Hacker News new | past | comments | ask | show | jobs | submit login
Average product lifespan of Google products before it kills them (gcemetery.co)
199 points by walterbell 17 days ago | hide | past | web | favorite | 99 comments

I noticed in the "recent deaths" section, YouTube for 3DS. This made me laugh, because at Netflix, supporting the 3DS was always a big pain. Regardless, Netflix still supports it to this day.

In fact, as far as I know, Netflix has only ever killed support for one platform -- The PS2. And that was only because there were only about 10 people left using it.

So we sent them all Rokus and told them we're discontinuing PS2 support.

Netflix is the last app that works on my Sony Google TV from 2011. The YouTube app shut off after like 2015, but 8 years later Netflix is still trucking.

Funny that it has google in the name of the OS and a third party app was supported for longer...

Why don't these devices run Android? Pretty sure you can run YouTube on a 2011 Android release. That's not important to Sony, of course, but even for new hardware it seems easier to make it run Android and shovel your bloatware on it than to develop your own half-assed OS for each TV model.

Google TV is a super-mutated version of Android... Google turned their back on it in 2014 in favor of something that was closer to normal Android.

Youtube's required Android version is listed as varying depending on device on Google Play, but on apkmirror.com it says that the minimum version is Android 4.4+. That was released in 2013.


That's for the latest version of YouTube, but I believe an old version of YouTube will still work fine with today's servers, at least for basic video watching (commenting etc. broke in the big Google plus debacle)

My mom still watch Netflix daily from my old Wii U I left home.

Every day with streaming still working is considered a win

Ahhh my 40" google TV was too big to haul across the country the most recent time I moved. Ended up in a dumpster. Really too bad.... That was a nice TV. Good picture quality.

On the bright side I bought a new TV and found, to my surprise, digital antenna works much much better now. I have two TVs, the flat screen from ~2010 gets 3 channels, the new one gets 30. They're connected to the same source.

> Ended up in a dumpster

Bummer. Put it on the side of the road and someone would have LOVED to take it.

I love that you sent them Rokus. Most companies would just tell them they are out of luck and to go buy something.

My old Panasonic TV has a handful of apps left (it's got like three home screens for apps, and like four still even exist). Thankfully Vudu, my app of choice, still works. As does Netflix and I believe Amazon Prime.

> I love that you sent them Rokus. Most companies would just tell them they are out of luck and to go buy something.

Yeah, I liked that too. It made me feel good an employee to know we did the right thing. Given the cost of the rokus vs the super happy customers plus the morale boost to the employees, it seems like a no-brainer.

> I love that you sent them Rokus. Most companies would just tell them they are out of luck and to go buy something.

Amazon did something in between those two when the Prime Video app on some Sony Blu-ray players lost support. I'm not sure if they lost support because Sony was no longer interested in pushing updates to those players, or Amazon was no longer interested in writing their app for them, but for whatever reason they were going to stop working.

What Amazon did was sent users a coupon, good for the next two months, for $25 off on a Fire TV Stick, Fire TV Stick 4K, or Fire TV Cube.

For quite a while during that two months, Fire TV Stick was on sale for 24.99, so with that coupon was essentially free if you didn't procrastinate too much.

Netflix support for the Wii is also being discontinued:


I didn't count that one because Netflix didn't actually discontinue it -- Nintendo did. They removed all support for streaming services.

Why doesn't Netflix support Switch? If you can't answer I will accept Nintendo politics as the answer.

Given hulu is supported, it’s pretty weird. Nintendo clearly doesn’t mind streaming apps. So I assume it’s netflix that’s uninterested in the platform, not Nintendo. But I’m an outsider, this is all speculation.

No idea, that happened long after I left.

So quick sign in on your Wii and get a free roku!

That's too bad. I liked Netflix's UI better on the Wii than on any other platform I've used it on.

The counter example would be Google Maps. They really care about this product. I think they even support Palm OS.

They care about it because

a) it’s basically Google Search just with a visual representation


b) the closest the West has to a ‘Super App’ (book hotels/restaurants/movie tickets etc straight from Maps)

I mean, they purposefully kept features off Google Maps on iOS to make Android look better, then brought them in once Apple made Apple Maps.

Do you know any articles about this? Would like to learn more.

Specifically turn-by-turn navigation, which Android had via Google Maps but you had to buy an expensive app on iOS like Garmin or TomTom. It was a real killer feature.

Apple and Google had been long time collaborators but the relationship was finally cooling off as iOS and Android competed more aggressively. Google continually pushed the rollout date for turn by turn on iOS so it was always “6 months away.”

Eventually Apple bought some mapping companies and when they were near complete Google finally added turn by turn to its iOS map app. Apple had a reaaalllll rough rollout the first ~3 years of Apple Maps I’d say, and for years after still nobody trusted Apple Maps. But they were no longer at the mercy of Google.

I believe vector images were also a point of contention.

That is impressive. I've seen it have hiccups before, but I just tried again on my (old) 3DS XL and it did indeed still work.

Not being able to switch users is pretty aggravating, though. (Both for me and the family member who's account the 3DS forces me onto.) That issue alone prettymuch keeps me from using it :(

> Not being able to switch users is pretty aggravating, though.

It should be noted that the final version of the 3DS app predates multiple users, hence the limitation. The app can't be updated.

Oh, does Nintendo not allow you to ship updates? That would make it a pain to support!

Can confirm. Found my Boxee a couple months ago and Netflix worked. I honestly couldn't believe it.

Love this a ton! How does Netflix ensure reliability for those platforms with an ever growing codebase and the evolving business? Was the platform built to be extensible from the start?

Most of the old apps can't be updated, so they are sort of frozen in time. They all hit APIs on the backend, so as long as the APIs they hit remain backwards compatible it "just works". The old apps don't get any new features that require UI changes or can't be supported on the platform if that platform is updateable.

So the short answer is, everything is an API, so as long as the API stays backwards compatible it just works.

Gotcha I thought that it might have been server driven clients similar to the way Spotify does it. But that makes sense :)

Last time I checked about a year ago. Netflix still works on my iPad 1

Okay, so the page/statement is inaccurate, and not just because it's been spammed to HN three times before today (seven months ago).

Since I run Killed by Google, I feel it's important to clarify that the average Google product lifespan is much longer than four years thanks to flagship products like Gmail, Maps, Docs, etc. I can't actually get a solid number despite a lovely spreadsheet of products that I actively track because I haven't even compiled them all yet. Anyway, I made a choice not to draw stats in killedbygoogle.com for a reason, and it's because they're misleading to most people. When I'm asked about it, I say, "The products listed lasted an average of about four years." KBG is cynical view of Google's product strategy by any measure, but at least it's not drawing empty and misleading conclusions.

Like always, I have issues with these lists. Project Ara is there despite never actually launching a product. Or Google Glass is on there despite a new Google Glass hardware that was released this year.

They also list Google Sky Map ( https://gcemetery.co/google-sky-map/ ) which they call discontinued because Google released it as open source and handed over reigns to another group of developers, but looking at the github it still gets updates too, so I can see this one being either way. https://github.com/sky-map-team/stardroid

As an owner of a $1,500 Google Glass - which now is pretty much a brick, as the companion app doesn't work anymore - knowing there is a new version in the market isn't very helpful.

About a quarter of the projects in the graveyard had replacements, and often automated transfers to the new tools (writely, a bunch of analytics tools, songza, etc.)

Another quarter were explicitly experimental (labs, glass, ara, etc.).

And another handful are products that no longer make sense in the modern world (google desktop, various toolbars, browser sync, the gmail notifier, etc.)

Hell, one of them is apparently the webconferencing software google employees used internally in like 2012, it wasn't a product.

> Another quarter were explicitly experimental (labs, glass, ara, etc.).

How many years was Gmail in "beta"?

Not a Google fanboy by any means, but couldn't this be read with the opposite conclusion as well - that the average time that Google keeps marginal products alive is 4 years?

Of course it's possible that everyone has some personal favourites amongst those 164, but if an idea really has legs what's to stop people from exhuming it from the Google graveyard?

It's probably more like a year with actual support and updates, two years where it's left to whither and slowly die, and then a year of notice before sunsetting it.

Yeah. By that logic MS has eternal support for everything, because MS has a legendary hardcore approach to backwards compatibility that they will release a product, push it hard, and then abandon development on it... but still have it "officially supported" for a decade while it bitrots into a nightmare... but never tell their users "stop using this, it's deprecated".

As much as I hate Google sunsetting products, I hate even more the companies that keep their zombies shuffling along in a state of undeath forever.

That makes no sense. Nobody is forcing you to use an abandoned product. So you're obviously better off with the Microsoft model than the Google one.

A good example is VB6, the runtime is still supported in Windows 10 -- and necessary for one our largest vendor products to run -- yet the VB6 IDE only runs in XP. Our business would have been really screwed if they just dropped support for it. We are, of course, working to move to a different product and have the luxury of time to do that.

Or worse when an old product kinda pseudo gets merged into a new product, in a way that makes both of them worse. Better just to cancel the old one and move on.

4 years for a product is abysmal from one of the richest companies in the world. This is Google not some garage shop with 2 employees.

I have no idea what the metrics are, but to continue my devil's advocacy, would the counterpoint here be "average lifespan of a marginal product that doesn't get acquired by $BIG_COMPANY"?

> but couldn't this be read with the opposite conclusion as well - that the average time that Google keeps marginal products alive is 4 years

Averages are misleading when the distribution is bi-or-multimodal.

So are you saying that it couldn't be read like that?

Think about how many Google products have lasted much longer than four year: Search, Maps, Gmail, Drive, AdWords, etc. This skews the average.

What would be more helpful is a median, or percentiles (eg 75% of Google products only last X years, 25% last Y months). I imagine that would be much more representative of Google’s fail-fast experimentations.

FTA: The average lifespan of a discontinued Google product is 4 years and 1 month.

"Living" products aren't skewing anything

Id be interested in seeing the median lifespan. I feel like it's possible there are long lived products bringing the average way up, although I could be wrong without diving in further.

I imagine Google views this as a positive thing, trying a million ideas, so that hopefully it increases the chances of having at least one good and popular idea. The downside is, a lot of products they killed were good ideas, but its sink or swim and they don't get the products the needed love and attention so they don't take off and thus they kill them. And now even when releasing a good idea or good piece of hardware people will always be worried if it will be gone tomorrow, which leads to less adoption and more axing of products.

Maybe, but some are exceptionally bad. Why did Allo exist at all for example? And the merging, demerging involved with Messages, Hangouts, and Voice have been a complete mess. There's more there than just trying some ideas, as some of the ideas actively damaged older ideas.

I don't understand why they announced that Google Hangouts (classic) is closing to move to Google Hangouts Chat, like it's a chat app am I going to notice a difference, but now you just made me think that the whole service is shutting down?

This is what happens when companies are disorganized and don’t communicate internally.

I made a post recently on another article saying I thought the chatter about them killing product was overblown, but the chat/video stuff is truly a cluster that's doing them no favors whatsoever.

We had a separate SMS app, Google Talk, the Hangouts app, Hangouts integration with SMS, the new Messages app pushing you to de-integrate from Hangouts, and whatever the hell is happening now with Allo and Duo, Allo dying, Hangouts moving to business use only, or to become a Teams competitor, or is that even happening? They successfully scared everyone away from Hangouts chat, whatever is actually happening to it, but all this time, the "chat" area of Gmail still presents what I assume is Hangouts? But I guess that's about to change also...

Now Messages itself is slowly pivoting to "chat" with the RCS rollout and I really can't understand how this could have possibly come to be like this. The branding nonsense of introducing all-new, quite similar features for consumers and pivoting the existing brands everybody already knew to business must have been based on some analytics about Hangouts video chats being a Skype competitor for remote meetings, I guess, but it's still just baffling to think about. And I haven't even mentioned Voice's overlap in several of these areas.

Putting aside the obvious confusion it creates, how can it be efficient to operate so many overlapping services that do such similar things?

or Google could have kept things simple with having just one app called Google Talk.

Yeah. The Gchat/Hangouts mess was what caused me to permanently move off all Google-based IM-platforms.

You better believe I’m not coming back.

Yes, I see this as healthy turnover, the circle of life. I can testify that every main project I worked on when at Google is gone now except Chrome, the last thing I coded for. However, I don't think my time on other stuff was wasted.

Each of those previous projects shipped and did something useful for the public and helped Google in some way, like providing technical research or laying down infrastructure later used by other projects that still exist. At the very least something got killed because the team was needed on a high priority project, where they made a big difference.

I am genuinely curious if we are only seeing tip of the iceberg so to speak and there are a much larger number of projects getting killed before they are even released? Although it does not affect end consumers, I am curious from an engineering point of view.

Any Xooglers care to comment? I am not asking for actual numbers, but just ballpark numbers / estimates.

If it turns out Google experiments in the open while others experiment internally and kill products too early, may be this isn't so bad after all. Society got to see what worked and what didn't.

These stand alone products get killed but I would be interested in seeing a similar summary by feature. It's always seemed like a strategy to me that they build and release an MVP, early adopters explore it, love or hate it, and the best of what's killed winds up integrating into the core suite.

For example, I miss Google Latitude but location sharing is a feature of Google Maps now. I never used Inbox but I assume some of the new Gmail features are from what they learned there.

For all the fun people made of Yahoo back in the day, this is more or less the trajectory their many products went through. With the difference being a lot more of the internal discussions, consolidations and "new direction" decision-making was known, making them seem headless and chaotic.

There seems to be a general problem supporting employee-led innovation at scale, when the bean-counters start to control the product pipeline (and rarely look beyond the next quarter).

I find it amusing that, in a comments section that is generally irritated with Google about product shutdowns, there are also complaints that some of the products _suck_.

well one of the popular things said on HN is that ideas are cheap and execution is the determining factor for success. I think Google was deceived by their first idea which was so good that mistakes could not derail it and now they expect every idea to be of the same caliber.

This is of course aside from all the other problems they have with a company culture rewarding ideas but not rewarding growing a product, and so forth.

I wonder what this is the average of? It is the average of just things that were killed, or the average of all products, including living ones like Search? And is it really average or is it median?

> based on discontinued products listed on our website

This makes me think it is only dead products, which seems like it would skew things a bit.

If anything this tells me it takes them a long time to make a decision to kill something.

Exactly. This website is either making a very basic mistake or is just plain misleading.

This is part of the reason why I decided not to include stats at killedbygoogle.com . When you included flagship products things like Gmail, Maps, Drive, Docs, etc the 'average lifespan of a Google product' rises significantly.

While some might be seeing 212 potential youtube scale success stories that google killed "for no apparent reason", I'm seeing 212 potential wework scale disasters that google avoided by killing off successful pitfalls and focusing on their core. The truth is somewhere in between, but obviously with google being what it is today, none of these cases chip-ed of anything from their success.

Meanwhile on AWS I was using SimpleDB until recently on a small project - I make AWS NO money - but they seem to still support SimpleDB even though it is not actually marketed? It's 12 years old and can't be generating a lot of new signups because it doesn't actually show up anywhere I don't think.

Does anyone know how AWS handles depreciating items on AWS. I've yet to be bitten despite beng a long time user (S3 still going / Simple DB was going last time I needed it etc).

As I've understood it, AWS keeps things alive as long as someone is using it. Once they want to get rid of it, they take it off the catalog of items you can provision, and start encouraging users to move to it's successor, and once it's abandoned fully they get rid of it. That's how they've always done old ec2 instance types, for example.

SimpleDB is indeed still up and running. Considering how it was described from the start as "retail store product catalogue" I wouldn't be surprised if AWS is keeping it around due to demand from a single large customer which starts with "A" and ends with "Retail".

I wouldn't be surprised if SimpleDB has been reimplemented on top of DynamoDB though.

It surprises people that Google kills products because people them as traditional products. But they're exploratory vessels for selling ads. Of course Google's ads are the product.

It's unsurprising, to me at least, that they explore lots and lots of ways to sell ads. That's exactly what you do with whatever widget your company peddles.

Average lifespan of products that have been killed, when they are killed. The actual average lifespan is probably much longer.

As I understand it, a successful product launch is a strong addition to a "promo packet". What we are seeing is the inevitable result of a system that prioritizes new products above all else. If google wanted to combat the stigma that they kill 3 out of every 4 products they launch, they could just start rewarding promo packets that demonstrate maintenance and growth. I still have Reader bookmarked as a reminder to not get to invested. Show people that they are committed to the longevity of their products and you might just get more early adopters for you next big launch.

I'm upset about Google Trips.

Trips was one of the few examples of how letting Google know everything about you lead to a great experience. It'll automatically figure out my time tables, hotel reservations, and flight times. The ability to download city info and store it offline was wonderful, and its recommended itineraries, while often silly in their listed time tables, were a great jumping off point.

Nothing else can exist that does the same job, because only Google has access to literally everything about you.

I never heard of it, and I'm one of those people who would definitely use it.

Nobody is pointing the massive flaw in the headline ? This count does not consider products that have not been killed (e.g. Gmail)

It's only the average lifespan of killed products.

If you'd like a formal survival analysis which takes into account that right-censoring, I did one a while ago at https://www.gwern.net/Google-shutdowns

RIP Google Inbox. I still think it was the best mobile email client they've made.

Hear hear!

"The average lifespan of a discontinued Google product is 4 years and 1 month."

"The average lifespan of a DISCONTINUED Google product..."

Note this does not mean the average lifespan of a Google product is 4 years. There numbers do not include products which have NOT been killed.

"... they might develop their own emotional responses. You know: hate, love, fear, anger, envy. So they built in a fail-safe device."

"Which is what?"

"Four-year life span." --Blade Runner

Here, though, the emotional response is our attachment to a Google service.

Most of these could have been a success but google has no idea how to connect with customers. An idea and a product is what we have. If customers accidently start using the product and it becomes a success it stays but if it doesn't meet some corporate goal it gets killed see g+(first they pushed it everywhere making everyone hate it, the moment groups of people started using it google saw it would never reach facebook and killed it. If they would have left it could have pivoted).

I don't know what many of these products are either. Cloud VR cloud, but it doesn't sound like something I would shutdown maybe sell.. maybe rethink.

Reminds me of the Autodesk Graveyard:


I'm still disappointed they killed off Google Reader, it was an excellent product as it was, and it had so much potential to be more.

A better metric would be rate of death, not total deaths. For example, how many products does Google have now vs 2004 Google?

Not only a better metric, but the only way to actually compute life expectancy. We need the mortality rate for each year, then we can know the expected remaining time for any particular "age".

I interviewed at YouTube and I kid you not, the interviewer had no idea what YouTube Leanback was. As wrong as it sounds, I had to school her, a team lead at YouTube on their own product. YouTube Leanback is awesome and will be missed. I integrated it in a chrome extension I made called YouTube Share Enhancer.

Does no one at Google HQ ever have the guts to raise their hand and bring this up? It's a meme by now.

There have been some comments on here about how career advancement optimization at Google involves hopping between projects, staying long enough to launch, and then moving on.

Whereas doing good work on product maintenance or evolution is much less rewarding.

True or not, it would explain a lot about how Google (as a whole) behaves towards products.

That's not a Google only problem. It's a cultural problem in Silicon Valley (ex drive to disrupt) and America (ex not respecting stay-at-home parents).

See the artwork of Mierle Laderman Ukeles: https://en.wikipedia.org/wiki/Mierle_Laderman_Ukeles

Feels like someone higher up should walk through some of the now dead projects, and fire management team members, who wasted resources on those. Especially highly promoted and/or repeat offenders. Might actually spark some real innovation.

While I was there, we used to joke that the claimed strategy was “launch and iterate” but the career strategy was “launch and leave”.

It’s made worse by internal churn. An unstaffed service will blow up pretty quickly (because someone in infra wanted to go for promo) so you either have to justify staffing a project indefinitely or kill it gracefully.

Raises hand.

"Hey, I think we shouldn't cancel this project."

"Not your decision."

End scene.

I did not know that Chromecast audio was discontinued. That was pretty fast.

Nice. So this is effectively what is keeping skynet at bay.

Inbox still hurts

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact