Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Google AMP case study: leads dropped by 59%

kinsta.com

Post reply on HN
Print view View original

91–100 of 163 posts

Re: Google AMP case study: leads dropped by 59%

#91

Post by lern_too_spel »

dep_b wrote: post #33
untog wrote: post #11

Earlier quoted context omitted.

It isn't though, really. Page load really is nearly instantaneous. I'm no fan of AMP, but I see the reasons why users like it.

You're right about that, but AMP'ed pages not coming from Google's cache are already really fast. Actually they're only preloading those pages while you're on the results page, which is something they could do as a reward for all pages that have a reduced page weight.

It is not safe to preload arbitrary web pages. The user will be deanonymized to a site they never visited. It is safe to preload AMP pages by design.

Re: Google AMP case study: leads dropped by 59%

#92

Post by yorwba »

drewg123 wrote: post #81
freditup wrote: post #58

Earlier quoted context omitted.

It's interesting that so many people would enjoy AMP, because on my Android, viewing AMP sites in Google's main app often feels much slower than regular web browsing in Firefox for Android. The difference? I have an ad-blocker in Firefox, but not in Google's app. So Ad-blocking > AMP > Regular (as far as user-experience from a speed perspective). Obviously Google has a lot of incentive to make sure people don't go th…

The best of both worlds is to use Firefox to handle the AMP site links from Google Now, or Google Feed, or whatever it is called now. Fast, and with ublock. The only problem I've ever seen is that stories from tomshardware do not load in firefox mobile unless I replace the amp in the URL with a www.

You can use https://addons.mozilla.org/en-US/firefox/addon/amp2html/ to have AMP sites automatically redirect to the non-AMP version.

Re: Google AMP case study: leads dropped by 59%

#93

Post by shaki-dora »

asdfologist wrote: post #15

As a user I like AMP because it's guaranteed to be very fast. There, I said it.

HN starts caring a lot about the integrity of the sacred HTML when it's their own content being mangled.

User experience is only valid as an argument in the context of that darn, biased "mainstream media" being dismembered by ad blockers, or held for ransom by "Brave".

Re: Google AMP case study: leads dropped by 59%

#94

Post by shaki-dora »

tyingq wrote: post #30
izacus wrote: post #27

Earlier quoted context omitted.

Your conspirancy theory completely ignores the whole shitshow of 15MB+ textual websites we've had to endure in those last years.

"Conspiracy theory" is an interesting take. Are you suggesting Google's motives here are purely altruistic...solely to improve the web?

The conspiracy here seems to refer to the accusation that Google is trying to actively make their search results worse, in order to force people to buy ads.

That post does seem to be paying lip service to actual benefits for users (such as speed), but then ... suddenly ignore them in favour of a Trojan horse metaphor that doesn't make much sense to me.

Re: Google AMP case study: leads dropped by 59%

#95

Post by notahacker »

criddell wrote: post #56
tyingq wrote: post #49

Earlier quoted context omitted.

I suppose, but that sort of reinforces the real purpose of AMP. No publisher in their right mind would give up the top part of their page and enable left/right swipe to navigate to competitors, give up the back button control, etc. It's basically a protection racket. Pay up (in control terms), or get pushed down the search results page. Feels a lot like the old "digg bar" ( https://moz.com/ugc/die-diggbar-die-die-sho…

> No publisher in their right mind would give up the top part of their page and enable left/right swipe to navigate to competitors Publishers usually do not make good decisions. That behavior is for users and (IMHO) it's great.

That behaviour is for Google, not for users.

Re: Google AMP case study: leads dropped by 59%

#96

Post by sureaboutthis »

mort96 wrote: post #23
sureaboutthis wrote: post #14

Earlier quoted context omitted.

Speeding a site up is one thing. Speeding it up substantially is another and I doubt your example of one who only knows how to insert jQuery snippets knows enough to both understand and implement all the details from server to network to browser.

You don't need to know enough to understand and implement all the details from server to network to browser, you just need to know enough to remove the tags related to ad networks and tracking to see huge load time and ux improvements. It's not really the web developers who are the issue; they just implement what the people above them in the org chart wants. You could argue that the web developers should push back on…

If one thinks all they need to do is remove javascript to speed things up substantially, then they fit into my category of people who only know jQuery snippets only think they know enough.

There are a multitude of issues that can be tweaked to speed things up more so than javascript elimination. You can switch to HTTP/2, implement server push, image compression, header compression, compressing HTML/CSS/Javascript downloads, preloading, above-the-fold downloads, and much more. A jQuery plugin user might not know half those things and I only mentioned maybe half the things one should know.

Re: Google AMP case study: leads dropped by 59%

#97

Post by PaulHoule »

lol768 wrote: post #55
privateSFacct wrote: post #41

Here's the reality. The Taboola news carousals at the bottom of most slow loading actual news sites are filled with total junk. One page per image websites, super slow, adblock goes crazy. I've literally NEVER had a good experience with these "Stories you may like" Google does something with a news carousel that loads fast and the content isn't crap. I get why publishers running things like taboola think it's terribl…

> The Taboola news carousals at the bottom of most slow loading actual news sites are filled with total junk. This, a thousands times. For both Taboola and Outbrain it's clickbaity trash - usually misleading and low quality content. They're designed to "disrupt" so that people click on them and have in the past featured on blatantly false news stories. I'd honestly be ashamed to work at one of these sort of organisat…

I think the ads you get from Taboola put you through a really awful funnel to weed out one-eighth wits and anybody else smarter.

It's like the "grooming" that a pedophile does or the badly spelled message you get from some Nigerian prince or somebody who says you won some lottery that you've never entered or even heard of before.

If you click on Taboolah links and manage to sit through a 45 minute video of somebody droning on and using words they can't pronounce (e.g. "Bih Seph A Nol A" as opposed to "Bis Phenol a") you just might be dumb enough to punch in your credit card number and get an $80 a month supplement that you can never cancel.

If it wasn't so offputting to ordinary people more complaints might go to the FTC and they might have something to fear from the authorities. As it is, public servants are paid too much to spend hours just to find out what the scam is if somebody reports it.

Re: Google AMP case study: leads dropped by 59%

#98

Post by cronix »

yorwba wrote: post #90
freedomben wrote: post #76

Earlier quoted context omitted.

Agreed regarding ulterior motives (keeping you in their ecosystem). In the excellent book "In The Plex" by Steven Levy, he talks about Google's motivation of essentially building a browser, mobile OS, etc just to ensure that some other party (like Apple) doesn't cut them out and reduce their search/ad domination. I believe protecting search/ads is at the base of their motives (as is, undeniably, vacuuming up tons of…

> But that said, why do so much for open source? It gets them more users for their services if people make use of their open-source libraries that happen to be optimized for Google, e.g. by integrating better with Google's cloud offerings. > Why not keep big things proprietary? They do keep big things proprietary. If it directly makes money for Google, you're not going to see it's source code, unless you work for Goo…

> Most people don't change their OS ever

Or the default search engine, which is Google. So most Apple users are still using Google, and Google pays a ton for Apple users data. $12B is not a trivial chunk of change. That's more than most companies make, and that's only from search! http://fortune.com/2018/09/29/google-apple-safari-search-eng...

Re: Google AMP case study: leads dropped by 59%

#99

Post by hnuser355 »

freedomben wrote: post #66
hnuser355 wrote: post #52

Earlier quoted context omitted.

We were looking for someone to do basic web dev for us, but when our leading candidate failed to even derive KKT conditions from scratch in our interview we simply had to eject him from the candidate pool

Did you find someone? "basic web dev" and "derive KKT conditions from scratch" do not sound like combinations that I would expect to find often, but it's always interesting to hear about different industries/expectations out there.

Sorry it was a joke :)

Re: Google AMP case study: leads dropped by 59%

#100

Post by TeMPOraL »

sureaboutthis wrote: post #96
mort96 wrote: post #23

Earlier quoted context omitted.

You don't need to know enough to understand and implement all the details from server to network to browser, you just need to know enough to remove the tags related to ad networks and tracking to see huge load time and ux improvements. It's not really the web developers who are the issue; they just implement what the people above them in the org chart wants. You could argue that the web developers should push back on…

If one thinks all they need to do is remove javascript to speed things up substantially, then they fit into my category of people who only know jQuery snippets only think they know enough. There are a multitude of issues that can be tweaked to speed things up more so than javascript elimination. You can switch to HTTP/2, implement server push, image compression, header compression, compressing HTML/CSS/Javascript dow…

And none of the things you mention will amount to anything if the site is still pulling dozens of trackers and ad scripts.

The average JS dev (like average anything dev) may not write performant code, but the functional parts of the website are not where the main problem lies.

Post reply on HN