Google Will Soon Start Punishing Mobile Sites that Use Annoying App Install Ads
techcrunch.com
techcrunch.com
Chrome app -> youtube.com (no prompt to install/switch to YouTube app, simply redirects me to m.youtube.com) -> search video -> play (which plays out of chrome and does not switch). The whole process is exactly how I'd hope/expect it to be.
I've exclusively used Youtube through the chrome app instead of the Youtube app because a few years ago I found that the search results were significantly worse/more heavily filtered through the Youtube app for some reason and never used it again.
Even if you try living your whole smartphone life inside native apps (which would be sad), you inevitably get sent out to the web via links. Opening in your phone's browser or through the native web view.
The native "web view" is often an inadequate browsing experience, and will have the same annoyances as the full browser anyway, only worse because it's a cut-down version of a browser, tethered to the native app that doesn't want you straying too far.
To the point about dev's eating their own cooking. It's a fair question, but one I think increasingly is about project managers and non-developer stakeholders calling the shots on things because "everyone's an expert" in the web dev kitchen.
luckily the foundation hasn't seen a negative impact on donations.
I wonder what percentage of the knowledge graph result excerpts are full entries, and don't trail off indicating more can be found on the source site.
Perhaps some users would explore a topic if they saw their many options to do so.
This is just pure speculation, though.
double scary since it seems google is in a position now to steer (or probably 'make it better for users', in their view) the whole www just by tweaking page ranks
http://www.searchenginejournal.com/google-penalizes-chrome/3...
Even more extreme is that every single time I click a link in a hangouts chat on iOS, it asks me if I want to download Chrome.
The EU investigation committee is going to love this.
[0] https://developers.google.com/web/updates/2015/03/increasing...
[1] https://unop.uk/dev/experience-with-mobile-chrome-app-instal...
Edit: In case it's not clear, this isn't just for web apps. As of at least a month ago (Chrome 44) you can add native app install banners to a website [2].
[2] https://developers.google.com/web/updates/2015/03/increasing...
I can't believe those forums don't just migrate to Discourse.
I'll have to disagree on this point. There are a lot of phpBB forums out there, and their quality and content varies widely as the Internet itself. Many of them, esp. some of the long-established ones, are invaluable sources for all sorts of arcana. Wikipedia doesn't even begin to touch the kinds of obscure practical knowledge embedded in the "dark web" of phpBB forums. There are a lot of Google searches for which a single post by some domain expert is the top result, many of which in old, old forum software of some sort, usually phpBB.
FWIW, I've never once run into a Discourse forum in the wild, i.e. via a web search, site link, etc.
<rant>And: Can someone please please create a new stackoverflow that rewards solving problems instead of going full Daesh in destroying old artifacts, enforcing dumb rules etc?</rant>
Ubuntu: http://discourse.ubuntu.com/ Rust (I think): https://users.rust-lang.org/
It's pretty big in the Ember community for obvious reasons.
Outside of nerd-land, however, it doesn't appear to have taken off particularly.
I lost too much of my adolescence to PHPBB and the like to truly hate it.
- Dual core CPU recommended
- 1 GB RAM minimum (with swap)
- 64 bit Linux compatible with Docker
- Postgres 9.3+ (2 GB RAM min?).
- Redis 2.6+
- Ruby 2.0+ (we recommend 2.0.0-p353 or higher)
Ultimately I wound up putting PHPBB with a different skin back on the server, since it just requires MySQL, PHP, and any web server.
Plus Discourse has a lot of strange features (e.g. auto-mod promotion) and it feels like it is still in beta based on the feedback they get on their own forums.
And don't get me started on the docker-based setup. If you want a local native install (no docker container) well, you cannot. Docker or nothing.
The whole product feels extremely convoluted and over-engineered.
I just wanted something that sucks less than PHPBB or SMF, you know? But this thing mainly seems more bloated, and the live demos didn't really quite convince me of the "sucks less" part either.
One thing I haven't even been able to test yet, once you have it running, how much is there to configure in order to tone it down? Because if you want to run a forum for pretty much anything but a tech audience, I would like to get rid of all the "dynamic" and "interactive" javascript stuff. Because it really makes it less interactive/dynamic for people on old, low-powered computers. They need basic HTML full page reloads (assuming the HTML is somewhat lean--leaner than PHPBB/SMF preferably :) ).
Because the unique thing about forums, that still gives them relevance in the 2010s, is that you can gather a community with extremely specialized niche knowledge. If the person who knows everything about, I dunno, stained glass dyes, psychogeography or uh Sumerian breakfast recipes, just happens to have a really old machine you don't tell them to upgrade or suck it, because he or she may be one of the only people in the world with the particular expert niche-in-niche knowledge that brings invaluable quality to the forum, much more than having a shiny slick UI that looks buttersmooth on a new laptop.
I don't agree with all the design decisions of Discourse, but it was (seemingly) designed to be a straightforward migration from phpBB and other similar forums (flat ones). Just having a proper search alone is worth the migration.
For you. A developer. The average deployer of a PHPBB forum clicks a button on their cheap host and sets the options in the admin page.
They don't even have the foggiest idea what Docker even is.
I just had to create our Docker setup at my work, basically from scratch (and I'm a front end dev) and if I can, you can :)
How can I help you grok it?
Edit: actually, it CAN have a lot of influence on the quality of the content. Ever found a phpBB thread with dead images because the user hotlinked them? Discourse downloads them and replaces the urls with the locally cached ones, ensuring the images are there in five years. That's just one example.
Maybe my use case is out of the ordinary compared to a lot of HN, I use it for a single forum that I am an active member of, and I can't recall the last time I googled something and was redirected to a result that was a forum entry that redirected to a Tapatalk ad (99% of my google searches end up at stackoverflow or blog entries).
Personally I close the page.
I really hope the creators of tapatalk go out of business.
This only applies to page covering ads for app installs. Not other interstitials. In fact Google recently did a study showing how 69% of people leave the page when faced with an interstitial [1], yet they also recently released an updated format for their own Google overlay ad [2] (although this is for apps). They don't want to lose web/search traffic to apps, but they make plenty of money running these formats themselves. A good summary of the various conflicts of interest is here [3].
1. http://googlewebmastercentral.blogspot.com/2015/07/google-ca...
2. http://adwords.blogspot.com/2015/08/beautiful-new-designs-fo...
3. http://searchengineland.com/google-goes-app-interstitial-pro...
What does an ad that shows between levels of angry birds on your phone have to do with ads blocking you from getting your search result?
This sounds like some weird sort of misdirection attempt.
Google is claiming to do this to improve user experience, however:
- they do these exact ads on every site they have
- they arent stopping other interstitials so sites can be just as annoying on the first page
- a site with great content should not be penalized for the site's own choices on ads
- they're forcing a change that affects everyone else except them
The PR signal doesnt match the execution. It's a monopolistic move and raises questions about their involvement in so many layers of the ads/web stack.
> the signal is that they don't want their search results to annoy users by showing them app install ads on the first hit
which is absolutely great. Well done Google.
Everything else in your post is weird and contradictory. They're apparently fine with advertising and non-annoying app install banners but they're going to make annoying app install banners less visible in search results. I'm really not seeing the problem.
> It might be good for the consumer but that's just a side effect.
See what I mean? So it's a great move and should be celebrated, but lets not congratulate Google on their motives? OK, I think we can all live with that.
Yes it should! If those ads are hostile to the user they should definitely be penalized no matter how great their content is.
Different statistic.
Not always, apps give them the ability to have push notifications plus the fact that having the app icon in your app tray is a constant reminder of their service.
Being nagged to install another black box has, for me, nothing to do with a "better experience."
One of the most important lessons of judo is that you never use force yourself – you just redirect the force of the other person so they hurt themselves.
Using that philosophy in life makes stuff a lot easier ;)
And, funnily, it fits here, too. Using their own newsletter to get it blacklisted. The more they send, the less are received.
"After November 1, mobile web pages that show an app install interstitial that hides a significant amount of content on the transition from the search result page will no longer be considered mobile-friendly. This does not affect other types of interstitials."
This implies the answer is No, since that's another type of interstitial (not an app install) -- but is one that blocks most of the page content. It's curious that they're analyzing the nature of the popup closely enough to determine that.
They're basically a new kind of pop-up window, an annoyance that you have to click away.
I don't remember browsers only blocking certain kinds of pop-ups based on content, they just see if it wasn't initiated by a user-action, it'll get blocked.
I suppose for Google there may be some reasons why they want to block the app-interstitials (I guess because they want to encourage keeping the content they index on the web, not in an app?), but even then, that's still no reason not to punish the other types of interstitial just as much. Trying to make the distinction is just making it harder on themselves, they will occasionally get it wrong, or people will complain that their interstitial shouldn't be punished because it is (somehow) not quite about the app--making the criteria a bit broader seems like it's easier for Google and better for the users?
Safari is the odd one out, there. No idea why it's allowed to avoid it.
It's not so much the curating that's the problem, nobody would care if Duck Duck Go did it, the problem is with the regulators that view Google to be the only curator out there.
> without specifying what metrics Google is and is not allowed to use?
Why not? If society depends on something, we regulate it. There are many different possible degrees of regulation, so this is not a boolean condition, even though painting such situations as a boolean choice between pure laissez-faire and nationalizing an industry (instead of merely regulating).
If you don't like this idea, I suggest that Google (or any other business) should find a way on their own to handle the situation (possibly by making their metrics more available, transparent, and verifiable?), or the choice may eventually be taken out of their hands.
However, focusing on my suggestion of transparency and verification misses the point. The method by which Google regains (and maintains) the public trust is their problem, and I'm sure that more than one solution exists.
My point is it would probably be in Google's best interest to fix this problem, because waiting too long has consequences. At some point, the people who feel that are treated unfairly by Google - it doesn't matter if they are correct in that belief - will fight back, which has already started in small, preliminary ways.
> The outcome would be ugly.
The outcome would also be ugly in a future where a socially-necessary service is controlled entirely by a single entity. As usual in these situations, the best situation is somewhere in the middle.
To your broader point, I agree that it's reasonable to want Google regulated to some degree, though I do fear regulators will take on that "transparent algorithm" point of view, which will just make search terrible. Governments generally aren't very good at regulating tech businesses.
So ranking measures derived from user behavior: great. Ranking measures designed to prioritize features in your browser, or to push people away from apps (that, just by total happenstance, your dominant competitor does better than you, and which threaten to undercut your dominance in search/display ads): not fine. Which really is classical antitrust: you can't use your monopoly (search) to juice your other businesses (browser, app store).
Users will bounce if they don't like ads, with or without countdown timers. Acting based on observed user behavior is great. If they don't like it, they bounce. If they tolerate it, they don't bounce. Over time, users will migrate to less-annoying sites. Ranking sites on observed bounced rates, whether caused by ads, low quality, or what ever else, is awesome and should have no anti-trust implications. That's not what google did.
After all, if these ads really annoyed users in general, that would show up in their behavior, and thus naturally be picked up on by ranking algorithms, no? (They annoy me, but I'm not all users.)
In any case, that concept of using monopoly power in one market to bolster power in another doesn't really apply here with Google.
I share a lot of the unease many people have with the amount of power Google has been able to amass as the de factor curator of the Internet, but I'm also somewhat amazed they are in the position they are in and have been as (mostly) benevolent as they have been. I suspect it is actually pretty hard work just to remain somewhat close to neutral when you are sitting on that much power.
Windows 7 in Europe.
Remember them pushing Chrome on you? having ads directly on Google.com? To this day forcing you to use Chrome for Hangouts? And so on?
Do you seriously think Chrome would have caught on if it weren’t for this push of advertisements?
This is direct abuse of near-monopoly in one market (search engines) to gain unfair advantage in another market (browsers).
Seriously, look up "Dear Sophie" to see how many awards it won. Google doesn't do TV, so that's not a monopoly.
It was horrible, and I really believe this was unfair competition. Especially as people using AdSense complained that they got paid next to nothing or nothing for people viewing and clicking Chrome ads.
When they launched chrome, the browser wars had long since been won. Everyone who knew what a browser was had installed firefox, and everyone else thought IE was the internet. Just because you ignored the ads doesn't mean the general public did. It sounds like you're being biased.
If you're going to attribute Chrome's growth to "accidental clicks on ad manipulation" instead of the massive awareness campaign, you're going to need to provide proof.
Several people on this page provided statistics and reports how they got nothing or next to nothing when Google replaced all ads in the AdWords network with their Chrome ads.
As I’m also not interested enough in the topic to research sources just to change your opinion, I don’t care if you’ll downvote me again.
Are you sure you understand what you're talking about?
That's not what you meant.
It could be hard to argue that an interstitial pushing an app (especially an app specifically for the site the user is visiting that gives them enhanced features compared to the mobile web version of the site) is more annoying, distracting, or otherwise harmful to a user than any other kind of blocking interstitial.
Thus, one might try to conclude that Google's actual interest here is trying to keep users from using apps to access content instead of using the web site where Google is more likely to be the ad provider.
I have no idea if this could rise to the level of an antitrust violation, but it's the first place I'd look for one.
Still, I feel like newsletter prompts are also easy to detect, so that argument doesn't hold a ton of water unless there's also a separate plan to address those—or maybe they're testing the waters with this one. I think I need more information to have a real opinion :/
Also, the blog post states they're essentially calling these sites mobile-unfriendly, which I think is entirely fair. All full-page interstitials are annoying, on the mobile web or desktop web, but only "install our app" interstitials are unique to mobile. So if you view "mobile friendly" as meaning "as good as desktop web", banning bad things that are only relevant to mobile makes sense.
This is anecdotal, but I almost always abandon pages that lead to an app interstitial, which means that I end up back on the results page. This is a pretty unambiguous signal to Google that the search result was bad. So, the remedy that they're proposing is good and obvious.
Also, Google's search engine has never, ever, ever been objective in the signals used for ranking. Their results have been different and biased ever since the start, which is what made their search engine successful in the first place. It's a little late to complain about that.
Why only this type of interstitial and not those pushing newsletters or paywalls or surveys? "I know you've only just arrived and read all of one sentence but CHECK OUT OUR NEWSLETTER!"
PS Another thing that may also hurt Google's core business may be once Apple starts blocking their ads. Google ads will become the next Flash.
If your developers are not capable of making your website mobible friendly in 2015, any app they develop is likely worse. This is especially true when your app demands access to everything on my phone even though it doesn't need any of it to function.
For many use cases (games, highly interactive experiences, camera or microphone using app, etc.), a native app may just be better, and guiding users there might be a better experience and a better conversion funnel.
Yes, you can make a website mobile friendly, but that doesn't mean it will be as good as the native app.
In the context of forums though, its entirely true and accurate.
If we were talking about apps, as in things that require more than an internet connection to work, I would agree with you.
Just because you can make it an app, doesn't mean you should.
http://mobile.weather.gov/ would disagree with you.
When it comes to websites that try to force their app upon you when visiting their mobile website, my experience has almost always bee n the application is as bad as, if not worse than, the website's functionality. This gets to my point that if your developers are not capable of building a decent mobile friendly website then they are probably not any better at building an app.
Also I hinted at the fact that in many cases these apps appear to be designed to track users and invade users privacy. Why else do so many need access to contact lists and more when their functionality doesn't remotely require it.
In the world of apps, I can fool you into giving up your location (or on android just sample it as needed without the end user knowing) and other privacy stealing tricks I can resell to advertisers and marketers. I can also embed ads right into my application and have IAP to make even more money.
Until the mobile platforms fix these issues, there will be a financial incentive to push apps over pages. That's the elephant in the room here.
Still, G's new policy allows small install the app links, which is fine because a lot of people do want the app versions of web properties.
https://developers.google.com/mobile-ads-sdk/docs/dfp/androi...
Edit. More here:
https://www.thinkwithgoogle.com/products/mobile-ads-intersti...
Left hand doesn't know what right hand is doing.
They can often be annoying, but that's different than interstitials on top of a mobile web site, right?
Also they only want to stop full page ads pushing to download the app, not any other full page ad.
It could then provide alternatives to the app / site, and possibly even an anonymous whistleblower service for employees .
On the other hand, using result long-clicks as votes is quite popular, and seems to work well.
The business wants to convert low-value mobile users into high-value app installers, and would happily sacrifice most to get a few installs if that's net positive on long-term value. The search engine, on the other hand, wants as many users as possible to be successful in finding what they're looking for. These perspectives are not aligned - high abandonment rates due to interstitials can be fine for the business, yet penalized as destinations from a search engine. The underlying question is who Google optimizes for in this case, the user or the destination.
It only works for certain scenarios and cases, but at the end of the day, there's significant data to show that they work.
I certainly can't argue that it does yield positive business results under certain scenarios, but for the average user it seems to injure common web experiences.
For eCommerce, this type of interstitial is a no-brainer: users have an easier time feeling more secure purchasing through an app vs. Safari or some other mobile browser. (Especially if it's Apple Pay enabled!)
For content heavy sites, it might be that the opposite is true: faster access to content can be used to hook the user. If you own one of these sites, you still need to design a method to get the user to come back repeatedly after you post updates. Email subscription modals do a good job at this, but people hate giving their email - especially on mobile.
Considering the security of most of the "apps" out there this is really depressing.
The Hotwire Android app was the first one I noticed doing this, raising notifications unrelated to actual use of the app (e.g. for bookings). If they have a general promotion, like "fall sale" or something, they push a notification to a targeted subset of users. In their case there isn't even an option in the app settings to opt out of the notification spam. Yelp was the second app I found doing this, but they at least have an opt-out in their settings menu. For apps that don't, you can entirely revoke their notification privileges in the central Android settings, but I doubt the average user knows how to do that.
My dislike for playing whack-a-mole with this kind of nonsense is why I don't install apps anymore (outside a few trusted exceptions, like Wikipedia's app), and just use mobile websites.
Don't give a fuck. If I wanted your shitty app, I would have gone to the Play store and installed it. If I want to go to your website, then I want to fucking go to your website. Don't be a bag of rancid dicks and redirect me.
> 2) It's a solid retention strategy.
[Citation Needed]
We could choose to drop the interstitial as a matter of principal or on the assumption that the difficult-to-quantify long-term benefit will pay off later, but we'd have to make that decision not in the absence of data, but actually in opposition to the data.
Or let Google make the decision for us...
Can you provide an example of an app that is infinitely better than its mobile website?
If you can just provide one example, and not any of the tech giants. One example where there's a "mobile web experience" as you put it, and a native app.
I was shocked to learn how effective interstitial popup subscription requests are: http://optinmonster.com/3-reasons-lightbox-popups-capture-mo...
http://googlewebmastercentral.blogspot.co.uk/2015/07/google-...
Well at least they admitted they are bad, even if they tried them.
It's like a car company saying "we used to only make black cars, but actually we looked into it, and now we think people want colours; so we're making them".
"So we're making them, and we're penalizing any dealerships that are still selling black cars."
If a car manufacturer used to sell black cars, then added other colors, if their own dealers refused to sell the additional colors, it would be entirely reasonable for the manufacturer to step in (since it hurts the manufacturer's brand). So in that analogy, the manufacturer is right.
The reality is that Google aren't a car manufacturer, they're more like a car magazine. Suddenly they decide not to feature trucks, and therefore Bobs BigTruckin modal X2000 coming out next month is penalised because they won't be featured in the next issue.
Is it "fair?" In general, yes, but in this specific case because Google holds a monopoly on the market you could argue it is unfair. A lot of sites will suddenly see a large drop-off of traffic if they didn't get the memo, because so many depend on Google for redirects.
So Google is punishing competitors for something they do themselves.
This is all looking like Google just wants to make the EU regulators angry.
1. http://www.searchenginejournal.com/google-penalizes-chrome/3...
You cut them out, they cut you.
As you can read here their response to the EU antitrust lawsuit http://techcrunch.com/2015/04/14/internal-google-memo-respon... google makes the argument that users spend a lot of time inside apps. I would expect they are worried about the ability for users to go directly to content and bypass search from which they make money.
I'm not sure which side of the fence I fall on. The web is arguably a more open technology than native apps but the experience in a native app is often better.
If people move away from web and toward Google's mobile properties, then they're probably pretty okay with that.
I think its more driven by the fact that more of Google search traffic comes from mobile, so its more critical that the results provide a good experience on mobile.
Until then, the snark seems unwarranted, though I agree that there's good reason to be concerned: they definitely have a lot of power here, despite not having seemed to abuse it yet. That seems like an interesting conversation to have, especially in a more substantive format.
Apple is responding with ad blocking technology and a news app. Facebook has responded with articles that load quickly from their apps. These solutions represent an existential threat to the open web, and therefore Google.
I really hope Google doesn't just stop at app ads, and starts punishing websites that do objectively horrible things to end-user experience.
http://sentrant.com/2015/03/12/safari-to-app-store-redirect-...
This all assumes the site is using the features the browser makers are telling the sites to use.
> If a browser feature is annoying you, the correct approach would be to disable that feature in the browser.
Yet when some of us shut off Javascript - a feature that is often annoying, enables many kinds of tracking, and occasionally creates serious security problems - half of HN freaks out. Given that Javascript is necessary for many types of surveillance-as-a-business-model, hostility to the idea of disabling javascript is expected.
Push notifications are still a newer feature, but once that feature becomes entrenched enough that some people's profit depends on the feature, disabling push notifications in the client will be regarded with similar hostility.
They are. They are completely opt-in. You don't get push notifications unless you ask for them.
Don't get me wrong, I dislike advertising; I'd rather pay for content, but this strikes me as anti-competitive.
And you might not even know there's a gallery to view, because without JS the fallback might be a single image rather than a prompt to swipe the slideshow.
When the simple act of swiping within a web page requires JS, you're crippling your own experience. Not to mention all the other JS-powered functions on a page. Adding things to a shopping cart etc, is much more user friendly with JS enabled.
Or maybe the UX used to suck on the two because of a secret non-agression pact between Google and Apple ;)
Maybe i'm looking too much into this ;)
"Stay on the native web so we can sell more ads!"
Maybe it's mutually beneficial for Google and its search users, but this is certainly the argument that's going to come from folks who are about to lose a ton of app downloads.
http://searchengineland.com/google-goes-app-interstitial-pro...
There's definitely some advantage to keeping people on the web, since App Indexing requires buy-in from content producers, but, if browsers start truly losing to native apps, Google are prepared.
note: i work for a publisher and when an app store redirect comes through AdX or AppNexus and someone makes a complaint it is one of the worst, most helpless feelings. this is why people use adblock and there is nothing we can do because the networks are, still, an important part of our entire revenue stream
======
I'd be willing to bet, this will one day be in an FTC/DOJ report about monopoly practices at Google...
Google seems to have forgotten they don't want to be looked at as a monopoly or abusing their power for the small things. While this is consumer friendly, at least at first glance, I still find it anti-competitive and generally worthless exertion of their near monopoly power.
It also effects them quite heavily (GMail being my prime example). If I had to guess, this is a decision by someone who is lacking the resources to fully grasp the decision and potential long term ramifications of their action. Maybe they do, but for a company of their size this seems petty.
The potential anti-trust lawsuit however will not be petty.
It's more like a common practice by retailers: using their market share to force changes by their suppliers. For example, Tesco (the largest UK supermarket) recently banned certain sugary drinks targeted at children.
That means it will most likely end up in some antitrust trial.
Or Google Search will penalize other Google sites that use those ads. It wouldn't be the first time Google-owned sites have run afoul of Google Search penalties.
Interesting Fact: A chunk of Google Chrome's install growth has been courtesy of the exact same type of bundleware offers with Adobe Flash, Java updates, and free antivirus installs/updates.
1. Apps can't be searched or content linked from Google.com
2. Apps don't show (as much?) Google ads
1 + 2 = Google penalty