Hacker News Reader

Comment by cromwellian | Hacker News Reader

It's always surprised me that Yelp, like Flickr, and other early services, became stagnant when the mobile transition happened, and failed to leverage the opportunity.

Why didn't Yelp early on see that they could become OpenTable, or GroupOn? In China, services like Dianping combine all three plus mobile payments: Restaurant reviews, booking, discounts, and payment, and even delivery services.

Surely Yelp could have seen how other crowd sourced content farms became commodities? Yelp is trying to blame their problems on Search, but the real problem to me is that they failed to adapt their business model in the face of a sea change in computing and from mounting pressure from competitors like TripAdvisor.

The Yelp CEO seems to be indulging in the classic approach when your business model fails to resort to patent lawsuits, copyright lawsuits, or begging anti-trust regulators for regulatory capture.

View on HN

It reminds me of the "Sears could have been Amazon" post from 2007[1]. I don't know if it's selection bias, but asian companies seem to be ridiculously diverse (Mitsubishi makes cars, elevators, air conditioners, and is a bank. Yamaha makes pianos, electronics, and motorcycles). US companies seem super focused on maximizing their core business. I'm sure they have different incentives that cause this.

[1] https://www.metafilter.com/62394/The-Record-Industrys-Declin...

GE? From television stations to jet turbines for aeroplanes.

The US does have a handful of relatively old, diverse companies, but when traveling or reading into other countries it seems like diverse companies are either more user-facing or more common. It just seems common in the US to hear about once innovative companies over invest in their main business model and die off. Retail like Woolworths, Sears, and Toys R Us come to mind, but so does Flickr. Maybe that's preferable since older companies carry along cruft, but it's interesting to compare the two.

GE also has a massive finance arm.

Which is a requirement for these conglomerates. Both Keiretsus and Chaebols have banks.

https://en.wikipedia.org/wiki/Keiretsu

https://en.wikipedia.org/wiki/Chaebol

https://en.wikipedia.org/wiki/Zaibatsu

> I don't know if it's selection bias, but asian companies seem to be ridiculously diverse (Mitsubishi makes cars, elevators, air conditioners, and is a bank. Yamaha makes pianos, electronics, and motorcycles). US companies seem super focused on maximizing their core business.

US companies tend to use wholly owned subsidiaries for stuff like that.

The steering wheel in my 90s pickup has the United Technologies logo on the casting. That's almost a completely different industry (aerospace and civilian automotive doens't have a as much common supply chain as you'd think).

The US has fluctuated between conglomerates being in vogue and not. GE being one of the last surviving conglomerates and on it's way out. You can chalk this up to a variety of reasons including: tax policy, activist shareholders, etc. Many of the Japanese companies you mention don't have the same pressures to break up.

This article talks a bit about the conglomerate break-ups of the 70s, which includes a great graph showing the rise and fall of such:

article: https://www.cbinsights.com/research/disrupting-management-co...

graph: https://s3.amazonaws.com/cbi-research-portal-uploads/2018/05...

Traditionally, American conglomerates flourish in high-interest-rate environments, such that the money emitted by profitable verticals can be used to subsidize growing/unprofitable verticals, all beneath a single centralized authority which overtly acknowledges its legal and financial control of the subsidiaries.

Conversely, when money is cheap, you can just incorporate all your verticals as freestanding entities (aka "startups"), often beneath a decentralized control structure (aka "cabal of venture capitalists who collude with one another") through which you maintain influence over them without incurring any of the unpleasant legal ramifications caused by explicitly linking the verticals together in the eyes of the law. You can do this latter decentralization thing since you're freed from the frequent need to move money between verticals, and can instead just pump money into them (if necessary) every year or two (aka "funding rounds").

However, I have no clue how Asia works.

Look up "Mitsubishi Group" on Wikipedia for some interesting insights into how "a decentralized control structure" works. I suspect that this sort of decentralisation is common not just in Japan, but throughout most large players in the global economy. The differences will mostly arise from how much the "home market" of any of these groups (cabals) will care for, or acknowledge antitrust practices.

Other good examples are Unilever, Google, Apple, Freeport McMoran, Tata Group etc. Once you bring China into the equation it becomes more difficult to trace the global connections due to massive state ownership, but I'd be surprised if they didn’t have interests in many global conglomerates.

This isn't really conspiracy theory, this is just how a global economy is evolving amidst multiple conflicting interests and regulatory bodies.

There are other conglomerates.

Berkshire Hathaway is an extremely good example. It includes a wide variety of companies that like to spin off cash, such as See's Chocolates, combined with a core insurance company that could need access to very large financial reserves. According to http://www.berkshirehathaway.com/letters/2017ltr.pdf they are operating under the assumption that a $400 billion catastrophe has a 2%/year probability, and they are prepared to weather such an event.

Another good example is Amazon. They are the humble bookstore that does everything from web hosting to delivering groceries.

My understanding is that the Japanese conglomerates are different from US conglomerates.

The US ones are largely siloed collections of companies across a spectrum of industries.

The Japanese ones are strongly vertically integrated. I was told when I was sponsored by Yamaha for international ski racing competition and at one of their testing camps: 'we mine the ore, make the metal, make the machines that make the tools, make the tools, and then use the tools to make the products'. It's about both controlling the quality all the way up and down the value chain, and also capturing the value-add all the way up the chain (not sure how much of each).

I was quite impressed by their quality and technology, which was the prime reason I went with them. At the time, they were one of only two companies worldwide that could make a pair of skiis that was actually indistinguishable left/right to racers at our level (the other was Fischer which dominated the Swiss, Austrian teams and we wouldn't get their best race stock), and Yamaha had amazing ability to tune the performance properties. I'm quite sure that they could do it because of their strong vertical engineering and QA integration.

(Always seemed like a superior model to me, but I'm not a big biz guy, so what do I know?)

Edit: typos,clarity

What kind of ski racing did you do ?

I've never heard about Yamaha skis. But I am from Europe so in alpine skiing (FIS races) I just know the European brands (Rossignol, Fischer, Head, Atomic or even more obscure ones like Stöckli).

The Olympic Men's Slalom in 1992 was won on Yamaha skis.

Same, FIS Alpine racing. Mostly Downhill for me, but raced all events. Mostly on the Europa Cup and NorAm circuits and a bit at WC level.

I'm not sure if Yamaha is still active in racing, but they had some really amazing skis, and they tried out a lot of interesting ideas. One of my favorites was a triple-core construction where two of the cores were a high-energy elastomer pre-stretched in the mold to ~5x their original length. The idea of this pre-load was to have a moderately soft flex but return a lot of energy as the ski un-bent as you exited the turn -- and it worked like crazy -- the best pair of GS skiis I ever had! I also had a pair of 220s to try for DH; they were not fast probably because they had too much energy and vibration on the straights, but they were an absolute blast for hi-speed free-skiing -- just crazy spring out of the turns -- I was sad to have to turn them back in since they weren't going in my race bag.

Fun times...

That Sears story was fascinating, thanks for sharing it.

I thought it was because when Japanese companies go for financing the bank insists that they take their name. So you have tons of basically unrelated companies with the same name.

Source?

I heard that third hand but it does seem to track reality.

https://en.wikipedia.org/wiki/Keiretsu

Mitsubishi makes a very good example. I remember back in the '80s being struck by an issue of Consumer Reports that reviewed two different Mitsubishi products - one was a car, the other a can of tuna fish.

American companies aren't diversified in the same way because investors realized that breaking up a conglomerate maximizes the stock prices - the underperforming parts put a damper on the stock.

I think it's a rather bad example because all that the different Mitsubishi firms share is a name and a logo. They're four independent companies.

I am not verse enough in the legal matters yo have concrete examples, but I think Japanese style zaibatsu wouldn’t fly in the EU or US from regulatory perspectives.

The main goal of having conglomerates is to modularize essential parts, in particular a financing branch that acts as a money pump through the whole group.

So by definition the companies from the same group are defined legally as separate units while fixing deals between themselves and protecting the other companies’ turfs.

Not sure about how the Japanese are structured so I can't comment about that. But large conglomerates aren't impossible in the US, GE is probably the best example.

Or the fifth largest US company by market cap: Berkshire Hathaway.

The fact it's fifth probably has more to do with it. Conglomerates have the overhead of vast companies but gain minimal benefit from that overhead.

Berkshire Hathaway is a poor example to use for that. They famously contribute essentially nothing in overhead costs and instead make a fortune for the overall entity through smart capital allocation and investing.

Again though they don't have dividends yet are worth less than Apple which is a younger company that hands out dividends. BH's overhead is not direct costs but relatively poor asset allocation due to it's vast size making the overhead of smaller investments unwieldy and lacks focus to maximize gains on any component.

PS: You can argue about performance. But, in the last 20 years their stock has done slightly worse than the Dow when you include dividends. Meaning they could have done better investing their profits in an index fund.

https://dqydj.com/dow-jones-return-calculator/ may 1998 - may 2018 = 323.965%

http://www.macrotrends.net/stocks/charts/BRK.A/prices/berksh... May 1998 - May 2018 = 320%.

> American companies aren't diversified in the same way because investors realized that breaking up a conglomerate maximizes the stock prices

Is Google or Amazon not at the conglomerate level yet? Is it simply because their services and products are all on the internet?

I've seen the trend go the opposite direction - conglomerates are in now, because they give you nice steady dividends and less shock when one part of the company is failing. If not for AWS, where would AMZN stock be today? I've seen no serious effort on the investor level to break up these companies, and practically none politically (fledgeling movements on the far ends of the spectrum).

Don't Google and Amazon still make substantially all their revenue from a single business? I could see that changing for Google if they're successful with self-driving cars.

The pressure to break up a conglomerate comes from outside investors who don't have a stake in running the businesses. There have been hostile takeovers motivated completely by the anticipated profit from a breakup. I don't see that happening for either Google or Amazon.

No need to wait for Waymo to get your example: AWS is already a large and fast growing business.

> AWS is already a large and fast growing business.

But isn't AWS the same as Amazon, just instead of selling a new vacuum, its selling cloud compute? Its still e-commerce, just rebranded as "cloud provider."

Leasing services opposed to mail order retail? Seems like a very different business to me.

Sure, they share a lot, like:

- Billing: One bills you before shipping, the other bills at the end of the month.

- Cart: One lets you add items to a cart, the other lets you use products via API.

- Delivery: One requires warehouses all over the world, the other requires data centers all over the world. Only warehouses and data centers are so specialized that they aren't interchangeable.

- 3rd parties: One is a platform for creating online storefronts where 3rd parties can sell. The other is a platform that runs any code you want.

- One has a mass market audience of consumers, the other has a highly specialized market of developers.

Hmm, on second thought, maybe they are not the same business.

> asian companies seem to be ridiculously diverse (Mitsubishi makes cars, elevators, air conditioners, and is a bank. Yamaha makes pianos, electronics, and motorcycles)

This is because of political influence/leadership. Such companies would not have existed unless there was a political will to build large conglomerates like the zaibatsu in Japan (same thing in Korea with Samsung that does about everything). This is hardly free market at work.

Besides, this is not what you think it is. Mitsubishi is actually composed of hundred of sub-group companies that have basically nothing to do with each other, and simply share a larger umbrella brand, but in effect act as semi-autonomous entities.

I am surprised that I don't see anyone peddling this idea:

in markets which are corrupt or there is expectation of weak enforcement of contracts between different companies - the only viable option often is to acquire what you need under your own umbrella. This is an explanation I read somewhere for why conglomerates may have an advantage.

Xerox PARC developed most of the technologies that would define the personal computer, but Xerox didn't have the culture, structure or expertise to properly exploit those innovations.

Yelp was founded in 2004 and went public in 2012. By the time the mobile transition was really in full swing, they had accumulated a lot of institutional inertia. Their developers had spent years working on a big web app. They had a huge codebase that hadn't been designed with mobile in mind. They had a particular revenue model and brand image.

Turning that ship around requires courageous leadership. You need to recruit or retrain a lot of developers. You need to convince your shareholders to support a major investment that might not pay off. You need to gamble on a new business model that might undermine your existing revenues. You might need to navigate existing internal conflicts or risk creating new rifts within the company.

This is what makes startups so powerful. They're focused, they're not weighed down with institutional baggage and they have permission to fail.

Despise yelp because every time I end up there (usually via Apple maps) they push so hard to force me use their app instead of website, I swear never to use it again.

Me too. I made a restaurant reservation using their site. It worked fine. Text message confirmation led me to a page to confirm or change the reservation, with some tricky language to make one think you need to confirm the reservation you already made (you don't) and that you have to do it using their app.

I hate it because I know three restaurant owners who've told me their personal experiences with Yelp. TL;DR: Yelp is an extortion racket.

That right there is why I find Yelp's "but think of the children^H^H^H^H^H^H^H^Husers!" argument to be ridiculous. Yelp is actively user-hostile. Doing what's best for a Google user would only make Yelp's situation worse.

Yup. And additionally, the mobile app is a sub-par UX experience compared to the desktop website.

I really don't understand why - for years you couldn't even enter a review via their app (and later you had to login to the website to complete it).

To this day, you can't so much as search the text of reviews in their app and be directed to the review containing the string you searched for. You just get results for establishments that apparently contain your search string somewhere in the reviews, and have to ross your fingers hoping that text was "the tonkatsu is great!" and not "too bad they don't have tonkatsu!"

You can! It’s just incredibly hidden. This is on iOS:

Business Entry > Click the XXX more reviews button (beneath tons of content) > scroll up and then search within reviews

Seems it even highlights the word just like on the website.

I remember I once found this and thought “Wow, their UX people don’t care”.

Okay I have to say this and it needs to be said right now:

Their website design has been and remains just as awful as their mobile experience. What matters most, if not the only thing of value at Yelp, is their data and nothing else.

In my not so humble opinion they have never, ever had an enticing UX or product beyond basic reviews. Ever.

Same...they'll ask me to open in their app...I hit the back button and read reviews elsewhere.

They don't just ask though, they allow no other way to access the information. Unless switching to a PC and finding your way back down the trail to it is an acceptable option. Never has been in my experience.

I wonder how easy it is to get an organization like Yelp or Flickr who already felt successful ... to understand that it is time to take their product and change it to a more, mobile facing type application.

Just trying to get small changes in small tools and processes at companies I've worked at has taken a huge amount of effort and and a great deal of "that's not how we do it ... for no reason" resistance.

SmugMug recently bought Flickr from Yahoo [0]. Their CEO Don MacAskill recently said in an interview with Amateur Photographer magazine that they're looking to remove hurdles such as the tie-in with Yahoo's authentication system but at the same time they're leveraging the need of Flickr's community for a solid desktop experience (for curation) and not necessarily to compete with Instagram as a mobile-centric app, at least in the short term.

Long term plans are obviously still in the works but the overall idea I get is that Flickr's mobile experience going forward will be focused on consumption, while keeping most of the community engagement on the desktop site.

[0] http://www.amateurphotographer.co.uk/latest/photo-news/flick...

Amazing how much it takes to do that and how long it took!

> It's always surprised me that Yelp, like Flickr, and other early services, became stagnant when the mobile transition happened ... Why didn't Yelp early on see that they could become OpenTable, or GroupOn?

I don't understand where this line of thinking comes from. Yelp has nothing in common with Flickr in regards to their business performance or somehow missing mobile.

Their business also hasn't grown stagnant in the mobile era, it has grown dramatically: 10x sales increase since 2011 (six fiscal years). If that's a failure to transition, it's one every business on earth would like to have.

2011: $83m in sales

2013: $232m in sales

2017: $846m in sales

I'm also not sure why would they want to become GroupOn? I haven't heard about anyone using that service for years. Yelp still offers a useful service for finding good restaurants that I use weekly.

Offering a single five star rating is a commodity these days. There’s just not much value there for consumers compared to every other index of businesses with a five level rating.

Why would I open yelp when my maps app has five star reviews built in?

A friend who was working with a large American CDN asked a similar question to his CEO around 5-6 years ago during an All-Hands. He wanted to know why the company was not interested in trying out an AWS like services. The answer given was - First, it was not the company's expertise and second, they were not going to spend dollars on a unknown market. No risks to the gross margin was the mantra.

In comparison, Asian companies tend to be sprawling, essentially having multiple operations. If there is a missing market, they will rather build it themselves than wait for a competitor.

Yelp's relationship with lots of restaurants was already toxic from shaking them down with the whole ads-for-reviews payola. If they tried to offer a GroupOn or Opentable product they'd be fighting an uphill battle compared to a no-name.

Why didn't Yelp early on see that they could become OpenTable, or GroupOn?

You're implying what they should have done was obvious. Given the fact that none of the 'business listing' companies pivoted to become an Opentable or GroupOn maybe it wasn't that obvious.

Seeing an opportunity, and persuading others that it's a worthwhile opportunity, and executing on that opportunity, is exceptionally hard. Yelp did do a lot of the things you'd expect, but they failed to do them well enough (eg in a way that the market accepted).

Incredibly bad product most likely caused by bad management and engineering culture, bad UX, lack of focus on user and zero desire to provide any value rather than earn money is probably some of the reasons.

They tried, they just weren't successful.

Here's a TC piece on "Yelp Deals": https://techcrunch.com/2011/06/29/yelp-deals-mobile-groupon/

I also learned the current Yelp is the pivot, as per Wikipedia, via Fortune [1].

> One day Stoppelman was looking for a doctor but had no clue how to find a good one. That gave him and Simmons an idea for a convoluted automated system in which people could e-mail friends asking for recommendations on, say, local doctors, and the answers would be logged at a communal site for everyone to see. Levchin floated the duo $1 million to build out the plan. It went nowhere. But the co-founders noticed an interesting tendency among the early users. People were writing unsolicited reviews of their favorite businesses just for fun. So Yelp switched tack. "I remember the moment that Russ said, 'There should be a way for you to write your own reviews without asking questions,'" Stoppelman recalls.

1 - http://archive.fortune.com/magazines/fortune/fortune_archive...

Reply on news.ycombinator.com