

All of this does matter, but per the OP may matter less than CAC + initial push into a market (brand-building etc.). Think of the cost to develop e.g. MS Word vs. the incremental cost to retain users (programmer salaries go up over time, they have to remodel the app ~annually, etc).

If the WeWork model works in a market, they can buy buildings later (perhaps during a recession).

> when a recession hits, it seems like the first offices a company closes has are the ones on a short-term lease at a co-working space

I'm sure the other side of this argument goes something like the arguments against car ownership. Many companies won't have any long-term leases anywhere, will be more remote-friendly, etc. In fact some companies will choose to not re-up on long-term leases in favor of more flexible options like WeWork specifically because who wants to sign a long-term lease during a recession when visibility is poor?

Good points. I do think car ownership might be different because it's a depreciating asset with a limited lifetime of use. I think real estate and buildings, that lifetime can be significantly longer...

It'll be interesting to see how this all works out for a lot of this generation "tech" companies.

>Also, when a recession hits, it seems like the first offices a company closes has are the ones on a short-term lease at a co-working space....

Apartment rents spiked during the last financial crisis because people downsizing increased demand for them. It's entirely possible that downsizing companies might want to retreat from leases to smaller WeWork offices during the next one.

Aren't commercial leases typically of much longer terms than residential leases and less convertible than homeownership?

The issue I have with WeWork, aside from the fact that they're somehow a tech company, is that they have competitors that does the exact same thing... but who are profitable.

You have a company like Regus, that's not a tech company, they have a MUCH low evaluation, but makes a profit and having more offices.

My understanding is that they make some bullshit claim about how they have instrumented everything and that helps them to apply machine learning to their community and app, blah blah. It's def not a tech company but I could see them going that way eventually.

BTW - the concept is "valuation", not "evaluation" :)

> BTW - the concept is "valuation", not "evaluation" :)

Oooh... thanks. This is one of those cases where I suddenly realise that I've been using the wrong word for year :-|

Without looking at the actual accounting rules (GAAP or otherwise) used by WeWork it is difficult to say how the line costs are treated.

In your example, you have treated $1200 as an expense.

Lot of companies tend to capitalize these expenses (for example "deferred commissions" on Salesforce B/S). So, they will create a reserve on the Balance sheet of $1200, then decide on the useful life of the said expense and depreciate it on the P&L.

Let's say WeWork thinks the useful life of the $1200 expense is 24 months. The calculation will work like this:

P&L

Month 0: Depreciation Expense 0, Revenue 0 and Profit 0

Month 1: Depreciation Expense $50, Revenue $100 and Profit $50

Month 2: Depreciation Expense $50, Revenue $100 and Profit $50 so and so forth

BS

Month 0 - Asset $1200

Month 1 - Asset $1150 (Depreciation $50)

Month 2 - Asset $1100 (Depreciation of another $50)

> For a typical software business, a pair of metrics that are much better than profit are CAC (cost of acquiring a customer) and LTV (the life time value of a customer)

Andrew Chen, ex VP Growth at Uber, had a blog up on just this: https://andrewchen.co/how-to-actually-calculate-cac/

Great post. In a recent interview this is exactly what Masa Son is talking about: https://www.youtube.com/watch?v=eDpdcWz_F_0 What I don't fully understand - isn't this covered by GAAP accounting in some way?

Usage of GAAP is entirely up to the company. It might choose to not use GAAP policies on revenue recognition and expenses. Many non-public companies chose to do that if it means they can show their revenues growing faster than it actually is.

One of the example of this was Groupon. When they filed for IPO they were using gross sales as their revenue. And then deducting money paid to their partners as expense. SEC had to step-in and ask Groupon to restate their P&L statement so that the top line was net revenue instead of gross.

GAAP does cover this. Your "high upfront costs" would be capitalized. It wouldn't appear as a loss.

It depends on what kind of costs are we talking about. If it was something like "I give one year for free to get the client, and I will make money from them in the years to come because of course there is no reason to think they may switch providers later" I don't think it will be capitalized.

A more succint way of explaining it is that WeWork is just a real estate company, and real estate is about achieving positive cash flow (rental income > lease or mortgage expenses), not "profits" since a lot of the profit comes from selling the appreciated properties down the line.

I believe WeWork leases their properties, which would make near-term profitability important.

Why do you think Uber has a high customer acquisition cost?