This is literally some of the most expensive real estate in the city, and office space in central Tokyo is not cheap. I'm amazed that they can find enough tenants to sublet this at a profit.
That said, WeWork does have a Japan-only advantage here, because startups generally find extremely hard to lease office space there: even if they can find a landlord willing to take a punt, they are often required to pay one to two years' rental as deposit.  This won't help them scale outside Japan though.
I'm sure some of the hustle and bustle is due to the novelty, much like how WeWork was when it first launched in the US. However, you're right that there is a real lack of flexible month-to-month space in Japan so that may be a lasting advantage.
But anyway, maybe I haven't seen the cheaper Tokyo co-work spaces. If you have some convenient points, I would appreciate a pointer!
Edit: I'm actually in Shibuya every few months. When I'm there I work in the Shizutetsu bus station. If you see a middle aged white guy there programming, come say hi ;-)
I don't remember the specifics off the top of my head, but I'm sure there are/have been ~300 yen/h ones in for example Shibuya.
There's a smaller one called Open Source Cafe that lies a 5~10 min walk from Shimokitazawa station - that it's off makes it quiet and nice, and they host meetups from time to time.
The market rate for a fixed-desk designated membership seems to lie around 40,000 JPY / month in the hotter areas like Shibuya and Shinjuku, which is ~20$ per day if you only go there for weekdays. I think that's quite reasonable. I think WeWork has been part of driving this change, but it's definitely overpriced as things are now.
He was charmed - it worked on Benchmark as well but at least they didn't bet the farm in followon
So maybe it's for small groups? But if there's only 2-6 of you, you could go to the literally hundreds of coworking spaces in Tokyo for 20% of the price of a WeWork hot desk. And use the savings to pay for beer.
The only big-ish group I've heard that used WeWork in its initial growth is PayPay which.... is a joint venture between Yahoo Japan and PayTM.
It feels like they're running on some culture thing, but there's basically no math that makes WeWork a valuable choice. At their price points, it starts getting cheaper to just hire someone full time to manage your office space. And if you have that kind of money you can make the deposit.
Startups or anyone in need of office space on a budget just isn't able to afford a proper office, especially if you're looking for something short-term, or unable to commit to 1-2 years.
WARNING, WARNING, WARNING!
It's a lot easier to be profitable when your owner buys 20+% of your services. Also easier to get more than 40% institutional customers when the boss puts his thumb on the scale for the first half of that...
I think your math is wrong. The denominators vary for the different percents. (Edit: nope, mine was.)
Softbank buys 20% of their Japan space, so in order to get to 40+% they just need to find institutional investors for another 20+%? The denominator for both is "capacity at Japan office"
It looks like they just need 25% of remaining office space to be institutional to hit that (20 of 80).
That doesn't really affect your original point though.
If Softbank is renting 20% of space, and less than 100% of space is rented, then they count for more than 20% of the customer base being institutional.
Pay no mind that:
> WeWork entered Japan when office vacancy rates were near a decade low. The supply has tightened since to around 1%, a level not seen since the 1990s, allowing landlords to raise rents, while requiring a deposit of as much as a year’s rent.
They've got a great new product too:
> The new Passport service would give tenants flexible use of desks, community spaces and meeting rooms, amenities for which visiting customers currently have to pay extra. The service is a work in progress, but if adopted, WeWork aims to set aside a certain number of desks at all locations for Passport users, the people said. The move is aimed at increasing occupancy, the key to making locations profitable.
And forget that:
> Moreover, SoftBank itself has been a big contributor to that high level of occupancy. The Japanese conglomerate and its subsidiaries take up about 20% of WeWork’s office space in Japan, the people said. Over the years, WeWork pushed its locations worldwide to attract corporate clients because those contracts are more lucrative and stable. But Japan had a higher proportion of enterprise customers from the start, beating the 40% global average.
At least the layoffs will be humane!
Ahh, those blokes came up with something, which IWG (Regus) offers for decades for a much lower price minus the "cool" factor?
Well, that makes sense and all, but it says nothing about the more fundamental problem with WeWork's business model, which is whether or not it will go bankrupt when it hits its first recession, occupancy rates generally plunge, and WeWork has more short-term commitments from its customers than it has to its landlords. Perhaps there's something in WeWork Japan that addresses this, but if so, it's not mentioned in the article.
This part kills it for me. WeWork pumped their valuation by pretending to be a tech business when they're actually a office rental business. Son is still pushing that, he genuinely believes a bit of machine learning pixie dust is going to push occupancy up and make the company profitable but there is no reason to believe this. Frankly, I suspect if you do create an AI that's predicting and managing your occupancy then you're better off letting it go trade the stock exchange.