Hacker News Reader

This. I am surprised to see hacker news jump on this as proof that advertising causes reduced happiness when this article can only conclude that they are correlated. The common statistical maxim is that correlation does not imply causation. A far more likely scenario, in my opinion, is that greater development is associated with both increased advertising and a reduction in happiness. And yes, the summary does state that they were able to show that it is uncorrelated with GDP but what's telling is that the prominent bar graph is "uncorrected". So, after correcting, the advertising effect may be so minor that they chose not to graph it at all.

I do my best to limit advertising in my own life because I myself believe that I am a happier person without it. This study, however, is misleading and doesn't give us evidence of much of anything.

> So basically it has a negligible impact on happiness even if this analysis is true.

You are wrong. The paper goes on to say, in the very next sentence, that 3% is "approximately one half the absolute size of the marriage effect on life satisfaction, or approximately one quarter of the absolute size of the effect of being unemployed", thus (while smaller), it's the same order of magnitude.

Presumably you do not consider the impact of marriage or unemployment "basically negligible"?

> It is really, really hard to actually change people's opinions with advertising.

That may be as it may, but that is not what the paper is talking about.

I saw that. I still think it's small. 3% is 3%, even if their measure shows marriage being 6%.

Changing peoples minds about their self worth is still changing their mind.

Happiness is the one thing we should be optimizing for. 3% is not negligible if there is no trade-off.

Again, I don't think the correlation holds in the first place. But if it did, there absolutely is a trade-off. It's more or less impossible to grow any endeavor beyond local scale without ads, unless you're lucky enough to be an industry with a strong viral or press component. Take a look at the S-1s of most companies that go public. 9/10 of the companies have been able to grow because of ads.

I don't think you understand what I mean by "there is no trade-off". Those endeavours are only useful if and to the extent they create happiness.

Also: Companies need ads to grow because their competitors also have ads. That doesn't mean they'd need ads in if ads didn't exist.

The one legitimate purpose ads serve – could easily be served through other mechanisms, e.g. by the press. Maybe the press would have to expand but advertising isn't free either. In any case, an independent press is much more suitable to actually inform people than the ad industry which at best informs people by accident while trying to convince people to buy $PRODUCT.

Ads are like military: It's mostly necessary because it exists.

Okay. So say you make an app that helps plumbers do their job well. If every plumber used it, plumber productivity would go up 50%, making life easier for all. It cost you years of your time and lots of money to make this great app, and you need a lot of plumbers to buy it to afford to continue working on it.

There's a big plumber app on the market already, but yours is way better.

How do you get plumbers to hear about your product?

1) The press? Uninterested. Plumber magazines or sites? Very low traffic - no way to scale.

2) Call or knock on doors of every plumber near you? If everyone did this, the spam would be unreal. Also inefficient and not scalable.

3) Referrals? You get some, but most plumbers aren't chomping at the bit to help their competitors. And because you started with so few, referrals are a snails pace and by no means exponential.

4) Organic postings and SEO? The big plumber app is way ahead of you. And 100 other worse apps clog up the listings. No one sees your posts.

See the problem? Without ads, you can't grow many many categories of genuinely valuable products. Even if no one else is advertising, you can't really grow without them.

This is barely a hypothetical by the way. I've worked with a dozen companies with similar situations.

This is fallacious thinking. If you had an app that would increase plumber productivity by 50%, plumbers would be lining up to get into the Beta. Find any plumbing supplier or trade group, and if the effects are actually that good, they'll tell their members at no-charge because that knowledge would be a service to them. Plumbers would be cold-calling other plumbers trying to get a beta invite. All you need is a marginally competent sales person.

Advertising is a waste money for quality products, only makes sense for inferior ones.

> If you had an app that would increase plumber productivity by 50%, plumbers would be lining up to get into the Beta.

How would plumbers know about the app without advertising?

> Find any plumbing supplier [...] they'll tell their members at no-charge

In other words: They would advertise it.

> In other words: They would advertise it.

It's quite clear we are talking about paid advertising, not the broader concept of people saying things.

This is just not true. Example: I was just hired to scale marketing for a particular app.

This app currently has over 100 real reviews online, averaging 4.9 stars out of 5. Retention is nuts - 98% of people who ever started using the product (which has a monthly fee) are still using it. The company is three years old.

Perfect to take off, right? But few industries are actually viral. We just doubled the number of new customers per month using FB ads in the first two months of scaling it out, and we'll scale much further over the next 3.

Anyone who works in start-up growth will tell you the same things I am.

Why isn't the app growing organically?

You can certainly pay for eyeballs, the question is whether FB ads are giving users with a retention level high enough to justify the advertising expense. Users from online ads tend to much lower quality than those that come from organic traffic, I would not expect your organic retention numbers to hold.

I've said why its not growing organically many times. Most businesses have no viral coefficient, so organic growth is linear and slow. Some consumer focused tech busnesses are the exception, not the rules.

And yes, all ad spend should be measured against true LTV.

> Uninterested. Plumber magazines or sites? Very low traffic - no way to scale.

That means plumbers don't care about plumbing news, including your app. You have no right to disrespect their choice by spamming them anyway.

In a world without ads, not subscribing to plumbing news would be a weird choice anyway. Refusing to learn about new ways to increase productivity – like your app – means giving your competition an advantage.

> It cost you years of your time and lots of money to make this great app, and you need a lot of plumbers to buy it to afford to continue working on it.

Society does not owe you customers.

Not subscribing to plumbing magaxines just means the signal to noise ratio in them is bad.

For the value part: What if plumbers use Facebook (which they do) and when they see your FB ad, they buy your product. Wouldnt that indicate they do care about your app?

If plumbers aren't satisfied with the signal to noise ratio of magazines they don't want to see ads either.

They still use FB even though they see ads there. And then convert when they see the Plumbing ad.

So? Maybe some ends justify the means. Selling an app isn't such an end.

Is being able to sell your stuff all you care about? Because that's the impression I'm getting. I'd summarize our discussion like this (exaggerating for effect):

> Ads damage democracy, make everyone unhappy, they're a huge drain on our economy and can be replaced by better methods of distributing information.

> But I get to sell a plumbing app.

I would completely disagree with your summary of our discussion.

If there were no explicit advertising, companies would use other, hidden means to reach customers' minds. Today in the press we have a more or less clear distinction between advertising and content, and I think if we banned advertising it would create an unintended incentive that could lead to the destruction of unbiased journalism as we know it today.

> companies would use other, hidden means to reach customers' minds.

This is currently done extensively, despite advertising existing. So this doesn't seem like a valid counterpoint.

I explain why this is a valid counterpoint in the second sentence of my post: Today in the press we have a more or less clear distinction between advertising and content, and I think if we banned advertising it would create an unintended incentive that could lead to the destruction of unbiased journalism as we know it today.

I think the distinction between advertising and content is a lot more blurry than you realize. The majority of 'news' articles begin from some company's press release, product placement is common in fictional worlds, and journalists frequently self-censor to avoid displeasing advertisers.

> Today in the press we have a more or less clear distinction between advertising and content,

You have heard of native advertising haven't you? Of course ethical journalists refuse to do this. They will still refuse when ordinary ads have been banned.

On the other hand, ads make the press less independent. No matter how much you try to avoid it: If someone buys ads in your publication they have influence over you. Be too critical of them and they'll withdraw their funding. They won't openly threaten you, of course, because that is not legal and not necessary.

Only companies with inferior products need to advertise: great products are spread quickly at zero cost in the age of social media.

Further, a proportion of the sale price funded the advertisement. Any product that has been advertised is necessarily overpriced.

Sentence 1 is untrue. See https://news.ycombinator.com/item?id=20034730 for an example.

Sentence 3 is also untrue. No advertising means people hear about much fewer brands, which means reduced competition, which means higher prices. This more or less balances out the cost of ads for the consumer.

> Happiness is the one thing we should be optimizing for

here s some more spurious correlations. Countries with very high happiness have some of the highest suicide rates. happiness might be dangerous to the survival of the species. We should optimize against it

https://medium.com/@dtg319/happier-countries-are-more-suicid...

I don't think you can debunk a scientific paper written by multiple professors with a blog post on Medium.

Why yes, multiple other professors have found it.

https://well.blogs.nytimes.com/2011/04/22/happiest-places-po...

I disagree - especially since optimizing for anything blindly/without constraints can lead to horrific outcomes.

Optimizing for happiness would logically call for denying unpleasant truths because pointing them out would make people unhappy and downright inhumanities - if the depressed commit suicide it would drive up the average happiness. Even if given tighter constraints it would still have disturbing ignorance is bliss promotion.

> Optimizing for happiness would logically call for denying unpleasant truths because pointing them out would make people unhappy

Not if you take the long term into account. Denying unpleasant truth might make people happy -- though I don't think this works if your subconscious still knows the truth -- but then you cannot deal with the issue and it makes you unhappy.

> if the depressed commit suicide it would drive up the average happiness.

If someone's life is so unpleasant it isn't worth living and they cannot be treated, suicide isn't inhumane.

> Even if given tighter constraints it would still have disturbing ignorance is bliss promotion.

Something being disturbing doesn't mean it's actually bad. Reactionaries find homosexuality disturbing.

Do you have a reason for considering ignorance is bliss promotion disturbing aside from lack of familiarity?

> I disagree - especially since optimizing for anything blindly/without constraints can lead to horrific outcomes.

I don't think an outcome that makes people happy, including in the long run, can be horrific.

> So basically it has a negligible impact on happiness even if this analysis is true.

Not to those 3%, and since advertising provides very little benefit to anyone who isn't in the ad business, I would be happy to that 3%.

I don't think that correlation is accurate. But even if it is, there is a trade-off: https://news.ycombinator.com/item?id=20034507

I understand. I'm not saying that I accept that the correlation is accurate. I'm just saying that if it is, the tradeoff to stop making 3% of people less happy is totally worthwhile.

But of course, still in the hypothetical, it wouldn't be necessary for ads to disappear to accomplish this. All that would be necessary is for advertising to adhere to the ideals the ad industry itself says it holds: rather than manipulating people, ads could just inform people what the product is, what it does, and what its strengths (and, as long as I'm dreaming, weaknesses) are.

But the vast majority of ads do none of that. Instead, they engage in psychological manipulation.

I'd take issue with that. Go to https://www.facebook.com/ads/library/?active_status=all&ad_t... and look up some random brands you've heard of. The vast majority of ads are on the level. The ones that aren't are when the product itself is a scam most of the time.

I don't do Facebook, and have no clue what FB ads look like. I'm talking about advertising generally, online and off. I very rarely see ads that are like that.

Either advertising is so powerful as to rule the world, or it's so useless that nobody benefits from it. I 'd say , without adequate research on the effectiveness of advertising, these correlational studies are only marginally interesting.

The spending on advertizing likely has diminishing returns. The marginal decrease in satisfaction that would come from doubling the spending is possibly smaller than the baseline decrease caused by the initial expenditure.

There's nothing in their data to support this, and if the effect was there they would have seen it (since they have yearly happiness data). They describe a linear correlation.

They describe a linear correlation because they fitted linear correlation. All kinds of interesting effects could be hidden in the residuals, but they don't show them.

GDP measures activity involving money, but makes no distinction between useful, useless and harmful activity. the cost of cleaning up pollution is added to the GDP.

So, halving it would make people 3% more happy. I say halve it 30 times in a row then. These silly stats make no sense at all.

That doesn't follow.

https://en.wikipedia.org/wiki/Reductio_ad_absurdum

Not really, though -- it was empirical, and the fact that an estimate of the first halving does X doesn't imply that further halving will have the same effect, so a reductio doesn't work here.

Percentages don't work that way. Don't blame the stats for your own misunderstanding.

Yes, that was sort of the point.

Sorry.