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

How much longer will we trust Google’s search results?

theverge.com

Post reply on HN
Print view View original

161–170 of 284 posts

Re: How much longer will we trust Google’s search results?

#161

Post by netcan »

jborichevskiy wrote: post #13

> Recode’s Peter Kafka recently interviewed Buzzfeed CEO Jonah Peretti, and Peretti said something really insightful: what if Google’s ads really aren’t that good? What if Google is just taking credit for clicks on ads just because people would have been searching for that stuff anyway? I’ve been thinking about it all day: what if Google ads actually aren’t that effective and the only reason they make so much is bill…

"Good" is hard to define, when we're talking about monopolistish scenarios, at least in this era.

Google's ads make good money because:

(1) The overture model for running a search ad market works well (Google acquired and adopted a mature system). (2) targeted ad markets get better with scale. The larger the market share, the smaller a& more targeted a segment is. (3) Ads are also search results. Over time, Google has nudged users to select ads... on high value searches.

On that last point: The value of a click varies in orders of magnitude 18c for this one $12 for that one. In the early days, this meant you could buy clicks for pennies. These days, Google limits the supply of low value ads and increases the supply of low value ad clickers.

On average, users might even see or click on fewer ads than 10 years ago. On high value search terms, advertising prominence & click-throughs are through the roof.

If Google is "playing dirty," this is where the incentive is highest... and achieves the goals which Google has strategically used to grow revenue for years now.

Playing dirty is even a weird term here. A newspaper may have a journalistic ethos whereby ad-selling squares don't get to editorialise journalism... whether they stick to it or not. By what ethos is Google not supposed to let ad selling concerns influence organic search?

Re: How much longer will we trust Google’s search results?

#163

Post by geddy »

They've been terrible for years. I yearn for the days when you'd look up "x video game review" and you'd get actual human beings with personal sites showing up, and opinions from random people. Now that SEO has been beaten into a pulp, the first 8,000 pages are either major corporations or spam.

So the modern internet in general, I suppose. Corporations in spam. Never mind, it'd be unreasonable of me to expect anything else.

Re: How much longer will we trust Google’s search results?

#164

Post by xp84 »

sdoering wrote: post #139
takk309 wrote: post #94

Earlier quoted context omitted.

My wife does this all the time. If she wants to go to Prime video she types "Amazon Prime Video" into her address bar then clicks the first Google result, often the ad for Prime Video.

If I do not like a company but need to visit their site, I also do this just to increase the cost for said company as Google Ads are 'cost per click'.

^ This works really well except for if the company you don't like is Google :D

Oh, I tried an alternate approach and it didn't work... Search bing for Google and you just get a second bing. https://imgur.com/a/YntBiYc

Re: How much longer will we trust Google’s search results?

#165

Post by d1zzy »

grecy wrote: post #21

Does anyone here even really use Google's search engine anymore? I do search for stuff a lot, but to be honest I know perfectly when it will be a stackoverflow page I want, or a wikipedia page, or it will have that link to maps. I can't remember the last time I searched for something and I didn't already know what domain I actually wanted to end up on.

> Does anyone here even really use Google's search engine anymore?

Is this question supposed to be rhetorical?

https://lmgtfy.com/?q=google+search+market+share

Re: How much longer will we trust Google’s search results?

#166

Post by jiveturkey »

JohnFen wrote: post #108
jiveturkey wrote: post #73

Earlier quoted context omitted.

No they aren't and never did. They don't, for example, install a supercookie (like AT&T has done) so that they can track you the person vs "you" some uniquely identified user. Where are the pitchforks for AT&T? To the extent they do collect as much data as they can, they limit it to identify you as a unique user, not you as a specific person. In the cases they necessarily identify you the person (e.g. Google Pay wher…

> No they aren't and never did. That you can say this make me think that we are using different definitions. Have they also stopped buying real-world credit card usage and correlating it with your other actions? > To the extent they do collect as much data as they can, they limit it to identify you as a unique user, not you as a specific person. That doesn't really make it much better, in my opinion. > They do an exc…

> Have they also stopped buying real-world credit card usage and correlating it with your other actions?

Thank you. This is a really great example of pitchfork-ism. People want to see things that aren't there. Their desire to see those things blinds them to what is actually happening.

Google doesn't and never did what you are implying. Yes, they correlate those 2 sets of information. But neither side of the transaction (CC, Google) knows or can know the identifying linkage in the other direction. It allows the vendor (advertiser) to get conversion rate of online ads to real world purchases, without revealing who purchased them to any of the 3 parties. It isn't that Google or the CC issuer has to anonymise the data after they correlate the linkage, so as to hide the linkage from the advertiser; no party has the other piece of data in the first place, thanks to homomorphic encryption. It's actually a wonderful application of technology, being decried for completely wrong reasons. We need more of this specific technology, not less. (to the extent we need this at all)

> I object to the collection.

As do I, and I stated as much. My anti-complaint is that there is a huge backlash against Google when Google aren't the bad guys. There is a whole industry of actual bad actors -- doing "bad" things with the data they collect. vs Google that treats the data as sacrosanct, and they use it to actually provide a service to you. The other folks are simply stealing your data and reselling it for their own ends, without providing anything useful to you.

Look, I'm not saying Google is great, or even good. As I said here and elsewhere, they are evil. But evil people (entities) can do good things. We (collectively) have all agreed to Google's business model. And Google is acting as a responsible steward for the information they collect. There are much, much worse actors out there that are far more deserving of the spite that is sent in Google's direction. (only re: privacy specifically ... again, they are generally evil and deserve to die the death that will come to them and none too soon)

Re: How much longer will we trust Google’s search results?

#167

Post by JohnFen »

Japhy_Ryder wrote: post #117
JohnFen wrote: post #50

Earlier quoted context omitted.

I don't use Google's search engine, but I do use search engines heavily. Unlike you, I rarely know what site will have what I'm looking for in advance -- it's rarely SO, Wikipedia, or maps.

Just out of morbid curiosity, _what_ kind of things are you usually looking for?

A very wide variety of things, actually. I have eclectic tastes. But most often, it's programming or electronic-related stuff.

Re: How much longer will we trust Google’s search results?

#168

Post by xp84 »

praxulus wrote: post #145
smichel17 wrote: post #76

Earlier quoted context omitted.

I've seen people type "example" into the address bar, hit enter, then click the Google link to example.com, without realizing that they could just add ".com" to the end of their initial query and go directly to their destination.

If you have a typo in the name, Google will fix it for you, whereas your browser will happily take you to a domain that is probably owned by a scammer.

One could argue that since Google is JUST as happy to take money from scammers for an ad for the keyword WellsFrago as the Registry would be to sell the domain WellsFrago dot com ... I'm not sure either approach is technically safer than the other if you're prone to typos. I think (gasp) bookmarking the important sites you need is the only safe approach.

Re: How much longer will we trust Google’s search results?

#169

Post by moultano »

> Last week we updated the look of Search on desktop to mirror what’s been on mobile for months. We’ve heard your feedback about the update. We always want to make Search better, so we’re going to experiment with new placements for favicons….

https://twitter.com/searchliaison/status/1220768238490939394

Re: How much longer will we trust Google’s search results?

#170

Post by lawlorino »

jborichevskiy wrote: post #13

> Recode’s Peter Kafka recently interviewed Buzzfeed CEO Jonah Peretti, and Peretti said something really insightful: what if Google’s ads really aren’t that good? What if Google is just taking credit for clicks on ads just because people would have been searching for that stuff anyway? I’ve been thinking about it all day: what if Google ads actually aren’t that effective and the only reason they make so much is bill…

> What if Google is just taking credit for clicks on ads just because people would have been searching for that stuff anyway?

At least for my company, we run blackout experiments and measure/estimate the incrementality of the ads.

Post reply on HN