79 comments

[ 3.2 ms ] story [ 132 ms ] thread
For anyone as confused as I was as to why the conclusion is 'inescapable', he seems to actually explain things in the previous blog post[1] and this one appears to just be an additional tidbit on top of that.

[1]http://brontecapital.blogspot.ca/2016/10/some-comment-on-twi...

It's the unofficial of "Jack Dorsey is fired!"
That article is spot on. It highlights the problems of the experience, and how insane the burn rate is, for something thats no doing anything new for users.
Been watching this train wreck in slow motion for years.

This has been a long time coming and I am positioned to profit from it.

Are you shorting or going long? Cause stuff is unpredictable!
Shorting.
Plot twist: he's actually hoping everyone else shorts, so that he can buy and go long ;)
And I would have gotten away with it too if it weren't for you meddling kids!
If you read the author's (good) previous post on Twitter, you'll notice this at the end:

"Twitter revenue is still rising and is running about $2.5 billion per annum. It is a website that once ran extremely well on less than $250 million in costs. (No I am not joking.)

If you can't make this have a 40 percent operating margin then - frankly you are inadequately brutal. Personally I think 50 percent is possible.

At this market cap that works extremely well for a financial buyer. Its a no-brainer even.

So expect it to be bought. By some Wall Street bastard armed with a lot of debt.

And that bastard will fire a lot of people.

If I worked at Twitter I would be preparing my resume and providing a list of really quick things that can be done to improve the user experience - with the code all mapped out. That largely involves getting rid of spam bots and the like. But unless it radically improves the user experience or monetisation and you can convince the new owners you can implement then you are out.

And the fashion-obsessed philosopher king. He is out too whether there is a buyer or not. That is necessary to save the company.

John

PS. Long for the takeout which I see as inevitable."

I suspect the lack of management focus is causing all of the waste. All of this investment in video, the various API services Twitter created and then discontinued, working on various Scala frameworks, probably a lot of internal data analysis tools, tweaks to the timeline format that nobody seemed to want...

I just don't think a lot of what Twitter is doing is necessary to running a core business. The problem is that they don't have a core business -- they have several factions pushing competing visions of a core business and none of them can gain the upper hand. They need to decide what their core business is and focus on it.

If you look at their list of active users, it's really something a lot of companies would kill for: almost every major politician, journalist, and celebrity in the US, and a similar presence in several other countries. It's just amazing that they can't figure out how to make a profit serving that kind of user base.

I should add that aside from the exceptional success of Steve Jobs at Apple, bringing a founder back to run the company is not a very good sign.

It wouldn't be surprising if the timeline stuff comes down to measured engagement going up while a modest number of vocal users make noise about change being bad.
I think that their focus has been on the experience for advertisers, which are less visible or feel negative to the end user (such as tweaks to the timeline—100% for advertisers and brands). However, those choices are probably highly influential to their revenue, and arguably they need to be working more on those things—which means even fewer visible changes that benefit end users.

Either they're working on the wrong thing, or they're not working hard enough on the right thing. I don't see too much of a split focus—they know the base they have and they need to extract value while keeping the essence of the experience they have.

Interesting how this is an example of "societies crumble when their elites want to go in different directions".

I agree that lack of focus is a big problem and wonder what those competing 2-3 visions are.

I think Jack came back because he/the investors thought he was the only person with the political clout, or moral authority, to whip people into line behind him, and get the company moving in one direction

This (and the original post[0]) sum up how I feel about Twitter. Every few months (since joining in 2009), I jump on and attempt to curate my feed to only get stuff I care about. I stick around for a few weeks, then get overwhelmed by the deluge. So I constantly curate and trim and it all becomes very tiresome.

I want Twitter to be a feed of news and opinions from organisations and people I care about, but it is far too hard to jump in and see how a conversation started or evolved. It is far too hard to filter out the noise.

Honestly, my RSS feeds give me everything I need and they've been going strong for over 10 years. Sprinkle in a bit of Apple News to fill in the blanks and Twitter quickly finds itself in the "too hard" basket.

The financial state of Twitter highlighted by this article is mind-blowing. How on earth have costs crept up so much with no meaningful improvement in product? As a very astute tech-savvy person, I cannot see anything that Twitter has done in years that have made it better. That stands in stark contrast to Facebook.

I'm sad about this because Twitter has been such an integral part of the modern internet, in democratising information. It's role in the Arab Spring alone was world-changing. It's a pity it hasn't been able to harness this.

[0]http://brontecapital.blogspot.ca/2016/10/some-comment-on-twi...

The Arab Spring thing is hyperbole to the hilt.

If anything, Facebook and Twitter acted as pacifiers. While people could use them, they stayed inside rather than get involved in the streets.

Only when the authorities cut the connection did shit hit the fan, as the masses ventured outside to keep up what was happening.

Thinking about it, i suspect Twitter was an artifact of the US mobile market. It provided a way to do mass texts at a low cost.

But as people moved to smartphones and push IM platforms, twitter's reason to exist kinda evaporated.

On top of that they had basically handed the task of innovating over to third parties via their APIs.

I still recall seeing people use Tweetdeck on desktop Windows with multiple columns set up on different topics, hashtags, and people during large events. This to allow them to track the full breath of what was happening at a glance.

Crazy thing is that Twitter did integrate the "column" concept into their service, via lists. But neither their site nor their apps allow us to access those lists in any way similar to Tweetdeck. In contrast, most third party clients will support a column view these days.

a) twitter's reason to exist is as a communication system where the primary mode is individual > public, in short form. No other communication platform exists that has those characteristics. The web is the next closest thing, but the upstart costs and knowledge required to present information on the web are greater, and the expertise needed to make your content discoverable (SEO) is higher. Blogging services get close as well, but I think the short form is important because it both decreases the barrier to entry (simplicity and low threshold), and democratizes the communication itself—my tweet looks the same and has the same limits and capabilities as CNN's tweet, or the POTUS's tweet. There's something to that.

However, it takes far too long to really understand the value of the service—took me several years before I really "got" it and my curated feed was something I actually enjoyed reading and participating in.

Third, I don't believe twitter understands the value I described above. They've made no meaningful progress in line with that value in their product that would indicate they understand it or are interested in increasing the value, or even preserving it.

b) Twitter acquired Tweetdeck, to their credit, and it still works for exactly what you describe. They really should have innovated on lists more in the actual product, though, completely agree.

> However, it takes far too long to really understand the value of the service—took me several years before I really "got" it and my curated feed was something I actually enjoyed reading and participating in.

This for me, as I've mentioned before on HN, is the main reason that Twitter is a terrible service and losing so much money. The users (especially new users) have to invest far too much of their own time curating their feed in order to make the information Twitter provides enjoyable/useful, hell, perhaps even fun.

Contrast this with Reddit, Facebook or any other social media platform. The heavy lifting is done for you. You create an account and a large amount of interesting, neatly categorised and useful information is curated for you according to your tastes. These websites are instantly sticky, and keep their users coming back for more.

If Twitter can't get something as straightforward as a fucking onboarding user experience right, I dread to think what else lives under the hood that is a festering pile of dog shit.

Add to this that a large amount of the Tweets being generated these days are automated and you've basically got a platform of bots talking to one another, with no value for us humans who aren't prepared to spend half a day curating their twitter feed.

Very soon Jack and his shareholders are going to own 100% of nothing unless some drastic changes are made to the Twitter platform.

>But as people moved to smartphones and push IM platforms, twitter's reason to exist kinda evaporated.

Huh? Did you read the articles? Can you explain how this jives with the massive increases in revenue?

Reason to exist != mindshare.

By that time, big names and whatsnot was on Twitter and you have snowball effect going. And you can observe how the shift goes from sms backed to web and app backed.

But once you mostly interact via web and app, the whole thing about 140 characters goes out the window and you basically turn into a cramped "facebook" (or perhaps and RSS feed).

At the same time the main innovation was not happening at twitter hq, but in third party apps (Tweetdeck etc) that allowed people more control over just what part of the stream they wanted to pay attention to.

Twitter is effectively coasting.

The revenue comes from investors. A lot of those investors are saying the same thing you are: "Look at all this revenue, they must be doing well!" For those people it's a closed cycle: investors see revenue increase and invest because they see revenue increase, other investors see revenue increase and invest because they see revenue increase.

But this isn't a sustainable business, it's basically an unintentional pyramid scam. Businesses need to make a profit in order for people to receive a return on investment, and Twitter has never, not once, ever made a profit.

What do you mean the revenue comes from investors? That is by definition investment, not revenue.
What the heck? The only thing accurate in this post is that Twitter has not yet made a quarterly profit... The rest is pure nonsense, most of it being quite malicious, and some being uneducated misconceptions.

Investment is not revenue. Their revenue comes from advertising.

Businesses do not need to make a profit in order for people to receive a return on investment. People will always be willing to invest in a company that shows that it may have good growth in the future. Costs can be cut at any time, resulting in an easy profit. The investors care more about growth than profit, as it is a young company.

It's really not ok to accuse people of scams without any evidence just because you don't understand how businesses work.

Regarding the original piece, I felt it's a mistake for the writer to go, 'well I used the timeline years ago and the timeline looks the same in 2016.' They built out a major advertising and metrics system in the meantime that's keeping them afloat with billions of dollars.

That said, their product execution has definitely been lacking to say the least. There seems to be some sort of mis-management that is aborting a lot of products and work they're doing internally. For example, the new 'user curated Moments' thing is a good step but they had exactly that feature, called something like custom timelines or collections, many years ago but let it languish. The new algorithmic timeline is exactly similar in some respects to their Discover tab from many years ago. It seems like the product ideas and engineering talent is there, but something's slowing them down...

I would say maybe it's 'too many execs' but then without those execs building relationships maybe we wouldn't see so many Tweets on TV and have governments all over the world using Twitter. It's all very complicated, but yes the gist does seem to be "there are more people at the company than needed."

Twitters role in the Arab Spring and its new(ish) Saudi investors are not a coincidence.
As somebody that actively campaigns for better cycle infrastructure in the UK, twitter has been phenomenally valuable, enabling me to establish connections with thought leaders in the UK and even, more vitally, from around the world. It has also enabled the communication of my own ideas nationally. Particularly this https://cyclebath.org.uk/map/ which has resulted in myself being asked to do workshops at conferences.

More importantly, twitter has evolved my thinking on urban design, housing developments, road design, and active travel/transport and am considered a bit of an 'expert' locally. I'm now looking at standing in local elections to become a councillor, with the ambition of becoming Cabinet Member for Transport and

I am aware I am standing in a big echo chamber and have crafted my twitter account to facilitate expanding my thinking in areas that I have a passion for.

Once in a while, I even tweet a picture of cats.

https://twitter.com/awjre

Yeah, that's one of the major benefits—it really can bring people together around specific topics, and democratizes and enables communication between people of all levels. I've connected with and formed relationships with experts via twitter to the point where we've met at conferences or talks and been able to strike up a conversation based on our latest twitter discussion; they knew who I was and were interested in talking to me. Not sure how I would have done that otherwise.

You're right it's an echo chamber, but echo chambers can help you be heard...

But not necessarily heard by people outside the echo chamber.
That's okay, I wouldn't be heard by them anyway :-)
> That stands in stark contrast to Facebook.

Although I agree that Facebook might have changed more than Twitter I'm wondering myself the same for them. I use Facebook since 2008 or so, and the changes/improvements were very marginally for me. The biggest thing was timeline (already long ago). Since then I only discovered minor stuff ( like "yay, after several years I can finally select emojis in the chat bar", self-playing videos, different like-types, after years finally another search field, ...), but nothing which really answers why Facebook needs that amount and growth of engineers from a pure user perspective.

What has worked well for me is creating constant churn in who you're following: follow anybody that you come across in retweets or suggestions if you think they might be interesting. Then be ruthless to unfollow anybody when you see the (1st or 2nd) tweet you don't care about.
Sounds like a lot of work to me for little in the way of value. I don't understand why Twitter isn't doing more of this automatically for its users.
Why does he believe that only one in three employees will be keeping a job post-acquisition? Not that the engineers will have to worry.

They have a very talented engineering team that will be sought out by every headhunter if that's the case but it all seems baseless.

Twitter is going to switch into maintenance rather than growth mode. You don't need as many engineers to do that.

If they cut their operating costs down, they might actually make money.

Those that leave, by choice or not, will go on to build up other places that don't yet have their product.

From my perspective, Twitter has been in maintenance mode for years. What significant changes have been made that requires thousands of engineers? I'm being serious here. The interface is still the same clunky mess, the home page has been rearranged a few times, and now we have even more convoluted rules as to what counts towards your 140 character limit.
Yet they've managed to massively raise their r&d expenses.
>They have a very talented engineering team

Do they? What have they been doing all these years? As a user the product just seems to get worse and worse for me and despite the massive dev team they actually outsourced the mac client...

They haven't been building things for you. They've been building things for advertisers.
It's possible; a very talented engineering team with horrible management that holds them back is still a very talented engineering team.
this is where an mba comes in handy... he's basically saying that the company is spending too much on prople (among other things) so a private equity firm (the financial buyer, as opposed to a strategic buyer) will see an opportunity to buy the company and fire a bunch of people (including engineers), which will improve the profitability of the company. because valuation is the net present value of projected earnings, and because pojections are typically based on recent performance, the company will have a higher valuation and will sell at a higher price. being armed with debt means that the PE firm can buy all the equity for a fraction of the price and make out like bandits on the sale.

so the quality of the engineering team doesn't factor in at all. =/

favourite line from the article -

"the plural of anecdote is data"

nice.

Yes, very clever - but also, by now, a venerable quip.
Anyone else get tripped up when the underlined "He simply does not deserve the job" turned out to not be a link?
Yes, I didn't understand the article so this was the only thing I thought about.
Have they ever tried to charge to follow certain accounts? Say someone popular has 1m followers, all new followers would need to pay a subscription fee to receive all their tweets. It's a bit like the old CelebIndex. The more followers someone has they more the tweets cost. I'm thinking something like 1 or 2 pence per tweet.

This is probably a terrible idea.

This would have bankrupted Ken Bone's grandmother overnight.
A celeb/politician/etc's popularity is what brings regular people to twitter. To charge the celeb would be counterproductive and probably move them over to a competitor for good.
Sorry - I meant that it would cost general users say 50P a month to receive tweets from say a major A-lister, but 25p to recieve the tweets of a C-lister. People could get them for free, just delayed by say 30 minutes.
Not being a Twitter user myself, it's hard to comment on how it is. However, a number of their open source projects [0] are really really good.

[0] https://github.com/twitter/

> When Facebook had $1.974 billion of revenue it had $1.008 billion of income before taxes.

> Twitter is kind of different.

> When Twitter had $450 million of operating losses and $533 million of losses before tax.

For a UK based site, the English here is incredibly poor. The first sentence implies a comparison to when Twitter had a similar level of revenue.

Making sense of it:

> Facebook had $1.974 billion of revenue and $1.008 billion of income before taxes.

> Twitter is kind of different.

> Twitter had $450 million of operating losses and $533 million of losses before tax.

If you want to keep the parallel structure you would want to compare Twitter's revenue and their income (loss) before tax, just like he did for Facebook. That tripped me up too.
The 'arab spring' is nothing but the US and other vested interests stirring up trouble in these countries. Revolution does not happen by itself. It takes money, effort and organization to get otherwise passive citizens to raise their head above the parapet and more often than not it's to futher the organizers interests.

Libya before and Libya now is a perfect example of a complete disaster for the people and another crime against humanity.

Saudi Arabia, the most repressive regime in the region and chief promoter of terrorist ideology and the one place that requires a 'spring' is the western world's best friend in the region and remains 'strangely' but predictably untouched by the 'arab spring'. This is just unadulterated propaganda.

We have far too many of these 'springs' repeated over and over again in the last 100 years to know what's happening and the modus-operandi and yet every story on facebook and twitter naively persist in rehashing this propaganda.

These interventions under the cover of 'human rights' are always to further some agenda and given the massive upheaval and suffering they cause to populations its difficult to see how any believer in humanity can't continue to support these blatantly self serving actions.

> Revolution does not happen by itself

I actually think revolution needs to happen by itself. The overlap between the skills needed to organise armed insurrection and to govern are similar.

Issues come up when still-formentig revolutions are prematurely encouraged. They may succeed in toppling the prior regime, but they never forged a shared identity and set of norms fit for rallying, organising and deploying force and order - the bread and butter of nascent governance.

The US generally didn't even support the Arab Spring, and actively worked against it in Egypt, a country with nearly as tight a relationship with the US as Saudi Arabia, that receives nearly as much military aid from the US as Israel. Everything is so much more complicated than you're making it out to be here.
800Mi in R&D for Twitter? Almost the same figure for sales & marketing?

Really, what are they doing? This makes no sense

OK, Apple is @ 10Bi per year approximately http://appleinsider.com/articles/16/04/27/apple-rd-spending-... (but developing hardware and silicon is not cheap

R&D is a fuzzy bucket. A lot of ongoing maintenance can wind up there. (Improves the optics on unit economics)
Yes it's crazy; if you add up R&D + Sales, the total is $1.6B; their revenue is only $2.2B... And that's without factoring in the costs of simply keeping the business running - They still need to pay datacenter costs and salaries related to the ongoing maintenance of the site.

When you look at the billions of R&D dollars which were spent over the past few years, you have to wonder where all the money went. I can't remember the last time Twitter released an innovative product... If ever.

It's not like they were spending their R&D on high-risk ventures like finding a cure for cancer or building a quantum computer. With the kind of R&D they do, you expect to see results at reasonable intervals...

I wouldn't be surprised if there is some kind of fraud going on.

Twitters manipulation of this election cycle will cost them large swathes of users in the long run. You cant set yourself up as a communication utility and then get so involved in politics, it is bad for business.
Twitter is the classic example of waiting too long to IPO. I understand there's a sort of cultish wisdom to delay an IPO for as long as (legally) possible; however, the underlying sustainability of a business either "just works" or it doesn't. The problems that plague Twitter today could and probably should have been caught at least a couple years before it IPO'ed at the very end of 2013. IPO'ing in the public exchange basically teases out sustainability problems. And regardless of what you think of American capitalism, the public markets are the best available detectives we have to investigate sustainability problems.

When sustainability is iffy, owners of the company get their vote, and then they get to implement leadership that does care about and pay attention to the numbers. Absent IPOs, a lot of startups are nothing more than glorified boys' clubs, partying on the dime of oblivious investors. It's lousy that the culture of so many companies in startupland hate the bean counters. The little numbers can usually tell a lot.

I don't quite understand your point in the context of twitter. It went public in 2013 and has been under investor scrutiny for the last 3 years. Even so, the sustainability problems have persisted despite the glaringly obvious balance sheets and steadily declining share price. This is a problem of investors (public and private) letting the board get away with hemorrhaging money for too long.
Jack Dorsey became CEO of Twitter in October 2015. The numbers in the article are for all of 2015. Should he have been able to fix the financials in the first three months back as CEO? I feel like I'm missing something.
Having one person be the CEO of two different companies at the same time just feels like an incredibly stupid idea.
Arrogant but necessary. That it's tolerated by investors should give you a sense of how urgent/dire things are.
Question: what happened in 2013 that made Twitter's operating costs increase 8x?
According to Google, Twitter has 4000 employees.

The total cost of R&D and sales is $1.6B per year.

Do the math:

1.6B/4000 = 400K per employee!

^ And that's assuming that all 4000 employees are working on R&D or sales.

That number tells me that there is something REALLY wrong with the company.

What kinds of people get paid $400K per year?

The cost of R&D covers more than just salary
Does that include employee costs that the company covers, such as health insurance, retirement accounts, etc etc?
I couldn't agree more and I'm glad someone is talking about this. I also read SEC reports (Ks and Qs) and I'm appalled at what I'm seeing.

I actually think it's bigger, though. Look at the original ethos of Silicon Valley: two mid-career guys in a garage working off-hours to get something going. There was a culture of cost-consciousness, a scrappy resourcefulness, built into these companies from day 1. Hell, they could've set up shop in San Francisco, but they opted for the suburbs instead, because they wanted to make it work. Quoting Paul Graham, in 2005: "For most startups the model should be grad student, not law firm. Aim for cool and cheap, not expensive and impressive." [2]

Fast forward 10-20 years. Have you been to the Twitter office? It's next to NEMA [1], an apartment complex where a cool $4000/month will get you 1-2 bedrooms. A juice from "The Market" underneath their office will run you $8, to go with your $15 sandwich. Everyone there seems to be OK wearing $100 shirts and expensive shoes. The office itself is impeccably painted and they have art all over the walls.

I don't care about the money per se; the issue is focus. If all the energy dedicated to all this flashiness was dedicated into getting actual business results, Twitter might be in a different place today.

Now, the standard argument against this is, we need to be in SF, and spend all this money, in the name of attracting the best talent. Perhaps, but, you have to wonder, when companies are climbing all over each other to get developers at $150-180K base plus cash bonuses, stock, etc., the case for being in SF gets more difficult.

What I'm really saying is two things: (1) Tech has changed...a lot. Gone are the founding myths of Amazon's door-desks and Bill Gates, whose famous restaurant was a burger place. Now we all eat at two-star Michelin restaurants at $150/plate.

(2) I wonder what this says about SF long-term. Sure, the good people are here, and sure, maybe it'll just be a slight correction. But I really think the case for locating a company elsewhere grows stronger each day given how expensive and difficult it is to run a company from here.

[1] http://www.rentnema.com/floorplans.php

[2] http://paulgraham.com/start.html

On your points about grad-student vs. law-firm: Yeah, all the 'real' hackers are out now. Just look at Google Maps, that thing was better 10 years ago than it is today. Complain all you want about the proliferation in devices, but the fact remains that it was better a decade ago. I say that to mean that the actual hackers and the neck-beards that care about the product are gone and have been replaced by bullshit artists. Reasonable people, people that have mortgages and get other people to change their oil. Let me repeat that: Google is full of BSAs now. Just look at the Burn, it's all cigarettes and blow now, not pot and LSD. It's a networking event! How in the heck did nepotism, sorry, I mean networking, get out to Gerlach? Where have the real hackers gone off to? Somewhere cheap, aka not the Bay Area. Look at Knoxville or Tallahassee, other Southern cities, Albuquerque and Reno, or Albany. Places with some sort of grad school (not just comp-sci, look at all that CRISPR-CAS9 stuff, oh man!) that does not have crazy rents nearby. A place where you can't tell if they are a grad student or a homeless dude. The 'real' hackers are out there being busy building things, thankyouverymuch, and not giving a damn about juices that someone else makes for them. It's not that Tech has changed, those toothy basement dwellers are a part of the human genetic code, it's that SF has changed. It's mature now, there are firms and suits and spouses that need to be kept. SF will continue along this path, it is not what it once was and really, all things evolve.
(comment deleted)