Ex-Netflix here. When I was there we had two metrics we cared about: hours watched and retention. Whenever we introduced a new feature we carefully measured it's impact on those two metrics. Ultimately, retention was the most important metric, but hours watched was a leading indicator for retention.
I always had a bit of a moral dilemma working at Netflix because my goal was to get people to spend more time watching Netflix, which I didn't really think was good for humanity.
Either way, it was nice having very clear goals. Netflix was a great company to work for.
I just cancelled Netflix for this exact reason. I find the entire platform to be anti-consumer. Auto playing videos with sound, automatically playing the next episode to encourage binging, changing cover art to give you the illusion of fresh content always arriving, television series that are 10+ episodes long for no reason other than to extend hours viewed.
Don’t get me wrong, Netflix does have some good content, but I finally realized that the way the platform is designed is gross.
What's seriously making me consider cancelling my subscription is the fact that they autoplay trailers on the kids channel. I don't want my daughter getting addicted through the ML algorithms on the site, period.
What's seriously making me consider canceling my subscription is that it's slowly but surely becoming a worse service over time.
It's clear that rising licensing fees are eating it alive, and its response, to create more original content, is mostly failing. Original Netflix series and movies are, by and large, fairly awful. The exceptions are notable but rare.
Next year, when the mass exodus of Disney content occurs, there will be a noticeable drop in the number of good movies on Netflix -- I'm predicting that I'll likely be one of many canceling Netflix in favor of Disney's streaming service, and that will be an inflection point in the beginning of Netflix's end.
Netflix original content is not enough to sustain it, and soon it won't have much else.
>Original Netflix series and movies are, by and large, fairly awful.
I've found it's not worth watching anything on netflix, originals or not, unless it's something I already know is good or if the rating is above 7/10 on a couple different review sites(even that's fairly unreliable).
I also find a lot of shows start good...are full of story dragging filler in the middle then end either in some kind of cliffhanger or just don't bother wrapping up most of the story.
I understand why they end that way....but most of the things i've watched like that don't end up with a second season...it just ends up feeling like I wasted a bunch of time with a story that will never conclude.
Not to mention the strange recommendations, the way netflix alters cover art depending what demographic they think you are...and something strange I asked about but was told I was crazy for thinking but..
But I have suspicions netflix associates unrelated accounts used on the same network in some way.
My girlfriend had been watching a series on netflix using my account, under my name on the roku app through the television.
She recently got a new laptop that came with a netflix app and used her parent's account to sign in. The two accounts have completely different information associated with them, my account had never been used on her computer, her parents account had never been used on the tv and thr show had never been watched on her parents' account yet when she started the show on her computer it started off at the exact minute she'd left off on the tv.
The only two things the two devices had in common were that they were used on the same network in the same house.
But I was told by multiple people I just have been mistaken and netflix doesn't work like that so what do I know?
There is too much awful garbage on Netflix. I don't want to waste my time with it, but it's right there--presented front and center. I can't count how many times I've started something only to exit it within the first few minutes. Or never made a choice in the first place.
I'd prefer a service with a few really good shows. Something a la carte.
> the suggestion algos (for me at least) have gotten worse.
They were ever good? At least for me, and with spending a lot of time rating, telling Netflix which genres I liked and which I disliked resulted in Netflix recommending I watch series of genres I explicitly told it I didn't want.
Their recommendations always were as bad as Amazon's which is to say the dumbest possible implementation (at least for usefulness for me as a viewer, I assume they both have reasons for it)
> Original Netflix series and movies are, by and large, fairly awful.
I disagree strongly. There are quite a few Netflix original shows that I like. The wife and I just finished watching Travelers last night, and it the 3 seasons were fantastic.
Why bother making that distinction when Netflix doesn't themselves? If they're going to put everything they license on the same level as Stranger Things or Roma, doesn't that show that they want all those "Netflix Originals" treated the same in the eyes of viewers?
At a base cost of 8 bucks a month, do you see a large number of people canceling it, vs. supplementing it with something else? In my case, I'm paying for netflix, hulu, amazon, and youtube. So adding one more item would be similar to changing up a cable tv package.
You probably make a lot more than most Americans (I'm assuming you're American since from what I hear Netflix content is really bad for everyone else).
Everyone I know already shares accounts because it's too expensive to have all of them. Adding another streaming service is hard to swallow. Does anyone have an idea how much Disney will charge? I feel it won't be as cheap as the entry level Netflix sub but who knows.
All of those add up to under $50/month. When I was a kid, everyone around (in small town midwest) had a cable package for over $100/mo and many were by no definition rich.
when you subscribe to cable tv, you pay for content and delivery, so i think it is far to compare the price of streaming service including the price of internet
While I get that that's more money than a lot of people have to spend, all of those combined are cheaper than a typical non-basic cable or satellite plan even today - after they know they need to compete with streaming services. It's a lot cheaper (especially adjusted for inflation) than those services were 10 years ago. So anyone that could afford content then is probably able to afford it now, but they have better options - so they might choose not to.
I've checked out Netflix in Russia, all the current and 1-2 y.o. shows are available with some random films. Also, the only language available to choose (except English and original if it differs) is finnish!
$8 a month is for the standard definition plan, which I doubt many people have at this point. The most popular (HD) plan just went from $11 to $13.
Subscribers might add Disney+ without dropping Netflix, sure - but all those services will reach a saturation point eventually, and subscribers will eventually have to choose which services they want to drop. Netflix still brings more to the table now than many of its competitors, but it's on a long, slow downward slide that will almost certainly accelerate this year.
I think they'll be OK this current price increase, but not the next one. I'm sure they have a good model on cancellations after price increases and they'll make up for the loss with the increased price. But I think a lot of people are going to start examining if they want to continue paying for Netflix every month without thinking. More people will likely cancel and re-subscribe periodically when they hear about a series or movie they want to watch. And the more competition from other streaming providers, the more people are going to be examining their total spend per month on streaming services, which if you have a lot of them, is beginning to approach a traditional cable television bill.
The price increase really wasn't the main driver for me; it was more that Netflix continues doing the exact same thing -- manipulating its users through user interface trickery, and continuing to spend money on content that is not very good. The more shows, including original series that I pass on, the more I realize that I'm paying an increased monthly cost to finance Netflix's overspending on B-rated content that I won't even be watching.
I also think Netflix took away the wrong thing from the Bird Box "success." Yes, a large percentage of Netflix subscribers watched at least 70% of the movie, but I think it was 1) to see Sandra Bullock, who is a great actor and 2) it came out over the holidays when you often want something in the background to watch because there isn't a ton of things to do that the entire family can do together. For me, I'm a fan of Sandra Bullock so that is why I watched it, but the rest of the movie and story was not that great or anything unique.
After I cancelled Netflix, I signed up for YouTube Premium instead, and I feel much better about it. You can turn off auto-play and video previews don't have sound. YouTube music isn't quite there yet, but it's getting better, and the iOS app is solid. And their recommendation engine is actually good. Plus I like that I can watch ten or twenty minute videos on an infinite variety of things instead of committing ten hours of my life to a Netflix show.
If Netflix had any faith in their own content, they would introduce some kind of impartial review system (either user-based, or Rotten Tomatoes, etc). Netflix's "% match" system is a complete joke; I frequently see 90% matches and start watching and go "why in the world was this recommended to me?"
Disney will also be launching their streaming service this year, and it will be rumored to undercut Netflix pricing. Plus Disney will now have all the Fox content, some of which is on Netflix and might be pulled down the line. I think Disney+ will also be more appealing to people with kids over Netflix, which tends to be more adult-themed.
The other issue they're going to run into is that they have almost reached peak saturation in developed markets. In order to continue getting new users in developing markets, where those people will be more price sensitive, they cannot keep raising the price. I wouldn't be surprised if they begin segmenting the pricing by region, or creating lower cost plans that only have limited content, or more restrictions.
Ah, I just found it, but it's on the website only under "playback settings." You cannot turn it off from the app clients, or at least not from the Apple TV one. I think this still shows an intentional act by Netflix to make this hard to do, because most people are not using Netflix from a web browser / computer.
You also cannot turn off the auto-playing previews with sound.
For me, turning off auto-play worked on both my Roku and Visio TVs. I actually turned it off on a per user basis.
As for the auto-playing previews, my partial solution is to hit the mute button when browsing the Netflix home screen. Still gets annoying when trying to read the text description of a show, but it disappears from the screen too fast and replaced by the preview.
I've also had the situation where the trailer would be for a tv show for older kids and it scared the crap out of my daughter (likely we don't let her watch on her own so we could shut it off). We are trying to limit her exposure screen in general, but commercials in particular, which is why we thought netflix was a good option...
Yeah I have a similar concern for my little one. I do find Netflix better than YT Kids though. Just like adults will watch others play video games and enjoy it, toddlers apparently like to watch others eat candy and unwrap presents.
I think he is not complaining about more content being produces, but the same content being diluted in more hourse than needed in order to inflate some metrics.
British TV, especially BBC and Channel 4 have always been exceptionally good at knowing when to kill off a series at the right point. Just compare the difference between something like the UK Office vs the US Office or the BBC version of Sherlock vs the US version called Elementary.
Another example, the BBC Series the Bodyguard (a different concept but similar enough) or Israeli Homeland vs US Homeland. The US has made the mistake on a number of times of taking a great concept and dragging it out too long. Maybe one of the few that was justified was House of Cards, where the longer US version was (arguably) better than the UK version...
The US version of The Office is apparently by far the most watched show on Netflix (US). Even if you acknowledge it went on probably one or two seasons too long (which isn't an uncommon opinion, to be fair), 7 seasons is far longer than most BBC shows last.
And, as another poster noted, I would argue that a lot of people think the BBC version of Sherlock has definitely gone on way too long.
I think British TV has actually started taking influence from American TV over the past decade or so (or at least realized that Americans are watching and have started to cater to those audiences). Downton Abbey I feel also similarly went on probably a couple series too many.
Wow, you mention the one time where a US show manages to give a character depth while the British version utterly fails at that and instead decides to keep it alive with cheap gimmicks.
Which of the GP's examples are you talking about? You could be agreeing with them on House of Cards or disagreeing with Elementary/Sherlock or The Office - it's not clear which.
I'd argue the same with House of Cards; the longer format was enjoyable at first but it definitely went on too long, and I think lost some focus along the way.
>Wait, one of the reasons you've cancelled is companies that make media to get people to watch... make... MORE of it?
it's not a bad reason at all. I prefer to buy media that respects my time and does not painstakingly try to get me to maximize my engagement with the service, because that's probably not very good for me.
Yes, I prefer paying for a service that respects my life and my time instead of one that treats me like a junky looking for another fix.
Back in the day Lay’s potato chips had a campaign called “bet you can’t eat just one” which clearly encourages overeating. And yet today we see companies, like Netflix, embracing binging as some innocuous thing. We say binge eating and drinking is bad now, but binge television watching is universally ok?
Padding and pointless busywork are a common complaint in video games. Personally I'd rather they cut the fat and leave the experience shorter but more striking and memorable.
A lot of TV shows have the same issue, Netflix originals (I'm mostly talking about the Marvel ones here) seem to meander a lot and could use some ruthless editing.
Can definitely tell scripts that honestly only have enough content to be a 90 minute movie at best are being extended and shipped as series for these metrics.
There is a setting in the account to be able to turn off auto-play, after doing that I noticed I drastically reduced the time spent watching videos. That little bit of effort just to hit the play button made a huge difference to me.
Only on the website, which almost nobody I know uses except to adjust billing information. Why wouldn't they expose this choice in the app client? Amazon Prime exposes it in the app. YouTube exposes it in the app.
That is clearly a deliberate move by Netflix, because they know it will reduce binging and reduce the metrics they so covet, like total hours watched per user.
My favourite thing about Netflix is how notoriously bad they are about fixing bugs in their Windows 10 player.
Hit pause at the end of a show to stop the playback, maybe for a quick bio? Fine! Then - you might have moved around in the house by now - it suddenly starts playing again!
Why? Because the auto play next episode timer isn't cancelled when hitting pause. Once it's counted down, playback of next episode starts irrespectively.
That's some genius systems development right there. I'm sure a lot of binary tree reversals were tested for to get to this level.
I wish the app would manage to adjust the display refresh rate to match the content's framerate in fullscreen. Just recently I tried to watch a highly rated, cinematic sensation like Roma on my projector and was surprised that I had to guess and set the proper output settings manually.
Setting the proper refresh rate is a solved problem for any decent software or bluray player but it seems to be completely ignored in most Netflix clients on countless different devices. (I hear the Apple TV does this well though)
I hear you! Also, until I figured out to engage Windows' tablet mode from the Win+A menu, when on my primary computer with a curved 21:9 I'd get a wide aspect movie squeezed into the 16:9 portion of the screen - a tiny slit of image surrounded by black frames. Ugh.
I don't have any experience with those wide screens and how they handle movies but I'd imagine they are very much suited for that kind of content if the player handles it properly.
Concerning the frame rate issue, I was really surprised that most people just don't care or notice at all. After some research it turns out that only about a dozen of Netflix clients on various bluray players, streaming boxes and TVs handle proper refresh rate switching. If your device's Netflix client supports it is apparently pure guess work and anything but certain. Instead they just seem to rely on the display's frame rate interpolation algorithms to fix an issue that shouldn't be there in the first place.
I can live with that for mindless, forgettable content, but if you're trying to sell your service as the future of cinema and put effort into high prestige projects like "Roma" or "The Other Side Of The Wind" maybe they should get the basics right in order to offer a worthwhile big screen experience.
Yep 34" 21:9 screens are rather ideal for watching movies, when at my desk in its sweetspot with headphones. Not quite as comfortable as couch and the Samsung 65" big screen for multiple people though!
And then: I read you, again. The attention to proper detail just isn't there - unless you can wrap it up with a hot buzzword such as _HDR_!
The interesting thing about these sorts of goals is that there’s no accounting for saturation of customer engagement and diminshing returns.
When your only goal is to make a graph go up and to the right, what does a plateau mean? And then after the plateau?
I’ve had a number of managers over the years, and some of them could only think in those terms, and refused to think in other ways. Decades later and those jobs are long gone, and whether there was any respect between us is long since irrelevant, but by now, those kinds of goals often come across as strong signal to me of a built in concept of disposability. Proof arrives with the fact that I’m no longer at those places.
If the graph is only allowed to go up and to the right, your days there are assuredly numbered.
Infinite growth is unnatural after all. On a side note, it's fascinating hearing investors worry about Apple's growth slowing (possibly declining even) while throwing around figures in the tens, if not hundreds of billions
You’ll get to a point where the Netflix staff would have to down tools and watch Netflix all day to get get the last bit of growth! I guess that’s after they’ve convinced all mothers to stick their newborns in front of Netflix.
Wouldn't the ideal Netflix customer watch 0 hours (thus consuming none of the company's resources) while still paying their monthly subscription? :) OTOH active viewers probably refer more friends and family to become new Netflix customers.
Also ex-Netflix, and no, that was not ideal. We worried about those folks because they were the ones most likely to quit and the least likely to tell others about the service.
The cost of actually delivering the video was so tiny it didn't matter. To put it another way, you could watch for the entire month continuously and it would make no difference to the bottom line.
The videos were fixed cost licenses, not per view, so it didn't matter who watched it or for how long. The actual cost of sending bits over the wire was trivial.
> The cost of actually delivering the video was so tiny it didn't matter. To put it another way, you could watch for the entire month continuously and it would make no difference to the bottom line.
I think cable tv is much simpler to deliver as you only have a few channels. Netflix has tens of thousands of videos so they can't just broadcast everything (I don't think cable TV is broadcasting all their channels either, I don't know how they do it in practice).
> (I don't think cable TV is broadcasting all their channels either, I don't know how they do it in practice)
Actually they do. Just a whole lot of MPEG-PS, several of which multiplexed into a single frequency channel (sometime called transponder channel, even in the context of cable, although the term comes from satcom), and then on the order of hundreds of frequency channels between 100MHz to 1000MHz (roughly speaking, it depends on the regulatory domain, frequency allocations, and doesn't stretch all the way to the limits of the range).
If the cost is so tiny it doesn't matter, why does Netflix have that annoying "Are You Still Watching '<showname>'?" pop-up that's shown after a few episode have auto-played? It's annoying when binging shows, and I always assumed it was a money-saving measure.
It's in case you've fallen asleep, so you don't suddenly wake up to find you've "watched" three seasons of whatever show you were on. This is only displayed if there's been no input on the remote for some number of hours.
I think it does this after 3 episodes of no input. It's a compromise; they don't want to do this after every episode, but they also don't want you to wake up having watched every season of the show.
It would be nice if it was configurable, though. In fact, it might well configurable; I've never really checked, because the current setting is fine by me.
> "The cost of actually delivering the video was so tiny it didn't matter..."
I don't want to get off topic but isn't this the ISP argument __against__ net neutrality? That is, they (and their collective) customers are bearing the cost of delivery, not (e.g.) Netflix.
The cost to an ISP of delivering a single bit is also trivial - the capability to deliver a bit very quickly is the cost - which is what their customers are paying for.
Net neutrality is about being able to charge both ways.
It's part of the US government but does not receive taxpayer funding.
"Zero tax dollars used. The Postal Service receives NO tax dollars for operating expenses and relies on the sale of postage, products and services to fund its operations."[1]
"The USPS in its current form runs like a business, relies on postage for revenue and, for the most part, has not used taxpayer money since 1982, when postage stamps became “products” instead of forms of taxation. Taxpayer money is only used in some cases to pay for mailing voter materials to disabled and overseas Americans"[2]
I'm no expert on any of this stuff. From what I can tell from searching online, the USPS is required to fund their own pensions[1]
> "for the most part."
This was explained. The government pays to mail voting materials to some people. How is that different from any branch of government buying some stamps to mail some letters and packages?
Their "monopoly" gives them the exclusive right to put things in mailboxes. In return, they are legally required to serve every single part of the country equally, at the same price. They also can't set their own prices.
I have no earthly idea how much this "monopoly" costs taxpayers and it hasn't stopped UPS, Fedex et al from providing plenty of competition. I think taxpayers, especially in remote and rural communities, are coming out ahead in this bargain but again, I'm not an expert.
In any case, none of this has anything to do with net neutrality which is where this discussion began. Pretend USPS doesn't exist - does UPS or Fedex charge the sender and receiver for shipping the same package?
How do they not pay for the cost of delievery? They pay for internet connectivity like everyone else. And they even have caching servers installed at most ISPs to optimize their bandwidth.
The issue is that ISPs want to double-dip. The cost of delivery is already paid for by customers, there's no reason to charge the company a separate fee. Why should it matter where a bit originates from as long as it's going to a paying customer?
While the cost of video delivery is negligible, there are definitely increased costs for delivering 4K video.
It takes more storage on the CDN box, and also takes more bandwidth on the outbound network port. That means you need more CDN boxes and more hard drives, which means renting more space and power and cooling. Also, you need to be closer to the customer for the 4K video since it's so high bandwidth, so that means more CDN deployments.
You also have to get the 4K video to the CDN, which is more costly because of the extra copies and the large files.
You also have to create all the 4K video files for all the devices, which takes a lot more processing time in the video pipeline.
You also need higher res files from the studio, so of course they charge more for that since it costs them more to move the large files around, store them, and create them in the first place with better camera equipment and editing equipment.
The plan that gives you 4k also gives you 4 parallel streams instead of 2 (or 1 for the cheapest plan), so it's popular with people who share the account with others. But 4k/HDR is also a value-add that they charge for, but it doesn't mean that it costs more to provide it.
4K content is better than HD content - to someone, on some axis of "better". How is that an "artificial increase" then? The customer is receiving a higher quality product for a higher price but isn't forced into the more expensive plan; it seems perfectly fair.
You missed another important piece of the comment you quoted: "The videos were fixed cost licenses, not per view, so it didn't matter who watched it or for how long"
In other words this was about the marginal costs of delivering content to additional viewers, not fixed production costs.
The customers are paying for the service. They bear the cost.
The idea that ISPs bear the cost of Internet traffic which benefits tech companies is at the centre of their agreement against net neutrality, and it's a fallacious argument.
The ISP is free to decide the price and the quality of the service offered. That's how they recoup their costs, by charging their customers the correct price for the service.
Maybe I was not clear enough, I was not defending ISPs. What I meant was that they will bear the extra cost if every subscriber maxes out their bandwidth usage so the cost (and prices therefore) will go up.
What customers pay is an averaged out bandwidth usage from my understanding.
The customer has already paid for a dedicated bandwidth. If the ISP has used an outdated model that doesn't account for current usage, that's their problem. That's how business risk and models work.
They can, and should, respond by modifying their plans accordingly. Once the consumer's contract is up, renew them with a more modern usage model. Until then, you eat the cost for not doing your only job.
The theory that "we didn't build in a margin of error and therefore someone else is responsible" is dangerous to society.
ISPs SHOULD see losses. Their investors aren't entitled to guaranteed profits.
But if everyone uses their connection in inefficient ways (such as letting their video streaming run for hours unwatched) everyone will simply end up paying more at some point.
Only in the United States, because of the lack of ISP competition.
In Europe, as usage goes up, prices stays flat, because ISPs there know that the cost of a marginal bit of data is almost nothing, and they actually have to compete with each other on customer service and price.
The idea that everyone should simply pay the same regardless of use is wrong.
Cell phone plans have data tiers and pay-per-use. There's no reason cable companies can't do this either, except it prevents them from fudging costs.
I probably use as much bandwidth as anybody else, and I'd prefer to be accurately charged for it rather than have my provider pull shady net neutrality things or try to become an ad network or buy a media company to "bundle" their services or sell my data to third parties to juice their income.
I don't necessarily disagree but go back to any of the discussions on here about data caps and read all the screams about stifling innovation, etc., etc.
I see both sides. Honestly, I think there was a better argument for it when the heavy users were a small percentage of users downloading a bunch of "Linux distributions" on bittorrent. Today, it's more likely to be a family that uses Netflix a lot.
Metering would change usage for better or worse and the fact is that a lot of domestic phone plans don't have caps, or at least meaningful ones any longer, although they may throttle data at some point.
I have often wonder why Netflix doesn't uses its Fast.com to offer CDN services.
They could also have offer their Encoding System as Services. I upload a video file, sets a few requirements and I get Netflix Quality encoding in return without me fiddling anything.
The answer to both of those is "because Netflix is a video delivery service, not an infrastructure service". It would be a whole new skillset for the company to offer such a service. Running those things, and selling them, are two very different things.
I guess so, It is just a nice idea. It was one day I notice how Fast the response time were in Fast.com compared to others. I am not even sure if Google Edge Node in the ISP even works with the GCP CDN, because their performance don't seems to be leading the pack.
The Market for CDN and Online Video Encoding that Netflix could get are properly in the sub annual $1B range, may be too small for Netflix to even consider.
I just thought the more traffic passing through those FreeBSD Server the better.
It's only incompatible when ISPs actually act on their natural preference for consumption to go towards in-house edge caches. It's not that much different from the ancient art of configuring http servers and caching proxies to get along well (back even caching proxies were still a thing).
So Akamai, CloudFlare et al. will locate servers in a large ISPs data centre. TIL, I just assumed once a resource had been cached then they'd effectively have that collocation (logically); I'm surprised if avoiding the cost of the first hit really saves anything for static content.
Obviously anyone streaming video, it's different.
I wonder if BBC iPlayer (major streaming service in UK) collocate with ISPs.
Will those servers have a local (to the ISP) IP, or do the Akamai IPs get physically routed to the AS?
It is a CDN, but what's new is that the openconnect appliances are distributed not just in datacenters, but directly to ISPs. Traditional CDNs aren't in your ISP.
Do Akamai has Servers right in ISP's DC / Network? I know they are within 1 hoops in lots of places, but never heard they are actually inside ISP's network, which is what Netflix OpenAppliance and Google Edge Node has done.
There's basically 2 costs to running a network, buying routers/switches/laying cable, and electricity.
The only thing that matters is peak usage because then you have to buy faster network equipment. If I buy 100MBit internet from my ISP, they basically need to spend, once, whatever 100Mbit of a port looks like, 1/10th of a gigabit port, 1/100th of a 10G port etc. They only have to do that once, yet I pay them every month, they're fine.
You’re only seeing part of the picture there. Hardware is approximately a one time cost, if you ignore upgrades and maintenance, and electricity is a consideration, but the other part is bandwidth to get data to that port.
ISPs also need a way to get your packets off their network and across to other people’s networks, which are possibly on the other side of the globe. A lot of this is done via peering agreements with other providers, where both parties will agree to connect their networks at a common data centre without charging for data going across that link, but there will still be some data that needs to go across transit providers - these are the companies providing things like transatlantic fibre, or interconnects between different parts of the US. They charge for a pre-agreed amount of bandwidth, and when you burst above that it gets expensive.
These agreements are the core of the net neutrality debate. Traditionally peering agreements would see both parties having a roughly equal amount of traffic going back and forth, however in a world with services like Netflix what you see is a huge amount of bandwidth being used on side of the link. This is why ISPs think Netflix and co should have to pick up the cost of those links.
I don’t agree with that stance since the reason ISPs have customers is often the ability to access those services. However, it’s not quite as clear cut as people make out.
But the same rule that basic cost equals hardware + setup + energy + maintenance also applies to the upstream provider, and their upstream provider, all the way along the selected route including the biggest interchange touched. The economic arrangements are different, but the cost base is just that.
Sure, and they purchase very expensive equipment to cover the peak bandwidth demand then spread out those capital costs in the form of a monthly bill.
If you want more bandwidth, it's "free" as long as there is room available, otherwise the whole chain needs to upgrade so as not to degrade the service of other customers. On a competitive market, the margins available and the prices should tend to go as tight as possible and the economics will lead to pay for usage.
They bear the cost to build the infrastructure, but their cost to deliver data is negligible. As long as their infrastructure can handle the maximum capacity spikes, they're basically printing money.
I used to work at a company that called these customers ‘zombies’ because you were afraid that any next marketing communication would ‘awaken the dead’ and trigger an unsubscribe. They’d even go as far as to not send marketing emails to ‘zombie’ accounts that had been long time inactive.
Bringing this back to Netflix, I imagine they would be concerned about a 0 activity customer leaving at any given time. Probably hard to forecast future revenue for them too.
In related news, MasterCard announced they were going to start forcing subscription providers to contact you monthly with a bill so this might be the beginning of the end for easy revenue from inactive users.
Ideal maybe, but if they want to maximize revenue they need to optimize around the marginal customer. Marginal customers most likely exhibit strong correlation between hours watched and retention.
Ideally of course because it's more revenue and less operating expenses. Similar business model of a gym, where they want many subscribers and nobody coming in, while making it difficult to cancel the membership.
I’m not sure that’s true. Unlike a gym which has a fixed amount of space, machines and resources, Netflix can distribute a fixed price production (tv/movie) infinitely. So it’s therefore in their best interest to make sure that customers are always bingeing.
Fair point and agree. It's not really comparing apples to apples. Netflix would want many binge viewers in order to retain their membership and get their friends to sign up to watch the same shows they're watching, while gym members don't necessarily convince their friends to sign up to their gym the same way they would if it were Netflix since it's more personal than social. There's a sweet spot between consumption and retention, just like how gym's don't like that one guy who's there for 12 hours a day taking up resources but would prefer someone who barely goes once a week which is enough for the customer to keep paying for the monthly membership.
interesting point. the problem is that no interaction means no data. and along the motto "if you can't measure what's important, you make important what you can measure" a more concrete metric needs to be used. even if - ironically - the optimization is opposed.
I was going to use that motto for a completely different email with my faraway manager but I found that the more popular version was "Measure what's important, don't make important what you can measure." It's attributed to Robert McNamara https://www.azquotes.com/quote/613247
sure but if you optimize you have to optimize towards a number. Now that number is either 0 in this case or x which is close to zero. Now which x would you choose? And if you choose an x (let's say 47 minutes and 23 seconds) you'll lose data moving towards it.
Ex-video game developer here. The games industry uses those exact metrics to measure their success.
I decided to leave the games industry after it was revealed that our latest release had a 14 year old kid who was playing, non-stop in 48-hour bursts .. and the game company execs loved it. Like, they were ecstatic that they had this 14-year olds' attention like that.
That was enough for me to realise that the video game industry is destructive, and has evil intent for our society...
Maybe computer games should show a pop up after a couple of hours of play, suggesting it may be a good time to take a break. Preferably at a nice save point.
I always wondered how some companies can rely exclusively on data like A/B testing to make decisions. Isn't that incredibly slow? In my experience there are thousands of decisions to make every year in a high growth company and there is just not enough time to test everything. This is why the best software companies that we are all familiar with revolve around an individual with incredible intuition/vision. If it was possible/practical to just rely on a bunch of A/B tests, every company would be successful.
Not really. With the amount of users these companies have, you can get statistically valid results very fast (and you can run hundreds of tests in parallel).
When I search on a specific title, I get anything that roughly matches any keywords in combination, then also any title that remotely associates (even conceptually):in any possible way with any individual keyword.
But what it won't tell me is "we don't have the title you want".
Specific example from earlier: look for the 2007 film "Sunshine". Then, go plug "sunshine film" into duckduckgo and see what comes up as the very first item.
>But what it won't tell me is "we don't have the title you want".
That's by design, not accident. The Netflix catalog is much, much smaller than most people think, especially for films. They have just under 6,000 titles available in the US, but a large proportion of that is cheap filler that most viewers will have no interest in. Searching the catalog using a third-party tool reveals just how thin it really is.
Netflix have focussed on recommendations over search because it makes their service look better. If you search for something, the odds are that Netflix don't have it. Recommendation allows them to promote their original content, which is Netflix's main moat - if you're hooked on Stranger Things or Orange Is The New Black, you're far less likely to cancel your subscription.
Do you really think that’s not on purpose or surprising? If you show them recommendations instead of an empty search result they might want to watch something else or get inspired to use other search terms to find some other result.
That may sound like a bad search for engineers but for the user it’s probably what they want and what Netflix measured.
Being an engineer is no excuse for not understanding basic product design nor being unwilling to see/accept how a feature benefits users.
(I'm saying that to agree with you, not to correct you)
This example is particularly good because they didn't even say "I get it, but I don't like it" or "I think it'd be better if..." Rather, they went straight to "wait, netflix devs are supposed to have skillz?" and "this must be incompetence", an assumption that represents one of the more toxic parts of our community.
I think the way they do it is very elegant. They surmise the film you want and if they don't have it, they put it in the top bar as "find films related to X". That way, they can push you into consuming something else even if they don't have what you were originally looking for.
Probably causes some customers to watch into some of the crappy search result titles, to see if it's the thing they wanted. Hence, the "hours watched" metric goes up.
Probably not so much toward retention (I'd have quit over this were I not getting it free) as toward redirecting you to content they already control at little or no variable cost.
Startups that are still searching for product/market fit don't really do much A/B testing - the user populations are too small for anything statistically relevant. They rely more on intuition, or as some would call it, blind luck.
In some ways, what an entrepreneur does is very coarse-grained A/B testing anyways. If you launch 50 different startup ideas and they all get zero traffic, and then you launch the 51st and it has 100 users sign up over a weekend, then there's a good indication that you should go with the 51st idea.
It's also what venture capitalists do at the level of companies. Fund a hundred companies, and then watch them. Pour more money into the ones who appear to be succeeding.
I doubt there are companies exclusively rely on A/B tests to make decisions, but even if there were, someone has to think about hypothesis and design the right experiment to then confirm with data. You can't just have a computer running around and make experiments and confirm them for youº.
--
º: you kind of can, if you have enough fresh data, to search for and derive statistically significant results automatically.
Maybe it's just me, but I don't think the secret behind this kind of success is just properly designing experiments. Again, if this were true, the best founders would be data scientists. But they're not. They are visionaries. So I reckon it must be the vision.
Vision is obviously necessary - but a big problem folks run into is failing to recognize when something is not working. A/B testing is one way to keep yourself from falling into the trap of thinking 'I know better than my users what they should want'
That's more of a systemic problem, it's debatable how beneficial getting people addicted to Facebook, YouTube's next video, or any of a number of engagement platforms is to humanity. But everyone does it, and the ones that dont, don't make it so big.
An excerpt from the moral philosophy of Goatee Spock Evil Dimension:
Littering, theft and pollution are also systemic problems that many separate actors contribute independently to - and no individual litterer, thief or chemical plant could solve the problem entirely by cleaning up only their own actions. That's why, for thousands of years, societies have agreed that litterers, thieves, and polluters are not culpable for their actions, and why no serious effort has ever been made to bring them to justice or discourage them. After all, you might complain when you see a mob boss using underhanded methods to get ahead, but cooler heads know that touchy-feely mob bosses don't get as far in their careers; they can't be blamed.
In my opinion, anti-litterers tend to be over-idealistc academics with no grasp on the real world. If I were to stop throwing cigarette butts out my car window, I would have more cigarette buts inside my car. They must not realize this, I guess, because idealists don't go on road trips. Besides, if I don't cover a square inch of grass with a chip bag, someone driving behind me will.
>everyone does it, and the ones that dont, don't make it so big.
Socrates was famous for posing this argument at Greek garden parties. Eventually it took root, and that's why the Athenians (and all societies inheriting their legal tradition) took every law off the books.
--An excerpt from the moral philosophy of Goatee Spock Evil Dimension
I think some of them are ignorant about the effects. Many smokers I have spoken with over the years believe that the butt / filter is just cotton and tobacco and paper and will decompose very quickly.
I'd like to see more education about this, and the amount that end up in the oceans.
Someone developed a backpack that can keep strong odors inside, there should be ashtray inserts and portable small proof boxes that smokers can use, so that their car does not get the old smoke-butt smell which is way worse than the smoked in the car smell - which I gather is one of the main reasons to get the butts out of the car as fast as possible.
I'd also like to see a change in how cigarettes are made, requiring the butts to be faster to break down / decompose.
Surely these things are possible, and less litter in the future as well.
Tragedy of the commons where the commons is human attention. Notifications are one of the ways to train people like pavlov’s dogs to check your app. If you don’t use it others will.
True and not true. Facebook, YouTube and Twitter make their money from advertisements. Netflix gets paid directly by the consumer. It's not obvious to me that it's in Netflix best interest to optimize the same metric as companies that rely on advertising revenue. If Twitter was paid, I'd already unsubscribed several times because I occasionally will try to "get off Twitter". It's a wasteful, unproductive time sink that makes unhappy in top of it. That's why I uninstalled the Twitter app. Now Twitter on occasion pulls me back in because I can just go to their website.
Netflix to me is a totally different beast. I feel many interactions I have with the platform are feeling more like a time sink. The interface feels increasingly optimize for wasting my time. It gets harder to distinguish shows I've already looked at from ones I haven't. They suggestions are more unreliable than ever. It's downright impossible now to decide at what point the obviously is nothing left on the platform I want to watch. In addition to the UI more and more of their content feels like generic, insightful material to keep you got to the screen. Not sure when that changed. I think it happened in tandem with Netflix being less about movies and more about tv shows. Not to mention the DVD to streaming move.
The more Netflix wastes my time like Twitter or YouTube the more likely I am to clean it from my life. Those services do not spark joy they just waste my limited time. No real reason Netflix cannot do better than them and treat me with dignity and respect, like a real customer and not like the product because in fact I'm not their product.
I had the same gripe about working at Twitter- the two main metrics were daily active users and user active minutes. User research was way understaffed so a lot of the side effects of optimizing metrics remained unchecked
So that why the auto play video feature just won't die despite everyone loathing it...
Most of our forays into Netflix these days are precision strikes. We know what we want to watch and we race to get to it without triggering those stupid autoplay videos. We no longer browse.
For a while I had my TV setup with variable framerate to match the content being played and of course the Netflix UI and Video operated at different framerates so every time a video would kick in the screen would blank for a few seconds and when you navigated it would blank back.
I really love how Hulu's current web interface makes it really easy to toggle Autoplay on/off. I hate how Netflix makes me search through account menus.
Hulu's current interface definitely isn't perfect, but I love the convenient toggle for Autoplay; and I love being able to search for episodes by name. As a fan of Seinfeld and It's Always Sunny in Philadelphia, I can remember an episode's name waaayyy more easily than I can remember its season/episode number
Maybe it's just in my head, but I feel like Netflix doesn't care what I want or think anymore. I feel like they've taken the position that their metrics and projections know better than I do.
> So that why the auto play video feature just won't die despite everyone loathing it...
Everyone loathing it? Is this really a thing? I never actually seen anyone complaining about it on Netflix. I've seen people complains about the "Are you still there?", my SO also often complains that it's "too long" between episodes and ask me to press continue with the remote.
I do hate the autoplay from Youtube personally, particularly since I've on Youtube premium (my phone just keep playing videos in the background, that's quite annoying and I'm really lucky it only has happened while I was on wifi). Still I haven't seen any people complains from it and I do know it can be easily deactivated on the website at least.
> We know what we want to watch and we race to get to it without triggering those stupid autoplay videos.
Are you on a special version of Netflix? Are you actually talking about the trailers that play on the background while you browse? You don't like that feature?
This seems like a particularly dangerous linear extrapolation in a nonlinear world. Shouldn’t you all have just had a minimum engagement time that you wanted customers to hit? Like ten hours a month should make you feel comfortable enough that the customer will not be canceling anytime soon. But 200 hours a month might mean you’re causing problems to their lives that they will address by canceling. Even the auto play feature just seem like a standard hack that Goodhart’s Law would predict. User hours may be a good metric to indicate user retention when it’s measuring people’s voluntary use of the product, but could become inaccurate when it measures their psychologically manipulated use.
Hmm not that I remember. Maybe if it hit the end of the season? Though it's actually been a while, at some point last year I felt like I'd run through all the interesting content on Netflix. Now I do the same thing on YouTube.
It's more like a screen saver. You haven't interacted with the app in X amount of time, it suspends loading the next episode until you confirm you're still there.
Hell, YouTube too does that on Android, pausing playback after ~1 hour.
I was quite annoyed when I saw it for the first time - I had music playing over my Bluetooth headset while cleaning my bathroom, so it was quite inconvenient to tap on the phone when YouTube suddenly paused it to check if I'm still there.
If it is a setting that can be disabled it would make sense even for Youtube Premium imo. Accidentally leaving it on would result in a lot of data for the user, potentially missing/losing where they are, would quickly drain the battery, and so on.
>I always had a bit of a moral dilemma working at Netflix because my goal was to get people to spend more time watching Netflix, which I didn't really think was good for humanity.
Ah, come on. We need entertainment. Netflix is far and above what cable companies have been doing for a long time in terms of effort and not being scummy. Bringing joy to people is good for humanity. Don't let it weight on you.
Nearly every place a person can work the product is designed to get people to buy more of it.
Netflix is chips, designed so that you can’t just eat one. Every product from packaging to taste to price to whatever is designed to manipulate us. We should definitely keep businesses in check but ultimately it is our responsibility whether we eat two chips or the whole bag.
Watching Netflix can be beneficial. There is really good educational content there. It is awesome for humanity and can help people in different situations. I am really surprised to hear that one can ask themselves if it is bad for humanity
With my 10 year old, he’ll need to watch some minimum amount of educational content online to
balance out the other stuff.
Generally he does this (or plays math/word/science games). And if he doesn’t, strangely our internet connection gets slow on his devices. Shucks, I guess it’s time to read a book.
It's not really addiction that worries me, it's habituation.
TV is absorbing, as a child with 3/4 channels there was often a time when the was nothing of interest and that helps one week interest elsewhere (aka boredom). If there's no point at which there's nothing interesting to watch it's easy to sit and absorb stimuli out of habit.
Maybe addiction can develop, does develop? If you do an action habitually you're still doing the same physically as perhaps you would as an addict.
Netflix user here. Honestly, people are going to watch TV no matter what. No amount of wishful thinking will make people do something more productive. Better Netflix than competitors, no?
I’d argue that, for at least some people, streaming/TiVo/etc. can lead to more deliberate TV watching rather than plopping yourself down and begging out. I guess the counterpoint is there’s all this TV just sitting there for the watching but I’m honestly not sure which predominates. Many
People historically watched a lot of TV, of whatever type happened to be on.
One issue I have is autoplay -- when a program stops you get a pause and time to think if you want to do something else. Which is why they don't let it stop ...
Continuous play is difficult with kids; you have to poll the TV viewing, rather than getting an interrupt. It's literally the only bit of Netflix I'm personally not happy with.
There was a time when people didn't watch TV. They used to listen to radio. And before radio they used to read books. It's entirely possible that a new form of media will come along, hopefully less focused on passive consumption of content.
did you ever resolve your moral dilemma? having done a very similar thing for another company, i felt the same way. ultimately i decided it was probably a net neutral, or possibly a slight minus, which definitely weighed on me. but the real reason i quit was because i decided it was incredibly lame. its so easy to find something thats an unambiguous plus, why even bother with questionable jobs? curious where youve landed, thinking wise
I always felt that it was a positive. I'm giving people hours of joy, allowing them to escape from whatever it is they are escaping from. I never even considered it a problem.
I was glad that I didn't work somewhere where the consumer was the product, and that we didn't really have to exploit anyone to deliver our product. Of all of the FAANG companies to work for, Netflix seems to be the one with the least things to complain about.
Could you give examples, I don't consider auxiliary workers to be really part of an industry (like non-specialist janitors aren't part of the satellite sector, say). Other than those roles I don't see how you can flit between industries chasing an ethical posting.
Let's say I'd like to work in environmental science .. what role can I walk into?
perfect example- i do data vis for carbon emissions on volunteer basis. there are lots of oppotunities like that that you could probably get paid a bit to do. email proffs working on cool projects and see if they need some cheap tech help. i repair and sell used electronics on ebay to make some money
I always had a bit of a moral dilemma working at Netflix because my goal was to get people to spend more time watching Netflix, which I didn't really think was good for humanity.
Either way, it was nice having very clear goals. Netflix was a great company to work for.