Is online advertising working? We simply don’t know.
November 14, 2019 9:08 AM   Subscribe

 


People really do click on the paid-link to eBay.com an awful lot. But if that link weren’t there, presumably they would click on the link just below it: the free link to eBay.com. The data consultants were basing their profit calculations on clicks they would be getting anyway.

I don't intentionally google ebay to get to ebay (or amazon to get to amazon, or whatever), but since the search bar and the location bar merged, sometimes it happens accidentally because of a typo or bad formatting or whatever when I meant to just enter the url. I do sometimes search for more obscure companies where I don't know the url off the top of my head. And there's always an ad for the precise thing I just searched for immediately above the link to the thing. I always stop for a moment and think about whether I like Google or the thing I was searching for more before deciding whether to use the native search link or the click ad.
posted by jacquilynne at 9:30 AM on November 14, 2019 [23 favorites]


I’ve been saying this for years. The whole thing is smoke and mirrors, the Internet “makes money” either via individual contributions or some massive investment funding with a reality distortion field around it. No one is willing to admit how blown glass fragile the whole thing is and are just hoping to have a chair when the music stops.

And then only the monopolies will remain.
posted by The Whelk at 9:37 AM on November 14, 2019 [48 favorites]


A few years ago, I was told that the most common search term on Bing by far was "facebook".
posted by Slothrup at 9:42 AM on November 14, 2019 [6 favorites]


For more than a century, advertising was an art, not a science.

No, advertising is the anti-matter to art's matter. Advertising is where art goes to die, where it goes to let companies abuse the power of the only good thing about humanity, to further the aims of the worst thing about humanity. That was before they brought psychology into advertising, now it's an outright cartoon villain's field.

I find the magic analogy much less disheartening. Advertising is magic-- eldritch magic, forbidden magic, mind-deforming psychic magic, magic that requires sacrifices, magic that saps the life and minds of those it targets while enriching the nefarious cabal of capitalmancers casting their unholy curses.

I’ve been saying this for years. The whole thing is smoke and mirrors, the Internet “makes money” either via individual contributions or some massive investment funding with a reality distortion field around it. No one is willing to admit how blown glass fragile the whole thing is and are just hoping to have a chair when the music stops.

It really goes how there's really just too much money floating around in the wrong hands. Personally I feel the sites and services people use to share all sorts of art and creations online should be public utilities. Fuck the advertising, fuck trying to make youtube profitable, fuck trying to capitalize on people who want to share social media about their lives to their friends... This shit clearly has a demand deeper than what makes sense for businesses to manage. The internet is for the people.
posted by GoblinHoney at 9:51 AM on November 14, 2019 [55 favorites]


"The worst, dumbest outcome (and therefore the one likeliest to be true, by the iron law of 2019) is if Facebook advertising doesn't actually work for anything except politics"

If the veracity of commercial advertising was regulated as little as political advertising, political advertising would work great! You could say whatever the hell you wanted.
posted by RobotVoodooPower at 10:01 AM on November 14, 2019 [2 favorites]


Also Advertising is nearly impossible to study effectively, it’s closer to alchemy but with a pyramid scheme built in.
posted by The Whelk at 10:02 AM on November 14, 2019 [8 favorites]


I'm not sure that running particular web sites as public utilities (aside from legitimate .gov stuff) makes a ton of sense; I'm not sure I'd trust Trump's Facebook any more than I trust Zuck's version. Although when you pare away the advertising, what you'd have left might not resemble Facebook very much, I suppose.

Rather, what would make a lot of sense is to de-privatize the Internet infrastructure, including (and perhaps most importantly) the last-mile infrastructure that delivers content to people's actual homes and businesses, and which uses public rights-of-way and public spectrum anyway.

Give everyone a 100MB symmetric connection—which is actually pretty modest by First World ex-US standards—and you could have self-hosted and peer-to-peer services, rather than the client/server architecture we're stuck with now and its easily-monopolizable server farms. Set up the network correctly and you wouldn't need constant state intervention to buy up the latest popular site, instead you'd get properly decentralized, democratized services from the beginning.

Today's Internet is basically designed for cable TV over IP. It's highly directional, designed to feed content from server farms down to eyeballs, with very little (other than content selection) going in the other direction. That's fundamentally a problem, and it's not how the pioneers of the Internet thought it would or should be. If we fixed that architectural issue, a lot of things would be possible that are currently not practical.
posted by Kadin2048 at 10:08 AM on November 14, 2019 [38 favorites]


One of the companies I work for just took their whole entire ad spend and moved it to digital - facebook, instagram, and youtube mostly. As I review copy, I don't dare share my feelings that approximately zero people will pay sufficient attention to these to make them worth it. As ads have filled more space on the screen and in the feed, we've gotten awfully good at wholly filtering them out.

(I am amused to see the Instagram mockups from our agency that show "749 likes" on our fake meme, when most of our real ads have something closer to 7.49)
posted by AgentRocket at 10:09 AM on November 14, 2019 [9 favorites]


This is interesting because it focused on the case studies of some really large firms (like eBay). eBay is famous! It's a word in most people's vocabularies.

In a previous life I did work with online marketers doing dropshipping (yes, sleazy!) and it was definitely the case that once you turned on the ads, the traffic picked up, and if you turned off the ads, it slowed to a trickle. And the remarketing features that facebook (for example) offers, (where once you've visited the shop they just bombard you with reminders that you've visited, that you added something to your cart but didn't check out, that you looked at X product, etc., and would you like to buy it now?) did seem to increase conversions.

It's hard to say that online marketing is doing *nothing*, but it might have limits to its effectiveness, especially with well established brands. It's like when I get ads for hulu while I'm using hulu. I know about this already, hulu, and you know that I know.
posted by dis_integration at 10:11 AM on November 14, 2019 [20 favorites]


I work in this area, and my best argument is always, graph it up rather than obeying tracking. Did sales go up when you spent a lot of money? If yes, Great. if nothing really seemed to happen, you’re probably cannibalising.
Repeat a few times, rule out other effects, and if it’s always cannibalising, pick the cheapest channel, and see what happens over a few weeks.
posted by eyeofthetiger at 10:17 AM on November 14, 2019 [9 favorites]


Advertising is art, sorcery, alchemy, psychology? Surely it's just sociology, no?
posted by Easy problem of consciousness at 10:20 AM on November 14, 2019


Rao awarded prizes to cult members. Those willing to accept their prize after Judgement Day – when the world would be annihilated and the faithful would ascend to heaven – were promised more money. Their belief in the apocalypse proved uncompromising. Even an extra 500 dollars couldn’t seduce the cultists. "Beliefs formed on insufficient evidence seem tough to move," Rao wrote.

This seems like it would be less surprising if it were framed as "beliefs people have built into their self-narratives." The entire managerial class has invested itself in a belief in the power of advertising and The Cyber to the point that any merely empirical challenge meaningless.
posted by Not A Thing at 10:20 AM on November 14, 2019 [7 favorites]


Yep. "It is difficult to get a [marketer] to understand something when his salary depends upon his not understanding it." Not to mention his sense of self as a critical mover-and-shaker in the company, a vanguard of the future, a swayer of mass sentiment, and a MOTU in general.

But advertising agencies aren't selling products, they're selling advertising. They've been good at that, because power projection makes the projector feel important; power projection is the heart of what propaganda is; and ads are propaganda for corporations.
posted by Tess of the d'Urkelvilles at 10:36 AM on November 14, 2019 [14 favorites]


It perpetually amazes me that with the vast amount of money spent on on-line ads, how often they advertise stuff I don't want and can't want because I just bought it.

I bought a TV on-line years ago. My ads for the next few months? TVs. No, I don't want a goddamn TV, I just bought one. How can billions of dollars be spent on this stuff and simple things like this get neglected?

I bought my daughter a desk for her to do her school work. Now I see ads for desks. What. The. Hell? We just went over this.

It seems that advertising should either reinforce brand awareness or my deep, abiding love for the brand or inform me of stuff (goods or companies) that I didn't already know about. Not tell me "You just bought diapers. Here're some ads for diapers! Yay!"
posted by It's Never Lurgi at 10:55 AM on November 14, 2019 [45 favorites]


It's hard to say that online marketing is doing *nothing*, but it might have limits to its effectiveness, especially with well established brands. It's like when I get ads for hulu while I'm using hulu. I know about this already, hulu, and you know that I know.

yeah, this. it's definitely a mistake to assume that online advertising works the same for a giant enormous brand as it does for a tiny little whatever. like do these people not know any independent creators?

for an enormous brand, attribution is basically nonsense; they just don't want you to forget about them. they do the graph thing if they're not dumb. you don't have to understand WHY advertising on a channel or with certain content or whatever is helping you for it to help you, you just have to know that when you stop doing it you start losing money.

for NOT enormous brands? advertising is about discovery. there's no "they would have clicked on you anyway," because they wouldn't have heard about you without the ad.

a whole bunch of people who sell books at 4.99 have kept their careers viable with advertising. and these people are not dumb. they do all the testing. a lot of them had careers in exactly this before they started self-publishing. and what should not be a surprise is that the people with marketing and advertising backgrounds seem to sell a shit ton more books than people who don't have those backgrounds, even when their books are...call them "less good."

I know this isn't relevant for the bubble argument, because Facebook isn't getting most of their money from small businesses who need targeted advertising for discovery. it IS kind of funny to me that targeted advertising is mostly only useful to the small fish who, on their own, can't possibly support the massive advertising-based platforms, but in a grimdark kind of way?

But I don't think it matters, because advertising is all we have. It's literally the only thing you can do to try to get people to buy your thing or support your whatever. You have to buy access to attention, whether you do that directly or you do that through a shit ton of grunt work. And it works often enough that people will keep trying. It's not going anywhere.
posted by schadenfrau at 11:06 AM on November 14, 2019 [19 favorites]


If people were easier to manipulate with images and videos they don’t really want to see, economists would have a much easier task.

Also, this circular logic bothers me. Of course showing people things they don't want to see isn't effective? That is...the entire point of targeting? Literally the entire rationale is "we'll show these ads to people who are interested in the product."

Whether the targeting works is a different question, but that's as much in the hands of the people choosing the targeting demos as it is the platform. Like if you're trying to sell religious baby crap to religious people who just had babies, and you are targeting Bernie Sanders voters, you are...not good at your job. (It's an open question whether Facebook's targeting is accurate, but, well, enough people have success with it that it's probably accurate on some metrics.)

Anyway. To paraphrase Gossage: people click on things that interest them. Sometimes that's an ad.

Whether you can make "sometimes" = "often enough" to be worth the money is usually a question of targeting and content.
posted by schadenfrau at 11:15 AM on November 14, 2019 [2 favorites]


There's a reason advertising copywriters are often referred to as "content creators" nowadays.
posted by SoberHighland at 11:16 AM on November 14, 2019 [2 favorites]


I've subscribed to Time magazine for +20 years. Nearly every page has an ad, that I ignore completely. And yet that's enough to fund Time magazine. Why aren't static ads online given the same courtesy (which don't even have printing and distribution costs)?
posted by Melismata at 11:17 AM on November 14, 2019 [6 favorites]




An anecdote: I freelanced as an art director at a big PR firm. They handle McDonald's, who was in the process of implementing a big menu change. One of the pieces of their PR effluvia was a video of the CEO giving a speech at some conference. Nicely edited down, digestible (heh) piece meant for business purposes, in other words, not a Super Bowl ad.

A young account person there told me with a straight face that the video was doing well and had "over seven billion impressions" at the time.

Goats, entrails, meaning. To be fair, many people in ad agencies have a similar level of skepticism as the replies in this thread. It's simply that slaughtering bulls for Mithra pays really well!
posted by SoberHighland at 11:25 AM on November 14, 2019 [9 favorites]


Regarding "humidifier collecting" it is true that when people buy new, shiny TVs and such, they tend to talk about them a lot. And then their friends and family member buys one partially based on the excitement. Or the purchaser buys one for his mom, or whatever.
posted by SoberHighland at 11:28 AM on November 14, 2019 [2 favorites]



Today's Internet is basically designed for cable TV over IP. It's highly directional, designed to feed content from server farms down to eyeballs, with very little (other than content selection) going in the other direction. That's fundamentally a problem, and it's not how the pioneers of the Internet thought it would or should be. If we fixed that architectural issue, a lot of things would be possible that are currently not practical.


Just my opinion, but the dream that computers and then the internet would open up a world where everyone could be a creator, was just a dream. Everybody just wants to watch TV and consume.
posted by njohnson23 at 11:28 AM on November 14, 2019 [18 favorites]


"The best minds of my generation are thinking about how to make people click on ads."

This is not only demonstrably false, but shows how inflated tech bros' egos truly are.

Rather uninspired, dull and witless "minds"—at best.
posted by Ahmad Khani at 11:30 AM on November 14, 2019 [12 favorites]


Advertising is where art goes to die

Counterpoint: nacho fries.
posted by grumpybear69 at 11:36 AM on November 14, 2019 [2 favorites]


This article completely ignores direct-response marketing which predates internet marketing by many years. It's not as immediate or quick, but it works. There's a reason we all get so much "junk mail." Marketers segment that stuff and get feedback on style, wording, format, images and messaging based on response rates. The concept is newer than billboards, print and TV ads, but it existed in the dark ages before dial-up.
posted by SoberHighland at 11:41 AM on November 14, 2019 [2 favorites]


Counterpoint: nacho fries.
Double counterpoint:
Skittles Pinata Man
Does it make me want to buy Skittles? No. Does it increase my awareness of Skittles? Well, the ad is 11 years old, so it totally increased my awareness of ads about Skittles. But to me it is art - excellently produced, and hilariously scripted and exists as its own thing outside of the product it is advertising.
posted by The_Vegetables at 11:42 AM on November 14, 2019 [6 favorites]


Is this where I can complain about the creative maker-types I follow on YouTube shilling for something called "skill share" when as far as I can tell the service offered by skill share is basically what these folks are doing for free on YouTube?
posted by St. Oops at 11:45 AM on November 14, 2019 [9 favorites]


Just my opinion, but the dream that computers and then the internet would open up a world where everyone could be a creator, was just a dream

I remember when the big line about the Internet was it would remove a lot of middlemen and rent seekers and now it’s just a machine that makes more middlemen
posted by The Whelk at 11:46 AM on November 14, 2019 [45 favorites]


I bought a TV on-line years ago. My ads for the next few months? TVs.
A theory I once heard that sounds plausible: a not-insignificant number of TVs and similar large purchases fail early or are otherwise returned. Thus, one of the single most likely indicators that someone wants to buy a TV is that that person has recently bought a TV.
posted by Hatashran at 11:50 AM on November 14, 2019 [8 favorites]


does that apply for everything though? anytime i buy something online from books to socks to a gizmo i get ads for that exact thing i just bought.

it’s not like i bought Death on the Nile by Agatha Christie and i get ads for Amazon’s recent Christie adaptation or a different Christie book.

i get ads for Death on the Nile.

i do not understand.
posted by affectionateborg at 11:53 AM on November 14, 2019 [11 favorites]


it’s not like i bought Death on the Nile by Agatha Christie and i get ads for Amazon’s recent Christie adaptation or a different Christie book.

i get ads for Death on the Nile.


The best minds are not working on this stuff. Is it easier programatically to look up what you bought, identify the ad for that product, and show it again? Yes. So that's what is done.
posted by The_Vegetables at 11:57 AM on November 14, 2019 [5 favorites]


I almost made a post about this article entitled The Google Squeeze, but I'll share it here instead. It's a thoughtful long piece about the increasing number of paid spots showing up in Google search, focusing on the example of Google as a middleman for people looking to book travel.
posted by exogenous at 12:00 PM on November 14, 2019 [4 favorites]


Just my opinion, but the dream that computers and then the internet would open up a world where everyone could be a creator, was just a dream. Everybody just wants to watch TV and consume.

I would argue the exact opposite is true. Over 400 hours of video is uploaded to Youtube every minute. There are billions upon billions of drawings between the major art sharing sites. "Soundcloud musician" is almost a synonym of "average person". I don't even know how to look up the statistics on how much non-professional writing is produced, between blogs, social media posts, and more formal outlets. Art has been as close to democratized as we are going to get. The problem is not that there are no creators - the problem is that there is no way to find the creations you would want to see.

In fact, the only things out there to guide us to the content we would want to consume are instead used to guide us to the commercial products we do not want to consume.
posted by FakeFreyja at 12:01 PM on November 14, 2019 [13 favorites]


It's not an ad for Agatha Christie, folks. It's an ad for Amazon.

Forest/trees.
posted by SoberHighland at 12:01 PM on November 14, 2019 [1 favorite]


I will say, Instagram's ads are eerily and creepily relevant to my interests. To the point where I do something I've never done on any other platform -- liked their posts and followed their accounts to see more and even BOUGHT something. Luckily, what I'm liking are spooky goth t-shirts and not racist propaganda.
posted by pelvicsorcery at 12:06 PM on November 14, 2019 [9 favorites]


It's not an ad for Agatha Christie, folks. It's an ad for Amazon.

Forest/trees.



Is this an ad for not reading the article?
posted by The_Vegetables at 12:21 PM on November 14, 2019 [2 favorites]


Is this an ad for not reading the article?

Also, tangentially, I listened to a talk by a current Harvard business professor, speaking of Amazon, who surmised that Amazon doesn't care how many people watch the tv shows they are creating because the TV shows are just an added benefit of having and reminding you about Amazon Prime, the free shipping component. What a world we live in where creating and distributing Hollywood content watched by relatively small numbers of people is more cost-effective than just making an advertisement for shoes!

At least the ad people put together some metrics for the money they are spending. I bet the content producers don't' provide that.
posted by The_Vegetables at 12:32 PM on November 14, 2019 [1 favorite]


__ ___ ___ __ ______ ____ __ __ ___________ __ _________... ____ ________.
          — Bill Hicks
posted by scruss at 12:41 PM on November 14, 2019 [15 favorites]


Most companies know that a lot of their advertising dollars are wasted. They just don't know which portion of them is the waste. That was the promise of online ads, and why paid search results are still >80% of Alphabet's revenue steam.

Having worked on that thing in past careers, I don't think the results are anywhere near as much of a slam-dunk as originally promised, which is why the panopticon is increasing. "Let's look at eyes to see if they even saw the ad! Did their facial expression change?!? OMG when will the dollars finally roll in and we can stop spending all this money?!?"
posted by fifteen schnitzengruben is my limit at 12:45 PM on November 14, 2019 [2 favorites]


"We're building a dystopia just to make people click on ads." -- Zeynep Tufekci's TED Talk
posted by chrchr at 1:05 PM on November 14, 2019 [9 favorites]


There definitely exist a broad class of paid search ads that are undeniably effective for the advertisers. The most obvious class is any business that you used to select via the Yellow Pages. If you want to find someone to clean your carpets, most people will simply enter “carpet cleaners” into the search bar and check out the first few links and pick someone. You still have to create a decent web page and compete on price, but in a community where there are 40 competing firms, being one of the 3 who show up at the top of the ad stack for that search is worth a lot of money to those businesses.

My wife owns a small business where we know almost exactly what each incoming phone call is worth to us by meticulously tracking what percentage become clients and what profit the average client generates. We’ve tested a variety of search ad campaigns and are absolutely certain they bring in far more profit than they cost.
posted by Lame_username at 1:15 PM on November 14, 2019 [6 favorites]


I'm pretty skeptical of the argument here:
  1. Keyword ad-buys don't work, as proven by these experiments with large brands: If someone makes a search with the keyword 'eBay', showing a paid ad for eBay is a waste of money.
  2. Therefore the whole industry is smoke and mirrors.
The experiments were interesting. I'd be curious to see results of the affiliate marketing, "#FitGirl" tests. The authors bring them up and then drop them immediately after the juicy quote: "If we run this experiment, and the results look like what you showed us with search advertising, I’m not going to believe you." Well? What were the results? Are the tests planned? Ongoing? Concluded and confidential?

The big improvement from Lewis they mention at the end is ghost ads, which is A/B testing with a control group. From his company's site:
The Ghost Ad itself is invisible to the user. He or she simply sees the ad that actually wins the auction. However, the Ghost Ad and its bid data are visible to ad platforms and advertisers. From that they can glean how users who should have seen an ad behave when they don’t see it. Ghost Ads do the job of PSA-based A/B tests, they just do it more precisely and affordably.
Which sounds like a normal thing? Not exactly a pin to the industry's balloon of self-delusion?

There's a story here about how the staggering increase in private surveillance has been in no small part to help marketing VPs justify quarterly budgets. Unfortunately, the story OP's trying to tell is about how hard-nosed economists from outside the industry cut through all that marketing malarky to uncover hard truths. They're missing a pretty dark forest for some old dead trees.
posted by postcommunism at 1:19 PM on November 14, 2019 [6 favorites]


I'm a firm believer that retargeting with things you looked at works on some level, because it's definitely worked on me on some level. I also get a lot of ads for, like, sales on items pertaining to a couple of my hobbies and those have probably contributed to some purchases. This is all pretty low-tech stuff though, based on my proven past purchases and interests.
posted by atoxyl at 1:19 PM on November 14, 2019 [1 favorite]


I don't think I've clicked on a search keyword ad in my life, though. I feel like those are really for people who can't tell which are the paid results?
posted by atoxyl at 1:20 PM on November 14, 2019 [1 favorite]


It's not an ad for Agatha Christie, folks. It's an ad for Amazon.

There is an item from a retailer (not Amazon) that I pretty desperately wanted, but which has been "temporarily out of stock" for some months. Periodically, I check the item page to see if it is back in stock. As a result, I continue to get ads featuring the thing, which in turn reminds me that I can't get this thing, and all my angry feelings toward the retailer. So while it is an ad for the retailer, it is not a good ad for the retailer.
posted by anastasiav at 1:29 PM on November 14, 2019 [3 favorites]


Re: the "humidifier collecting" (or TVs, or refrigerators...)

There are actually three (or more) complementary reasons that people run those ads.

1) Sometimes, the retargeting algorithm knows that you've searched for the thing, but isn't actually connected to the retailer you wound up buying it at, so for all it knows, you are still looking.

2) Hatashran mentioned this above, but a significant-enough percentage of large purchases get returned that even if the algorithm knows that you bought the thing, you're still probably slightly more likely to want to buy another than the average pair of eyeballs on the internet.

3) Obviously, neither of those apply to Amazon's internal advertising, since Amazon absolutely knows whether you bought the thing, and knows whether you've returned it. But, amazingly enough, being someone who has purchased a humidifier/fridge/tv/printer actually *does* make you more likely than the average consumer to purchase another again soon. Not hugely, perhaps, but it happens enough that companies like Amazon can definitely measure it. Lots of people wind up buying things as gifts, for relatives, or just because they now want another item for a different room (or second house). This is obviously not true anything like 100% of the time, but the thing about modern digital advertising is that it operates at such absurd scale that even fractions of a percent can add up.
posted by bluemilker at 1:34 PM on November 14, 2019 [4 favorites]


Advertising to me has always been a faith-based exercise. Faith in ages-old business advice and practice is the underpinning for conventional business wisdom.

Now, in the age of the internet, the faith is placed in qualitative measurements. But without a working behavioral framework, stats and numbers are just window dressing. Nothing really changes.
posted by ZeusHumms at 1:38 PM on November 14, 2019


If I use Google to look for the web site for the thing I am looking for and the direct link is not presented, I try another search engine or just extract the true web address from within the crap link returned.
posted by Burn_IT at 1:43 PM on November 14, 2019


I'm not sure that running particular web sites as public utilities (aside from legitimate .gov stuff) makes a ton of sense; I'm not sure I'd trust Trump's Facebook any more than I trust Zuck's version. Although when you pare away the advertising, what you'd have left might not resemble Facebook very much, I suppose.

Kind of a perfect time to test your values, I've never trusted the government less and don't even consider major parts of it to even be legitimate at he moment. I still trust an entity that is meant to serve me and the people, even if it's currently corrupted and deeply failing in it's primary function. The primary antagonist of government and why it is so bad right now, is businesses and capitalistic-forward practices. The government is corrupt, sure, but that doesn't make turning to their corrupter any better. In principle, the government is for the people. In principle, capitalism is against people. That's very reductionist but at the end of the day that's how I see the push-and-pull of government & business vs. actual human people.

As for paring away things from Facebook, I'd say Facebook doesn't even really resemble Facebook anymore. There was a time when it was basically just people swapping personal information around. In any case, until people come to own their own personal information they generate, and must be paid when it is used, scraped, or otherwise made productive, advertisers have us by the short hairs.
posted by GoblinHoney at 2:11 PM on November 14, 2019 [1 favorite]


I remember when the big line about the Internet was it would remove a lot of middlemen and rent seekers and now it’s just a machine that makes more middlemen

brb getting first round funding for a startup making machines that make machines that make middle men
posted by PMdixon at 2:30 PM on November 14, 2019 [5 favorites]


Every single number associated with the internet should be met with skepticism. At Patch, eventually owned by AOL, there was relentless and contradictory pressures to meet a new number: unique views were the path to financial success; no, this week, it's user engagement; no, we meant, THIS week, the number that would mean glory was returning users; no, page views? I run my own site now, and I'm bombarded endlessly by companies measuring different stats and how preaching about which stat would mean success. It's all a house of cards.
posted by etaoin at 2:37 PM on November 14, 2019 [3 favorites]


The 'ads show me what I've just bought' thing seems pretty clear to me - maybe Amazon doesn't actually share your purchase history with advertisers, because that's valuable data and also there's privacy implications. Whether you've looked at a product, well that's a different story. And the Buy button is on the page, so you have to 'look' at a product in order to buy it, so the only data advertisers can get is flawed and they don't the ability to correct it.
posted by Merus at 2:38 PM on November 14, 2019 [2 favorites]


I remember when the big line about the Internet was it would remove a lot of middlemen and rent seekers and now it’s just a machine that makes more middlemen

I swear there was one dial-up night in the 90s when I read an "end of middlemen" article and then clicked to a major publisher's website to find them selling books at retail, not wholesale, price, and that's when I knew once again that there was no Santa Claus.
posted by mattwan at 2:46 PM on November 14, 2019 [5 favorites]


I asked this of a google recruiter in college and it still feels true (and rather brain-dead obvious with examples like an ad for ebay at the top of a search for ebay)

The dream that google aspires to for search is that no matter what you might search for, google returns exactly what you need to find. This has an inherent contradiction with advertising that other ads based mediums (like radio or newspapers) never had. When you search for something on google (or the app store, or bing, or whatever) ads subvert the legitimacy of the organic results because they're related to your query but for a different reason than the organic results are. Googlers will talk abut how great search ads because they are "more relevant" than ads in the paper but if they were really relevant, they would be in the organic results. At least when I hear ads in podcasts they aren't trying to pretend to be relevant to the subject of the episode itself.
posted by macrael at 3:11 PM on November 14, 2019 [7 favorites]


"brb getting first round funding for a startup making machines that make machines that make middle men"

It's like Uber, but for middle men
posted by klangklangston at 4:26 PM on November 14, 2019 [8 favorites]


Two brief anecdotes: I used to work in journalism, and every so often would want to start my own publication. But one fundamental barrier was always that I don't really believe that advertising works on any measurable scale, and is better understood for most brands as a status signal — we are lucrative enough to throw this money at something whose cache will provide us with a halo.

This is not a winning pitch to advertisers.

I also used to work as a communication manager at a large non-profit, where I worked on a bunch of digital campaigns, and was regularly frustrated by the inability of management and board members to understand that the metrics they cooed over were absolutely useless to the actual work of the organization. Getting likes on Facebook was entirely unconnected with either raising money or passing laws or even mounting public awareness campaigns. But what's most effective in getting volunteers or donors is actually both a lot of work itself, and a lot of work to accurately collect data on. Bragging about adding 1000 likes in a month is easy.

OK, so three anecdotes: A buddy of mine on Facebook, he's a musician who used to run his own label before he went into academia. He was saying that the only way to be successful as an artist under capitalism is to build a community that values the art you're making, and that community is the only thing that's monetarily worth anything with art. I think you can expand that out as the default ethos that a lot of modern marketing works under, with brand communities and affinities and identities. I tend to find that really creepy, but it's definitely a mindset among marketers, and can shift a lot of the discussion on the goals of advertising.
posted by klangklangston at 4:39 PM on November 14, 2019 [11 favorites]


A few years ago, I was told that the most common search term on Bing by far was "facebook".

I would have bet money the number search on Bing was how to change the default search engine to Google.
posted by DirtyOldTown at 5:52 PM on November 14, 2019 [4 favorites]


The dream that google aspires to for search is that no matter what you might search for, google returns exactly what you need to find. . . . Googlers will talk abut how great search ads because they are "more relevant" than ads in the paper but if they were really relevant, they would be in the organic results.

Yes but the part that Google can't control is the faulty search terms that users put in. If everyone were a perfect search term creator, Google could return perfect results, which would kill off their ad business. But people put in extraneous words, they leave out necessary words, the misspell more than just a little, they fail to use tricks like putting phrases in quotation marks, etc. Search ads often use keywords that anticipate these failures, which improves the results for imperfect search terms. So really, Google's business depends on people not being very good at search.
posted by beagle at 6:01 PM on November 14, 2019


To think about keyword advertising, think about starting a business from scratch with only a cursory knowledge of office stuff. To get desks and paper and computers, etc., I'd probably start with one of the office supply stores and work out from there. But to figure out who is going to clean the carpets? I'm clicking on the first three janitorial services I see and checking their rates. Likewise, a locksmith, when you're calling, you're usually in a jam (this is where a common scam comes in actually), you are not looking for a name brand locksmith.

I remember being told once (I don't know if this is still true) that the most expensive term to advertise for was mesothelioma. Because anyone googling that who is not looking it up for school probably has it or knows someone who does. And mesothelioma is a nearly guaranteed settlement for whichever law firm gets it. I wish I could see some studies on generic industry stuff. Lawyers, plumbers, janitorial services, funeral homes. People use the address/search bar to search for google so they can then enter their query. Looking up a name brand isn't going to make much a difference. Finding a local auto mechanic, that's something else.
posted by Hactar at 8:22 PM on November 14, 2019 [3 favorites]


I have to say that for the small advertiser it works at about the same effectiveness as putting up posters on kiosks on the main drag in a college town. Can I sell out or fill a live show, or a beer tasting, or a book launch with Facebook ads? Almost every time and for not all that much money. That end of the process is fine because if I want to reach a crowd in a certain place with certain inclinations I can do that, and since I know how to give them images and text that they'll want to see, it works.

It's when potential customers that don't know me yet are looking for something like me or my stuff that the system sucks.
posted by Mo Nickels at 8:23 PM on November 14, 2019 [3 favorites]


Ebay probably wants that top advertising link to prevent another auction site from having it.
posted by amtho at 10:08 PM on November 14, 2019 [1 favorite]


Ebay probably wants that top advertising link to prevent another auction site from having it.

Keywords are not exclusive. So if someone else starts using it (a) Ebay can move back in with a higher bid and push them off the front page of results, and (b) Google may not even show it if the link has insufficient quality score relative to the keyword.
posted by beagle at 7:47 AM on November 15, 2019 [1 favorite]


Now, in the age of the internet, the faith is placed in qualitative measurements.

I think I intended this to be about "quantitative measurements". Either way, in this era of human history, we use numbers as signifiers of legitimacy and rightness.
posted by ZeusHumms at 8:38 AM on November 15, 2019 [1 favorite]


The best minds of my generation

oh, honey
posted by dusty potato at 9:27 AM on November 15, 2019 [2 favorites]


For more than a century, advertising was an art, not a science. Hard data didn’t exist.

Hmm... how about this example of hard data collection later in your article?
Picture this. Luigi’s Pizzeria hires three teenagers to hand out coupons to passersby. After a few weeks of flyering, one of the three turns out to be a marketing genius. Customers keep showing up with coupons distributed by this particular kid.
posted by Emmy Rae at 10:35 AM on November 15, 2019


Keep that in mind the next time you read one of those calamity stories about Google, Facebook or Cambridge Analytica. If people were easier to manipulate with images and videos they don’t really want to see, economists would have a much easier task.

This is also a claim I'm skeptical about. Buying something is not the same as believing something. Even getting someone to believe "Hillary Clinton runs a pizzeria basement crime ring blah blah blah" is more difficult than simply raising questions about her trustworthiness - that can be accomplished simply through endless headlines. No click required.

Otherwise, based on the very ridiculous advertising attempts I've seen online, it is not surprising that it's a pretty mysterious and in no way reliable industry.
posted by Emmy Rae at 10:49 AM on November 15, 2019 [2 favorites]


Ok one more thing: I know this is not the same as the "do you want to buy 7 more TVs?" phenomenon but I was shopping online the other day and I put a belt in my bag. Then the website showed me a screen saying "complete the outfit" and the products they recommended were 3 more belts. So get ready for a rather scandalous and confusing complete outfit! Can't wait!
posted by Emmy Rae at 10:51 AM on November 15, 2019 [9 favorites]


Then the website showed me a screen saying "complete the outfit" and the products they recommended were 3 more belts. So get ready for a rather scandalous and confusing complete outfit!

Maybe they think you're a Final Fantasy character.
posted by tobascodagama at 11:08 AM on November 15, 2019 [7 favorites]


Speaking of targeted ads, not long ago I was on Amazon and was appalled to see that Amazon was suggesting that I might be interested in purchasing a product that is associated with a very slightly less than 100 percent legal product that some people—not me!—consume to enhance their imaginations. Here's the thing: I knew I had never bought that item online. Indeed, I've made a point of only ever buying it in person, in cash, precisely because I would never want anyone to think that I am involved in such a shady activity—which let me repeat, I most certainly am not! And yet, Amazon's algorithms had figured out—falsely!—that I am involved in this shady activity that I would never have anything to do with.

There's going to come a time when nobody will be able to get away with anything. Not that that won't be a very good thing!
posted by Transl3y at 1:55 PM on November 15, 2019 [2 favorites]


I laughed at "new hobby of humidifier collecting", but maybe they are doing us a favor. I remember reading years ago about an (to me) amazing finding about ads in photography magazines. Some kind of marketing research instrumentation reported that people spent the most time looking up, perusing, spending time with, ads for cameras that they had just purchased! They had a term for it, like "confirmation bias" but not that, exactly. Some kind of comforting ego-stroke, apparently.
posted by Chitownfats at 3:34 PM on November 15, 2019 [2 favorites]


If you like this, you should really be reading Ad Contrarian.
posted by rednikki at 2:05 AM on November 16, 2019


Early in Dreamwidth's history, the founder wrote Why Monetizing Social Media Through Advertising Is Doomed To Failure (part one) (part two), pointing out key differences between "ads when you are looking for a car" (which can include ads for cars, for accessories, and for places you might enjoy driving to in the car you want) and "ads when you are reading about your friends' recent school shenanigans" (in which case, ads for the school are not likely relevant to you, and ads for textbooks are right out).
Web 2.0 happened because content-providing services wanted to shed some of their costs, and the costs they chose (consciously or subconsciously) to jettison was the cost of keeping a paid editorial staff. They wanted to shift the burden of content-creating to the people who were also consuming the content, and take a cut of the profits.

The shift in conceptualization of "who makes the content", though, wasn't accompanied by a shift in conceptualization of "where the money comes from". Web 2.0 services generally keep the same model as their Web 1.0 predecessors: subscriptions, advertising, or a combination of the two. And for a whole lot of Web 2.0 services, that model works -- or fails -- about as much as it did the first time around.
The article's over 11 years old now, and still annoyingly relevant.
Advertising on the Internet is an ever-escalating war for attention. Ad-blocking technology is growing more and more commonplace, and we're in a war of attack and counterattack between advertisers and ad-blockers. But there's no technology that can be created to get around the ad-blocking capability of the human brain, no matter how much an advertiser tries.
posted by ErisLordFreedom at 9:25 AM on November 16, 2019 [4 favorites]


« Older 3D Scan of Nefertiti Bust Now Available Online   |   and now we tear one of them open Newer »


This thread has been archived and is closed to new comments