[finiteloops]

There's no long term risk to google and other ad exchanges because the advertisers are in a prisoners dilemma.

If the advertisers collude or trust that the competitors won't increases ad spend, the relative market share stays the same and all companies participating enjoy increased margins.

The first company to break gains market share, so as a result they all "overspend" and google is the only one that benefits.

The worst part is the cost of google's margin is baked into product pricing, so the end result is we pay more to have the companies compete to advertise to us.

[tmpz22]

OR ads aren't that effective in the first place and the whole industry is a house of cards propped up by middleman making their fat margins convincing people ads are necessary...

[pembrook]

I've noticed HN often has two opinions about advertising depending on the day of the week:

1) It's an unchecked, all-powerful evil, making people buy or believe in things they don't want or need by hoarding their data

or

2) It's a giant ineffective scam that stupid companies who aren't led by engineers waste VC money on

The two are diametrically opposed, yet, I've seen the same person argue #1 on Monday when it fits the narrative, and then on Tuesday start arguing #2--blissfully unaware of how both cannot be true at the same time.

Could it be, that both 1 and 2 are wrong, and that advertising spend is simply reduced during recessions in reaction to the reduction in spending by consumers? Why pay money to acquire customers when the customers aren't willing to spend money on new products?

[AnthonyMouse]

It's possible for both 1 and 2 to be right.

Suppose that if you're peddling cheap junk and snake oil, advertising is effective, because nobody will have heard of your product by word of mouth (no one would recommend it and previous victims are ashamed to admit being suckered), but if you spam enough people you'll reach enough suckers to exceed the advertising expense.

But if you're peddling a popular and quality product, everyone has already heard of it and additional advertising has low marginal utility because you were going to get most of the sales anyway.

This furthermore doesn't get you out of the prisoner's dilemma, because even if buying advertising is only break-even rather than profitable, your competitor is doing it so you have to do it too or they gain a volume advantage over you and use that to kill you on unit pricing. But then you all do it and all that happens is that everyone pays money to cancel each other out.

[cletus]

I appreciate the attempt at reconciling these this contradiction but I don't think it holds up.

For one, if you accept that advertising can sell "cheap junk" or "snake oil" then you've accepted that advertising can sell something. That could just as easily be a useful product no one has heard of so the issue isn't the advertising, it's what's being sold and advertising is effective (which invalidates (2)).

For another, you use the example of a product "everyone has already heard of". You could point to something like Coca-Cola here. But this argument has two problems:

1. There are variations companies make to keep their product "fresh". Think Vanilla Coke, Cherry Coke, Coke Zero (or whatever the current form is) and so on. By virtue of them being new, potential customers won't have heard of them and advertising solves that problem; and

2. A lot of advertising isn't about direct customer conversion but "brand lift". Now companies have dreamed of the ability to accurately measure the brand lift of advertising spend but it hasn't materialized yet.

This is also why common comments here like "I don't ever click on an ad" don't really mean anything. Now you can argue that the ability to make you desire something you don't need is "evil", which is a reasonable argument to have. I think there are cases where this is true, such as advertising to children, and these should be restricted as some countries have done.

[AnthonyMouse]

> For one, if you accept that advertising can sell "cheap junk" or "snake oil" then you've accepted that advertising can sell something. That could just as easily be a useful product no one has heard of so the issue isn't the advertising, it's what's being sold and advertising is effective (which invalidates (2)).

It can sell something unknown, because then the advertising makes it sound good, they don't know anything else about it, and they wouldn't have heard of it otherwise.

Which is the opposite of what's happening in case 2 when the product being advertised is well known. It's not causing you to hear about it for the first time and if the product is low quality then the advertising is less able to overcome your existing negative impression of it than for something you've never heard of.

> There are variations companies make to keep their product "fresh". Think Vanilla Coke, Cherry Coke, Coke Zero (or whatever the current form is) and so on. By virtue of them being new, potential customers won't have heard of them and advertising solves that problem

This doesn't really explain all the ads for Coke Classic, or for that matter why so much advertising even for new products emphasizes characteristics that are either meaningless or unrelated to the product. There isn't really any information content in telling the customer that a new cola is "refreshing" or showing random people dancing.

> A lot of advertising isn't about direct customer conversion but "brand lift". Now companies have dreamed of the ability to accurately measure the brand lift of advertising spend but it hasn't materialized yet.

"Brand lift" is the prisoner's dilemma thing. When everybody does it they just cancel each other out.

[coliveira]

The reality of advertisement for established products is that they need to do it to maintain good will from the media. For example, in the aftermath of the Deepwater Horizon oil spill in 2010, BP spent billions of dollars in advertisement in all kinds of media. It became very difficult for the media to be too harsh on BP, since in the midsts of the Great Recession, they were one or the largest advertisers. Similarly it is very difficult for the media to say bad things about Coke or the Big automakers when so much of their revenue comes from these big accounts.

Another example (this one not in advertisement) is the situation of Bloomberg. They have one of the largest financial journalism organizations and, AT THE SAME TIME, they sell overvalued software to the largest financial companies in the world. One can only guess what can happen to a financial company that stops paying the fees to Bloomberg. Even though there is no real threat (and most certainly this was never expressed by the company), every financial outlet wants to be on the good side of Bloomberg reporting. The conclusion is that having a journalist institution receiving money from companies that they are covering is a kind of moral hazard that very few people understand, unless you are part of the business.

[pembrook]

Except, in the real world there's more than just A) unknown snake oil and B) high quality products with perfect awareness.

Your model only includes products on the narrows of each extreme. In the middle is the wide spectrum of most products...the ones that don't meet either description.

Even if you make the exact same product as a competitor, there is no prisoner's dilemma if you're targeting a different niche market to sell that product to. Perfect competition does not exist in the real world. The only thing that comes close might be a commodity like oil or water. But even water can be targeted to different segments of the market.

[AnthonyMouse]

> Except, in the real world there's more than just A) unknown snake oil and B) high quality products with perfect awareness.

There doesn't have to not exist more than A and B. If A and B both exist then 1 and 2 are each true and the further existence of C and 3 don't change that.

Moreover, even if some additional classes exist, the two examples are still central and problematic, not least because they're more likely to represent a higher percentage of ad spending.

The first because advertising is the only way to sell crummy products, since the only way to get anyone to recommend it is to pay them to, so their incentives to use it are higher.

And the second because the existence of the prisoner's dilemma is what drives up the ad spend on both sides. If you're targeting a niche that no one else is then you buy a small amount of advertising, reach those customers, make your sales and are done. If you're locked in a prisoner's dilemma with a direct competitor, you spend a little so they spend a little so you spend a little more until you're all spending a huge amount. And the fact that you're selling Fords and they're selling Chevys and they're not completely identical products doesn't really matter when they're both still cars.

[ethbro]

Relevant: https://en.m.wikipedia.org/wiki/The_Market_for_Lemons

The presence of pure mercenary advertising increases the global noise floor, which increases the information asymmetry by decreasing the visibility of quality signals.

The irony is that Google sees itself as a company that increases access to information.

You know what that would look like in the ad space? Product testing, reviews, and endorsements. Something like "Verified by Google" (aka Wirecutter).

Instead, Google absolves itself of responsibility via algorithms, steered by marketing folks in charge of their primary profit center.

And we're surprised by the corporate decisions they make?

[viraptor]

Another way both can be true is that many/most companies just post an ad because they need to to get started, but don't actually optimise how it works. You can easily confirm this by looking at some discussion boards about google ads, where people often ask how to fix a very bad / accidental ad placement decisions which burned through their whole budget.

On the other hand, there are companies with teams dedicated to make their ads as effective as possible and track/raise the ROI.

There's likely lots of companies even in between those extremes which don't even break even on the ads they pay for. (or don't know if they do or not) They may pull the ads and realise nothing changed.

[Polylactic_acid]

This doesn't seem right.

>quality product, everyone has already heard of

They reached this status due to advertising, Its likely they could drop advertising for a while and coast along fine but eventually that brand recognition will start to drop.

There is also the fact that a lot of products are purchased not based on fixed needs but flexible wants. Maybe I haven't purchased a soft drink in a while and I don't think about it, an advert could make me think about it and make me want it again.

[friedman23]

It's amazing the lengths people go to reduce things into two dimensions.

popular and quality product vs cheap junk and snake oil (and presumable unpopular)

An honest person would note all possible combinations in their simplified model of businesses seeking advertising

1. popular products with bad quality

2. popular products with good quality

3. unpopular (or unknown) products with good quality

4. unpopular products with bad quality

And of course the real world is more complex than this.

your analysis is bad and wrong.

[wdevanny]

The parent comment was presenting one possible reason to explain the grandparent's observation. They were not trying to perfectly describe the real world, but only to point out that two seemingly conflicting points of view can be reconciled. I think the parent probably wrote their comment starting with the same thought as you: that the real world is more complex than a simply stated opinion.

Saying "your analysis is bad and wrong." comes off as dismissive and I think you should read the HN Guidelines.

[mywittyname]

Well, we are a collection of people with our own independent thoughts and feelings on subjects, so it shouldn't be a surprise that some people feel the opposite of others are a variety of subjects.

Though, to be honest, I don't think I've heard #2. I think any rational person understands that advertising has some degree of effectiveness. After all, many bright minds from a variety of hard sciences have spent decades of their lives, and extraordinary sums of money studying human behavior for the explicit purpose of selling more shit. Many of the largest companies in the world are advertising companies; before Google and Facebook there was the TV and Radio giants, who were massive, in spite of pretty serious regulations.

None of this would have happened if advertising wasn't effective. At some point in time, people would have realized it didn't work and spent their money elsewhere.

There are two big issues that I see:

1) More views have driven down bids for ad views. The ad industry is largely driven by companies bidding for views, so without a large influx of cash, the spike in views was going to depress ad impression prices.

2) Consumers have less discretionary money to spend, even if they wanted to, so the ROI on each view/click is going down.

Each of those alone would be cause for concern, but combined, they do pose a significant issue for Google, Facebook, and smaller content creators.

[TeMPOraL]

I sometimes say something vaguely like #2. And it doesn't conflict with #1. The argument is this:

- A lot of advertising is effective, and a lot the effectiveness boils down to lies, manipulation and general dishonesty.

- Ad attribution - i.e. tracing how much money spent on what advertising resulted in how much profit and when - is a hard problem. It's easy to make mistakes with it, and it's also easy to lie and not get caught.

- Most people involved aren't exactly experts in statistics. That's especially true for small businesses, which don't have money or institutional expertise to hire talent just to evaluate their ad spend.

- Which means people trust they aren't being bullshitted by the very industry that specializes in lies and manipulation.

I've seen this play out in real life myself; I vividly remember working next desk to social media marketers who were clueless at maths. They'd take the numbers and graphs from Facebook's panel, write up stories that made these numbers always sound like everything is going perfectly, and send such reports to the customers who were even more mathematically clueless, and thus incapable of verifying whether the numbers and the story presented add up.

The way I see it: some advertising is effective sometimes, but you don't know which one is effective and when, it's mighty hard to figure that out, and the advertisers have every incentive to confuse the issue for you.

(Note that they'll also happily confuse the issue for themselves, too. The industry consists of a lot of players building their products and services on top of each others' products and services; there's a lot of competition happening, and there's plenty of incentive to use the same advertising tactics within the industry as outside.)

[ethbro]

Maybe it's an unchecked, all-powerful evil that hoards people's data, ineffective at making people buy or believe in things they don't want or need?

[xabotage]

I find myself generally unaffected by advertising and would normally fall into the "waste of money" camp, until I started hanging out more with non-engineering types. Seeing them made me realize A) advertising does work, at least for a critical mass of people, and B) advertising doesn't take away their ability to judge a product, no matter how well targeted it is. I've even found they have a superior ability to judge products and services than I do at times because they intuitively jive with the marketing in a way that my engineering brain doesn't (they can piece together subjective qualities about a product/service being marketed to them, and I just think "show me the specs and price." Both approaches have benefits and downsides).

[codexon]

Sometimes it isn't as nefarious as convincing people to buy a product they wouldn't normally buy, and simply just showing the right people that your product exists.

[roywiggins]

I bought precisely one item from a Facebook ad, because it was the exact kind of item I had been looking for and failed to find. I was really pleased.

But it happened one time. Now I get ads for chicken feet sellers on Alibaba.

[scarface74]

I find podcast advertising be very effective and targeted without being scammy. I would have never heard about Backblaze, Hover (instead of GoDaddy), Warby Parker, SquareSpace, Linode, Eero, or probably SquareTrade if it weren’t for podcast ads. I’ve used or recommended all of those services.

I also find Overcast (podcast player) ads very useful and have discovered a lot of great podcasts because of it. Marco Arment creates his own ad network so he could control both the content and he wouldn’t have any mystery meat binary blob advertising SDK.

[missedthecue]

I think a lot of people proudly claim that advertising doesn't work on them because an advertisement has never compelled them to go out and buy that product immediately. What they don't realize is that that isn't the goal of marketing at all. The goal is brand association. Charlie Munger famously says that 3/4 of all advertising works on pure Pavlovian conditioning.

[timClicks]

What's interesting in that statement to me is that you don't consider the spec sheet to be advertising. Therefore, you feel that you're "unaffected" by it.

[kbos87]

Thank you. The reality is that in 2020 the people buying the ads usually (not always, but more often than not) are expected to account for the return on investment. There’s a little bit of a holier than thou attitude on HN - but if you go deep enough and talk to the people responsible for these budgets, they aren’t stupid and what they are doing is very much scrutinized.

[TeMPOraL]

> if you go deep enough and talk to the people responsible for these budgets, they aren’t stupid and what they are doing is very much scrutinized

In my experience, it's not that they're stupid - but they have neither the required math skills, nor the visibility into the whole pipeline, to be able to correctly evaluate the RoI, so they trust their providers in the ad industry a lot. So the correctness of their calculations depends a lot on the virtue and honesty of people working in the industry specializing in lies and manipulation when placed in front of an easy opportunity to make extra profit by being dishonest.

[ac29]

I think you've intentionally been hyperbolic to make two nuanced positions sound like complete opposites.

Here's my take on the two sides here:

1) It makes me uncomfortable that companies can track me online and use info that to target ads (or whatever their business model thinks it needs my info for). There are things about me like my location, identity and politics that are personal and I want to be in control of, not tokens to be sold.

2) Online advertising returns much less than $1 in profit for every $1 spent, so whats the point? Only big tech companies are benefiting from it. The fact that advertisers are pulling back during this economic contraction only proves this - if $1 in ad spending bought you >$1, they'd keep up their spending.

Not everyone will agree with me, but for me, both of those are true.

[rahimnathwani]

"Online advertising returns much less than $1 in profit for every $1 spent"

Are you saying (A) no single advertiser increases their profits by more than the amount spent on advertising, (B) that, in aggregate, the amount spent on advertising is less than the additional profit earned by all advertisers, or (C) something else?

It's unlikely that (A) is true.

It's possible that (B) is true but that any individual advertiser would be net harmed if they were to stop advertising (because they can't stop their competitors from advertising).

"The fact that advertisers are pulling back during this economic contraction only proves this - if $1 in ad spending bought you >$1, they'd keep up their spending."

It doesn't prove that. Perhaps they are pulling their spending because:

- they're not selling any more due to social distancing (theatre tickets? massages? dating services?)

- they're not selling any more due to supply constraints or inability to operate their business's physical locations

- the stuff they're selling is stuff people cut in a downturn

Just because a piece of advertising isn't worthwhile when no one can go out and many people have just lost their jobs, that doesn't mean it wasn't worthwhile before.

[ac29]

C. I think its impossible to prove that any given $1 of advertising earns back $1 in profit for any sufficiently complex advertiser.

Targeted ads for largely online companies: If your customer sees an advertisement on Facebook and YouTube, and sees a few sponsored search results on Bing, then opens an ad in an Amazon mobile app six months later and converts, which ad was effective? Which was priced right? Lets assume you can 100% correlate all of this activity. Can you justify spending $X on platform Y will return >$X?

Brand awareness ads for largely offline companies: Your products are largely sold at retail, and you are a large multinational company like Coca Cola or Nestle. Which of your ad campaigns this quarter drove sales? Can you justify spending $X on platform Y will return >$X?

[rahimnathwani]

"I think its impossible to prove that any given $1 of advertising earns back $1 in profit for any sufficiently complex advertiser."

Even if that is true, that doesn't mean it's impossible to prove that your total advertising budget of $X improves your profit by more than $X.

[jariel]

Both 1 and 2 are correct.

We live in a consumption-driven society mostly driven by aspiration, not need. We are 'induced' into keeping up with the Joneses to a very great extent. This is not a conspiracy theory.

Much of ad spend is a waste because its effectiveness cannot be measured. Marketers will tell you "I would cut my budget in 1/2 no problem I just don't know which 1/2".

And some companies will spend ads on you from you're a kid until you're an adult hoping to lure you into a few purchases (Ford Trucks, BMW etc.).

[roywiggins]

They absolutely could both be true. Maybe 90% of advertising doesn't work, but the remaining 10% does, but nobody can tell which. It's a lemon market.

[martin_bech]

Both are actually correct. The point of beeing ineffectice is price.. add pricing is so high for most searches/placements that it will cost you 100usd in adds to sell a product with 80usd in margin. Companies will then rationalise this, with an assumption that the “lifetime” value of the customer is > 100usd, or you are paying to gain marketshare etc etc.

[dntbnmpls]

It's even worse when it's the same person pushing contradictory narratives.

The exact same person will write 1 in one thread and 2 in another thread.

1. "Online ads are ineffective and don't work. Online ads are a waste of money".

2. "The russians bought Trump's election by spending $50K on facebook ads".

It's hypocrisy based on agenda.

You might have also noticed that when someone they disagree with gets censored, they claim "google/facebook/etc" are private companies. When someone they agree with gets censored, we cannot allow tech monopolies to drive public discourse.

There was a recent submission where france ruled that google must display and pay for news headlines. The same people saying "nobody has a right to google's platform since they are a private company" were vociferously defending france forcing google to carry news and pay for it. No doubt many of them were hypocritical news industry workers.

But that's human nature, we are all selfish.

[DHPersonal]

I think most people are uncomfortable to admit how effective advertising really is on every person, so as a result, they think up multiple reasons to protest its existence, even if they don’t always make sense as a whole. My opinion on this apparent contradiction is that some who say advertising research is dangerous also believe it is inherently ineffective — that the intrusive advertising methods used now are as helpful as general surface-level research on customers — but it’s the treasure trove of personal information that is dangerous to us all, as it is readily available for abuse.

[ethanwillis]

To answer your last question. I work in adtech, sort of. One of the reasons you'd want to pay to acquire customers right now is firstly dependent on your product.

As an example, lets say your product has a conversion time of 6-12 months. If you strike right now and have the right content to keep people engaged over that time period while theyre in quarantine they'd be ready to "convert" right near the end of some of the most stringent quarantine restrictions.

If you don't spend, then you just burned an entire year's worth of leads in your funnel.

[robocat]

> the two are diametrically opposed

Your (1) is effective advertising that works, and your (2) is ineffective advertising that is a waste of money.

It is perfectly OK to know that both exist, and to argue against both.

[randomsearch]

Na, you’ve misinterpreted what people say.

They aren’t so much concerned by “people buying what they don’t need” - hardy a major concern of most hackers - they’re concerned about invasive data tracking being used to target ads. So I think you’ve gotten the first one wrong. And there’s no contradiction with the second in that case.

[ashtonkem]

A minor tweak makes these two statements work together:

1) The ad industry is insatiable for customer data, and is willing to violate every last persons privacy to get it

2) The utility of the advertising industry’s targeted models is suspect at best.

[searchableguy]

Not sure if this is correct but my hypothesis is that the difference between those threads is that in

A. They are thinking of what they consider an average user (most likely stupid, blissfully unaware of tracking and how companies are controlling him) whilst using ad blockers themselves or relative strong measures to avoid tracking.

B. They are thinking of themselves as the customer to some extent and knowing that they don't ever click on ads as often, it's most likely wasted money.

Again, I am not completely sure but it feels like the person is thinking of different subject in A and B when analysing.

C. It could be that the plan is to make Google look evil in both cases. By robbing both you and companies out of their money and data.

[refurb]

Considering the fact that budgets, including advertising budgets, then to be a percentage of expected revenue, I’d say your probably right

[jonny_eh]

> I've seen the same person argue

Really? You remember who says what on HN?

[nostrademons]

Comment histories are public. It's not hard to look at their /threads page and see what they were saying a day ago.

[K0SM0S]

Please accept this physicalist's take:

- Advertising as a first-principle works, because our biology is amenable to familiarity.

sheer repetition ⇒ familiarity ⇒ positive bias ("things I know / like / trust / ...", common pattern)

There might be more to it, but that's enough, and proved enough times in many a scientific study afaik.

- Applying the technique to a real-world problem has varying levels of efficiency; whether it's the wrong solution or badly executed.

Heed this naive comparison: not all bridges are good, and some are scams. But there's no question whether building bridges is useful in the first place. The question is and forever will be, what should be the rules for making good enough bridges, safe enough?

Advertising as a domain and market is not a special snowflake in most regards. It's actually boringly common, dare I say predictable.

____

Where it gets tricky. Where first-principles aren't enough because complexity is at a whole other level: advertising is the primary revenue for a bunch of industries, most notably the press (the media) which is otherwise considered "the fourth pillar of democracy", i.e. that quality of information in a democracy is as necessary as government, congress, and justice, the 3 branches of the republic ideal form.

How do you reconcile that the biggest "influencers" of public opinion, the press, is itself mostly influenced financially by the most interest, biased, self-tauting side of the entire economy? Wherein not rational engineers, not sane financiers, not level-headed CEOs or even just Jane and John your co-workers next door have a voice, not even sales who know that lying and deceiving is not the way to build a sustainable business... but marketing, in other words those whose job is to create a Hollywood-fiction of fabled greatness... it's not lying, it never was, it's been elevated as fiction —see: artistic awards for the best ads, superbowl hype, and the actual real cinematographic value of some of that, hands down. Nevermind that the products are asking for real money, however.

And then we wonder why the media has become such a theatrical ongoing masterpiece of sensationalist storytelling. Well, lines were crossed.

Why infotainment has become such a norm that it is now capable of higher quality than "editorialized" (read: advertiser-leashed) newsrooms. “No, Jane, we can't say that. We'd lose ad money, you don't want us to fire people, do you?”

The sheer complexity of that makes me want to duck in quantum machine learning and call it a day on politics.

I don't have a perfect solution, I can only see red lines in law and a certain sense of ethics, like we value life, we should value information. In short, bug is in human code, thus fix as well.

[modsWork4appl]

Marketing is all powerful evil. Even if ads don't work.

Some companies are better than others. Instead of "buy my product" they teach consumers they look cool if they buy. We wouldn't be dealing with Apple's BS if people only cared about quality.

[ipsum2]

Google is a hundred-billion dollar company precisely because it allows advertisers to see how effective their cost per sale (or action) is, compared to TV or newspapers, where advertisers are flying blind.

[hardtke]

Measuring the effectiveness of brand advertising on YouTube is more difficult that measuring TV/newspapers. TV and newspapers have higher repetition at an individual consumer level so the "which newspapers do you read?" or "which programs do you watch?" survey questions work. Additionally TV and newspapers have older audiences who are both more tolerant to ads and susceptible to brand advertising. YouTube is not particularly effective for CPA advertising so the technologies from other Google ad products are not that relevant.

[thehappypm]

This is actually untrue. Have you ever seen a survey in a YouTube ad, asking a question like "have you ever heard of Brand Z?" That's one of several mechanisms for measuring brand awareness lift, seeing the lift in recall over a control group.

Likewise, many advertisers aren't doing brand campaigns, they're doing direct response, that is, trying to get people to directly click to their site and convert.

Finally, advertisers on video sites tend to pay per "completed view" for some definition of completed view, usually at least a significant chunk of the video and not just the first few seconds before the skip button appears.

[thebean11]

Why would you need a survey question to track YouTube habits? Google already has that data, why wouldn't they just correlate that to a future purchase you make?

[hardtke]

Statistically reliable purchases studies are very challenging. People have multiple credit cards and sometimes pay cash. If you are buying a car you don't put it on the credit card. Additionally, you have multiple cookies/MAIDS that may or may not be associate with the same person/credit card. Even companies with Google or Facebook resources can't get this to work reliably. If it is a direct response advertiser, of course, you tell Google/Facebook that a purchase is made, but these advertisers are not the bulk of the market and many advertisers don't want to give Google that data.

[ethbro]

Because reconciliation across sessions and platforms is non-trivial. Especially with browsers become increasingly adversarial (aka pro-privacy).

https://www.quora.com/How-can-I-reconcile-Facebook-advertisi...

[edmundsauto]

Possibly true, although the ad industry has survived a number of significant transitions (print/radio/tv/digital/placement), plus 2 primary incarnations (direct & branding). Over the same time, the industry has gotten much more sophistication in measuring return on ad spend. All while growing by leaps and bounds.

I certainly conceded it's possible that your statement is true. I'm skeptical of most marketers claims. However, I have to acknowledge the scope of the deception if indeed it is a house of cards.

That's a lot of smart people wasting a lot of money in a lot of different ways over a century of the biggest growth phase experienced by civilization.

[naravara]

>Over the same time, the industry has gotten much more sophistication in measuring return on ad spend. All while growing by leaps and bounds.

The trick is, the ad industry people are the ones who design all the KPIs that determine the efficacy of ad campaigns. So there's a bit of a self-serving incentive there that could be degrading the quality of the information.

There is clearly some evidence that ad spending does help, particularly with building brand awareness and goosing demand. I think the jury is still out on how much and to what extent specific user tracking/targeting strategies work though.

[freepor]

I had a whole company that I sold to an FAANG that only existed because of Google Ads. It solved a very niche technical problem that you could never go and market on a billboard — the target customer was too niche. But with Google Ads I could find them Googling “How to do X to file type Y” and sell them my software right at that moment.

[codexon]

Ads really work. I have a hard time believing there's any company that put effort into having decent ads and found that they did nothing.

The only question is if they are fairly priced. It could be argued that they were overpriced due to over-funded companies overbidding on them trying to growth hack and bot traffic being mixed in.

[pascalxus]

There was an article written up about ebay and how they paid 20 million$ per year for that first google slot for certain keywords.

Then some consultant came to Ebay and proved, without a doubt that the ads don't work. So, they created a test: they removed the 20 million $ worth of ads for that top slot and watched the traffic afterwards, it was pretty much unchanged! It's because All those people that clicked on those ads, were going to come to ebay anyways, despite the ads, not because of the ads.

[TA129875]

Advertisers aren't idiots, they measure the effectiveness of their ad dollars. If they couldn't show a positive ROI they wouldn't keep spending money on it.

[tonyedgecombe]

I'm fairly confident many advertisers don't have a clue about the effectiveness of their ad spend. This isn't helped by all the dark patterns in the Adwords control panel, every new option they introduce is defaulted in their favour.

[thehappypm]

Does this sound like a real scenario?

CMO: We spent $100,000 on AdWords.

CEO: What was the ROI?

CMO: No clue.

[karatestomp]

Considering how terrible almost everyone seems to be at measuring things generally, or at setting up the conditions for meaningful measurement, or getting buy-in for more effective measurements because they come at some cost and “what Jim’s been doing seems fine” (it absolutely is not) I’m skeptical that advertising folks are somehow much better at this than everyone else, and if they’re not a lot better then they’re still fairly bad.

[root_axis]

These things are pretty easy to measure, you literally just measure the amount of conversions when ad spend is on vs when it is off.

[caseyf7]

I was in the room when a conversation like this happened except it was a million dollars.

[sidlls]

Of course not, but that doesn't mean the "analysis" used to derive the numbers the CMO presents justifies the claims.

[viraptor]

Not $100,000, but if you browse through some forums/subreddits/... about online ads, you'll see a lot of people who want to improve their ads and have no idea if their campaign worked at all in the last months/years. They don't have tracking and don't know how to implement it. There are easily $1000s spent that way by lots of companies which adds up anyway.

And there are also those "I spent $1k in one day by accident with no return, help!" posts.

[ac29]

No, because no one who wants to keep their job would say that.

But take a large advertiser like a Coca-Cola or Procter Gamble - they might run many thousands of ad campaigns a week across all sorts of venues. Their revenue is largely at retail, so there is no way to track directly from an ad campaign to a purchase.

How do you prove the return on any given ad dollar? You cant.

[xiphias2]

Advertising is effective as long as you have lots of VC money to compete with other competitors by overspending. It's far less effective if the company has to be profitable.

The future of ad revenue will depend on when the VC bubble will pop. As long as the interest rates stay 0%, it's not over yet.

[scarface74]

Look at what happened to Yahoo after the dotcom bust and startups weren’t using VC money for advertising.

[root_axis]

Of course they're effective, how could you even imagine that they are not? All available evidence overwhelmingly demonstrates that they are highly effective, it is the fossil fuel that powers literally all popular media.

[friedman23]

> OR ads aren't that effective in the first place

Is this what you want to believe or do you have data to back it up?

[TeMPOraL]

No, you need data to prove that they're effective. Preferably not data helpfully provided by the very company you're buying the ads from.

[oarabbus_]

Ads aren't effective in the first place? How do you justify such a statement?

[Spivak]

I know, it's bonkers that people hold this position when so many startups live and die by their customer acquisition rents.

Do you think startups like Casper and Blue Apron want to hundreds of dollars just to get a single person in the door to buy something and pray they stay for at least 6 months / don't return the thing just so they can hit their break-even point? Like if this was all a house of cards it would have collapsed by now since there are millions of eyes trying desperately to reduce the need for their ad spend.

So yeah, that quirky overproduced Doritos TV ad probably isn't doing all that much but for businesses that don't already have a critical mass of mindshare it's your lifeline.

[thehappypm]

It's definitely important for Doritos. Walk into a convenience store and you'll see a bazillion brands of chips, most of which are healthier or tastier than Doritos. Yet time and time again people go for the old faithful brands. You stop doing brand advertising, that dries up.

[ac29]

Blue Apron seems like a terrible example - their stock has collapsed by over 90% from IPO. And they were actually doing worse before the covid-19 crisis.

[Spivak]

I mean I would expect a company doing food delivery to have an uptick when people are trying to avoid going to the grocery store.

[AmericanChopper]

This is only true for established brands, selling products that don’t change, in saturated markets. Even then it’s only partially true, because the markets themselves are always changing (chances are at least somebody in the world drank a coke for the very first time today). It also doesn’t account for the fact that even if all those factors are truely static (or close enough), that product differentiation strategies can still be used successfully (which is again something you see coke doing a lot). That paradigm really only applies to a very specific type of advertising, under very specific market conditions, and doesn’t apply to the majority of the worlds advertisers.

[DeathArrow]

That doesn't work for me. I don't generally buy products I see in ads or commercials because I think that a good product doesn't need advertising. I try to discover products myself and do my own research on them. Whether is about food, clothing, cars, art, books, computers or services. If I can't research it myself, I'd ask someone with more expertise.

[jagannathtech]

We are a miniscule minority.

[tastygreenapple]

The demand fell, supply went up dramatically and there's a new equilibrium.

Very few advertisers are actually in a true "prisoner's dilemma" in the sense that they're fighting tooth and nail in the same market as another company and parsing back advertising will cost them dramatic marketshare. First, a lot of demand is supply constrained - lysol is going to sell their wipes no matter what, so advertisers can easily defect. Secondly, advertisers may be locked in a competition broadly but not in a specific channel. Your youtube channel probably isn't your best ROI or its easy to cut back spend and just invest in your most performant channel - it's often not youtube for these companies.

[aherhe4haeahe]

You are referring to the game-theory model of advertisement spending; but when game-theoretic decisions are made stochastically over time, it becomes a congestion game[1]

[1] https://en.wikipedia.org/wiki/Congestion_game

[vikramkr]

In layman's terms (the wikipedia is a terrible layman's description) - a congestion game is one where a particular strategy gets less attractive the more people choose to.

Advertising spending, for example in the context of tobacco advertisements, is actually a pretty classic example of what the person you are replying to said though. You'll have to clarify what you mean by making decisions atochastically and how exactly this ends up a congestion game. If everyone else is advertising their cigarette brand, your optimal strategy is to advertise as well, which is not a congestion game. You just end up with a dominant strategy of "advertise" for a differentiated product, which means it is better for you to advertise no matter what. It can also go the other way in advertising though. For example if you sell some undifferentiated product like a commodity, no individual wants to advertise since driving demand for the commodity helps everyone in the market while you have to bear the cost.