Hacker News Reader

Comment by calcsam | Hacker News Reader

This is bad, if not entirely unexpected, news. What is out of the ordinary is the public markets' reaction: the stock is down 3% in after-hours trading.

Typically when a company announces layoffs its stock goes up. More mature companies have sometimes slashed headcount on the way to a better market situation. But when younger companies announce layoffs, the markets stop assuming they have lots of growth ahead and adjust expectations accordingly -- usually resulting in a significant drop in stock price.

Remember, back in July, Twitter announced earnings beating market estimates. This was at 4pm, its stock surged 4%. About half an hour later, CFO Anthony Noto said Twitter would not see "sustained, meaningful" user growth for a "considerable period of time." TWTR promptly plunged 10%.

http://www.cnbc.com/2015/07/28/twitter-earnings-7-cents-per-...

View on HN

> the stock is down 3% in after-hours trading.

Any fluctuation within 5% is perfectly normal and totally meaningless.

Especially when the stock was up the day before

I suspect that's just people having their delusions of rampant future growth and expansion of twitter being crushed in the gears of mundane reality. These moves are perfectly fine for a normal company that intends to continue operating at about the same level as it is currently indefinitely. But people view every tech company through the lens of apple, google, amazon, etc. They expect growth, they expect new stuff, and so on.

The truth is that with modern technology you can build a company that has hundreds of millions of high-volume users and yet is not very expensive to operate. Companies, and their revenue, don't necessarily scale the way we're used to from the experience of industrial companies.

They turned off their APIs to application developers. What they should have done is ramp that shit up and figure out how to monetize the query traffic or figure out a way to decentralize their infrastructure to allow everyone to run Twitter together.

>What they should have done is ramp that shit up and figure out how to monetize the query traffic or figure out a way to decentralize their infrastructure to allow everyone to run Twitter together

So, I'm genuienly curious. How do they make money by "ramping that shit up"? I've seen a lot of companies that have 'get users' as their M.O., but I have no insight into the "profit!!!" phase. Opening their user base seems contary to "profit!!!", from a laymans perspective.

It seems to me like API/User info availability is their leverage w.r.t. profitability, much as it pains me as a potential developer who would like that information for free. How does decentralization make them (and their investors) money? I can't see a way to do both given the incentives.

While not stated eloquently, I do think there's some truth in that.

1. Open API's, good documentation and community engagement goes a long way in establishing developer trust. I don't believe developer trust is correlated significantly with revenue (directly), but it's absolutely correlated with acquisition. I'd argue that Angular or React were partially successful because of the stickiness of their communities, and that's a powerful self-perpetuating force.

2. Monetization is in the queries. Facebook and Google know this. Throwing aside all of their auxiliary services, emerging markets, hardware, et al... between 70-90% of their respective revenue is from direct advertising. If there's a way to scale or "ramp" their revenue model, this would be the portion to focus on.

Monetization of the queries doesn't appear to be a significant revenue opportunity for either Google or Facebook - why would this be different for Twitter? Not that this can't be a revenue source, but how big is it?

An alternative: I agree with everyone that they should have turned the APIs back on - they just should have made it mandatory to carry advertising as part of the feeds in order to access the feeds. They could have also layered on federated advertising for anyone consuming the API - meaning, allow those folks to show their own advertising in-stream. They simply pay Twitter a cut. Both of these may have helped them both grow the audience and advertising at the same time.

Monetization of the queries isn't a significant revenue opportunity because of scaling payment issues. Solve that, and you solve getting paid for doing something you are good at, which is throwing millions of statuses at millions of users.

I'm not getting why you think this data is so valuable or who you think the audience is - are you suggesting that millions of people would pay for the feed and it is just a problem of being able to pay?

The data is valuable only if someone has fair access to it. What an individual or a thousand may do with that data is an unknown and that is the real opportunity here, not the promise of a million subscribers paying you a dollar each. Twitter took that unknown, turned it off, and tried to own all the opportunity and all the answers. They did this because they IPO'd and then the need by investors to know what's going to happen took over even more than before. If there were answers to be had here, they'd already be doing it. I'm suggesting a move back to innovation, which is where they belong.

The desire by some individuals to make insane amounts of money on an idea are directly responsible for a massive SUCK experience for the end users of the software because those users are then forced to use the software in a way that makes the revenue predictable. I used to love to use Twitter because of all the wide variety of apps and things I could do with it with code. Now it's just unusable and dying. I would pay to make that different, but there isn't anything they offer that's worth my time and effort, so I'll go elsewhere.

"The data is valuable only if someone has fair access to it. What an individual or a thousand may do with that data is an unknown and that is the real opportunity here, not the promise of a million subscribers paying you a dollar each."

Why is the data only valuable if someone has fair access to it? Right now, basically anyone can go license the Twitter data and have access to the full firehose. And I don't know how you know if the data is real opportunity here (vs. advertising).

"The desire by some individuals to make insane amounts of money on an idea are directly responsible for a massive SUCK experience for the end users of the software because those users are then forced to use the software in a way that makes the revenue predictable. I used to love to use Twitter because of all the wide variety of apps and things I could do with it with code. Now it's just unusable and dying. I would pay to make that different, but there isn't anything they offer that's worth my time and effort, so I'll go elsewhere."

Are you offering a better path for Twitter and their shareholders, or are you offering a path that feels better, but may have a much lesser outcome for those parties? If so, why would they take it?

And while I agree with you that Twitter is dying, it's not unusable, it's just not usable in the way that you want to use it.

I'm sorry but I'm not quite sure what you mean by this?

In 2011, 96% of Google's revenue was from AdWords -- in recent years that has lowered slightly due to their diversification and cloud services, but it's still at least a 70% share of their gross revenues.

Facebook is currently ~80%.

The money is in the ads.

You and I agree.

The original poster is talking about queries in a difference sense than a Google query - he's referring to it as an API call in a "firehose" sense - so I'm buying access to the Twitter data for my application. I'm suggesting that the number of people who will pay for those API calls is limited.

But no doubt that money is in the ads - that's why I suggested that they should turn the API back on, but require clients to display the ads. Sadly, I think this opportunity is gone as they have killed their developer ecosystem.

I'm the original poster, and I was talking about ads, not metered queries. :)

[edit] and perhaps that was a poor choice of words. as a developer of high-volume low-latency exchange feeds, "query" has a bit of a different meaning to me.

Nope - the misinterpretation is my end - my bad. Re-reading your original I think we're on the same page. I think they need to reopen their API (if it isn't too late) and then just send ads as part of that package to clients and require them to display them.

Facebook's token user API system has now created the billion dollar juggernaut that is Tinder. That app is now irrevocably linked to Facebook.

You are completely right, there is a huge strategic reason for promoting OAuth API's as a large company.

Trust initially makes you decide to use something (marketing) and trust also makes decide to stay with it (loyalty). Marketing itself is mostly related to raising interest, but it frequently 'hacks' trust by using biases to establish quick and cheap trust with consumers. If a company's product is trustworthy (real trust, not biased trust) then the loyalty becomes strong with the customer. Companies who acquire other companies regularly have identified their bad habits of using too much marketing to try to get more customers and instead focus on buying loyalty through an acquisition.

I think it's supposed to work like this: assuming more users == more revenue, then opening up and promoting the use of APIs creates alternative channels to acquire new users (partner apps have their own incentive to attract users etc). Where it can hurt is when the integrating apps either don't substantially move the needle on new user acquisition, or worse they cannibalize your user base by creating reasons for your users to NOT visit your site.

IMO though the API "shutdown" was shortsighted by Twitter. If people are using a competing interface instead of your own, then maybe you should fix your own rather than just shutting down the API. Of course that's easy to say as an outsider. I'm sure the truth is much more nuanced.

Facebook's API gave birth to the billion dollar juggernaut that is Tinder.

Tinder piggybacked off of Facebook OAuth logins in a useful way, and now Facebook has further cemented their permanence as it will now be very difficult for that dating app to sever themselves from Facebook OAuth.

Note I said "or" there. Decentralizing the infrastructure would make things cheaper and might give you revenue opportunity (and growth) that you wouldn't have otherwise. For example, perhaps you get into the business of aggregating ad sales to downstream providers.

As for directly monetizing the APIs, one thought that springs to mind is charging for the firehose in a federated way. You could, for example, have a query endpoint that looks like this to guarantee standard issuance of data, identity and payment:

  GET /api/v1/status/access
  { 
    "payment_address": "13kU1qPtM3WWXrNugWfRPNM1BniMQDw9Ds",
    "rate_bytes_sec": 50000000, 
    "ask_expires_at": 1444492687, 
    "access_token": "h458Yf301",
    "ask_in_satoshis": 1000000,
    "ask_window_secs": 3600
  }

Make a payment to the address, then start querying at 50MB/sec for the duration of your valid token (in this example, 2 hours):

  GET /api/v1/status/h458Yf301
  { 
    "payment_address": "13kU1qPtM3WWXrNugWfRPNM1BniMQDw9Ds",
    "payment_received_address": "13kU1qPtM3WWXrNugWfRPNM1BniMQDw9Ds",
    "payment_received": 2000000,
    "access_expires_at": 1444499887,
    "identity_screenname": "kordless",
    "status_data_begins": [
    ]
  }

And yes, I'm proposing Twitter should use Bitcoin to monetize query traffic. Even better, charge people for posting lots of statuses, or following and unfollowing lots of people. Link ask price to signal/noise ratio.

"Note I said "or" there. Decentralizing the infrastructure would make things cheaper and might give you revenue opportunity (and growth) that you wouldn't have otherwise. For example, perhaps you get into the business of aggregating ad sales to downstream providers."

Well, maybe, but you can't save your way to $2b in revenue. Stock prices don't generally go up on cost savings, they go up because of future revenue AND profit expectations. Silicon Valley and Wall Street both prize growth in revenue way over profit. But you're correct that they might discover another revenue stream - but now you're betting a lot on the value of federation.

"Make a payment to the address, then start querying at 50MB/sec for the duration of your valid token (in this example, 2 hours):"

I think you are vastly overstating the number of people or companies that would be willing to pay for this firehose. While it may be useful to some companies, overall the data isn't likely to be enough to be a $1b opportunity. More importantly, the number of people willing to buy isn't likely to grow - so a B2B play can be a minor part of their revenue but I don't believe they'd be able to depend on it for massive growth.

"And yes, I'm proposing Twitter should use Bitcoin to monetize query traffic. Even better, charge people for posting lots of statuses, or following and unfollowing lots of people. Link ask price to signal/noise ratio."

Given the current cost structure of other services (Facebook/Instagram/SnapChat), I'd argue you're unlikely to get anyone to pay for it, and you're reducing both the size and any potential growth of your audience dramatically.

> Data licensing and other revenue totaled $48 million, an increase of 95% year-over-year.

This is from their Q1 2015 report, so data licensing and other revenue is probably $200M ARR now. Clearly people are willing to pay for access. If they improve the way they provide that, it just might be bigger than a $1B opp.

I would be willing to pay for on demand abilities to control who I follow and unfollow without being throttled.

That's approximately 10% of total revenue. I also think you have the growth incorrect - in their 10Q, Twitter said the product was growing 65% YoY. So you're making two assumptions:

- That the opportunity is there - and that the growth will continue. This isn't a given, but it is possible. I think the fact that Twitter is obviously investing more in the advertising business tells us what we need to know about the data business - that it is good, growing, but will always remain a relatively small portion of their overall revenue mix. Now you may argue that that is because they're blocking it - but my assumption is that they are investing based on Total Addressable Market (TAM) and experience. - That the problem is the payment system. There's little to support this - there's little sign of demand or willingness from individuals or small companies to spend on this at scale.

The fact that you're willing to pay for on demand controls is great and fine, but it does not mean that anyone else is willing in the same way.

I thought they did exactly as you said: the APIs are available as long as you pay Twitter for them...

> The truth is that with modern technology you can build a company that has hundreds of millions of high-volume users and yet is not very expensive to operate.

Investors had to dump more than a billion dollars into Twitter to get it to where it is today.

Uh, most of that money was from the IPO and that was probably done because founders/early investors wanted to cash in. I'd say the vast majority of that funding was not necessary to scale to where they are today.

Really? Wikipedia shows they raised over a billion from investors before IPO (including individual rounds of $800M and $300M.) Of the $800M, half went to cash out.

I'd agree though that at least technically, to deliver the core user product, this money simply wasn't needed. (It might have been needed for advertising or building up contracts, etc.)

It didn't help improve the interface much. For example, on this site the interface shows me quickly who is replying to whom. A billion dollars in and Twitter still can't do this. They tried to be the world's largest chat room and basically came up AOL.

And the UI changes they have made are demonstrably worse, such as their seemingly random threading, constantly losing page scroll position, and repeating earlier tweets.

1) Operating is not the same as building, especially if you have to figure things out.

2) Twitter reported half a billion in revenue last quarter.

>Twitter reported half a billion in revenue last quarter

And yet managed to lose 136 million dollars. While revenue is interesting, it just doesn't matter if you can't turn a profit.

How is it even possible for a company like twitter to spend a billion dollar and change in one quarter? That's enough to run a small country complete with golden bathroom for the dictator.

Time to go read about their OPEX I suppose. Anyone got a link to a good write-up?

Half a billion but...

People are expensive, office space in their 30+ offices is expensive, hundreds of thousands of servers are expensive...

How in the world do they have hundreds of thousands of servers!

Twitter is notorious for being ridiculously frivolous[1]. They also pay their employees very very high wages[2], which isn't a bad thing, but I assume that doesn't scale well.

1. http://money.cnn.com/2014/03/06/technology/social/twitter-ca...

2. http://www.glassdoor.com/Salary/Twitter-Staff-Software-Engin...

And? Aside from the fact that that's not really true, what does it matter? Twitter already IPO'd, the original investors got their money back, now it's just owners. And it doesn't require a team of 4000 to operate twitter. It requires far fewer, and as time goes on twitter will generally just get cheaper to operate.

Twitter can easily bring in enough revenue to keep doing what it's doing indefinitely. And that's totally fine. But it's unlikely they're going to grow by 10x in a few years or pivot drastically. Yet those unrealistic expectations were built into the current stock price.

That shows the company is stagnating.

Google's stock price would not be at where it is now if the company "just" stopped at search. It's everything else they kept pursuing that's allowed the company to reach stratospheric valuation heights relative to its real revenue.

> What is out of the ordinary is the public markets' reaction?

But is it? Maybe the public is realizing that tweets were just a cool fad for a while. For twitter to be profitable, I am guessing, there has to be people that want to use it, and (very important) people believing that people want to use it (so they'd pay Twitter for advertising).

Compare with Google. Google provides something momre back (gmail, search, calendar, docs, photo storage, directions + many other things I forgot about).

Twitter provides 140 status messages. There is nothing wrong it, that is very cool. But I think there is a realization that it just isn't as cool as it was in 4 years ago. And it just became 3% less cool.

I don't know, I'd be willing to bet twitter usage is up compared to 4 years ago. Also, stock price != coolness, and further coolness != making more money. Isn't that what this story is about?

But then again stock price != making more money. Particularly for non-dividend-paying firms.

Would you care to expand on this one?

It seems to me that stock price is correlated with more money: if I have X stock options worth Y dollars each, I effectively have X*Y dollars worth of stock. If stock price increase by 1 dollars, I have X dollars more than I did before. No?

I think they mean the stock price of the company is not well correlated with the amount of money the company is making

Apple often beats expectations and trades lower after the earnings call.

No, Apple often outdoes its previous quarterly results, but fails to meet lofty analyst expectations, so its price falls after earnings.

Exceeding expectations with a price drop afterwards is inexplicable.

I think I remember that happening to Google at least once. There are expectations and "expectations". Analysts' published expectations are often different than the ones they believe privately, and the expectation might be to exceed the "expectation".

Beating published analyst expectations isn't the same as beating the actual expectations of investors. One might reasonably expect the actual market expectations to be loosely correlated with analyst's published expectations, and they probably are, but its probably only loosely correlated.

If stock price increase by 1 dollars, I have X dollars more than I did before. No?

No. Only if you sell it right now. A lot of problems come from people thinking stocks are more liquid than they are.

> It seems to me that stock price is correlated with more money: if I have X stock options worth Y dollars each, I effectively have XY dollars worth of stock

Sure, but the "market" price of stock isn't really the value of every* share, its roughly the marginal price of the last/next share purchased or sold, but the more you want to trade, the farther from the market price the value will be.

And changes to the market clearing price don't tell you about changes to the supply and demand curves, just where they intersect. So, for any but a small quantity of stock, you can't tell what the real change in the realizable value of your holdings is just from the change in the market price, since there's no guarantee that a change of $1 in market price equates to a change of $1/share in the price you'll get if you try to sell a particular quantity of a stock.

We tend to ignore that when placing a valuation on a portfolio because if you don't ignore it, you can't get a valuation. But those valuations are approximations, at best.

> bet twitter usage is up compared to 4 years ago.

It probably is. But probably not as much as everything believed it would be.

Coolness is about making money in case, but indirectly. If investors see others not being interested in Twitter and not regarding it as cool, they will pull out. I understand Twitter's primary revenue tool is advertisement. That requires users to advertise to.

> Twitter provides 140 status messages. There is nothing wrong it, that is very cool. But I think there is a realization that it just isn't as cool as it was in 4 years ago. And it just became 3% less cool.

I can't tell if you're being ironic or facile, but Twitter has become quite a bit more than "140 status messages"(sic). It's a medium where politicians, dissidents, celebrities, the hoi palloi, brands, and anarchists all interact. Its a place of record quoted in headline news briefs and long-form journalism. It is used to organize rebellion and restricting access to it is viewed by many as a human rights violation.

Twitter as a business has problems, but being a fad isn't one of them.

They no longer allow sign ups via Tor, so I don't think it's as dissident friendly as it used to be.

That's more due to stupid retards abusing tor for spamming... with tor you have to draw the line somewhere.

> being a fad isn't one of them.

I'd heavily disagree with that. The fad's not over, but it is a fad. The reason it sticks around is that people have an easy way to communicate with celebrities/popular people, and that was certainly not twitter's goal. The moment something better for this comes along, the fad will quickly fade.

... in messages of 140 characters designed to provoke outrage and retweets, not to engage meaningfully with any of the complex issues of the day.

Their entire userbase could do a MySpace overnight.

Twitter is actually becoming quite popular in science. It's a great tool to hear of new papers, find out about new findings reported in conferences you can't attend, and communicate in a public forum. Don't think you would have had much of that in MySpace...

> restricting access to it is viewed by many as a human rights violation.

Restricting access to any website (except in some extreme cases) is viewed by many as a human rights violation. I don't see how Twitter is unique in this respect.

Don't make the assumption that because you don't like, or see any use for, a particular product other people won't.

While I agree with you, with Twitter the opposite seems to hold true. Many people like Twitter and see it as an integrated part of how they work and interact and that has driven the evaluation of the company. Not once has I heard anyone in the tech press or end in the financial news question Twitter.

Twitter should never have been a publicly listed company. They don't have a business plan, there was never any prof that they would be profitable. But Silicon Vally types, news media and bankers liked using the platform so: hey presto Twitter is a billion dollar company.

Twitter could be profitable if each user paid just $10 per year, but sadly I don't think people like it enough to pay up.

Twitter is suppose to be a growth company. Layoffs and downsizing are not what growth investors want to see. Thats why it went down.

> Typically when a company announces layoffs its stock goes up.

They haven't announced layoffs. The article is very clear about this.

Reply on news.ycombinator.com