4 ms·

"The price is usually given in CPM (cost per mille - cost per thousand impressions), so if an ad has a cpm of 10$, one single impression is worth 0.01$ ! (The a

"The price is usually given in CPM (cost per mille - cost per thousand impressions), so if an ad has a cpm of 10$, one single impression is worth 0.01$ ! (The average cpm, at the time of writing, is around 1$ if you were asking)." This is the fundamental core of why journalism never made the jump from newspapers to the internet, and why we have the absolute cesspool of clickbait and seo-chum of a 'news media' that we do now. Its got nothing to do with ethical standards or cultural shifts or the reporters/writers/editors not wanting to do much much better. Rather, at less than a penny per pageview, the only possible way to pay even a single salary is a plummeting free-fall down to the lowest common denominator and the basest of instincts. That is why facebook was able to make a biz model of it and every other site can't. ($45,000/year salary + 33% biz-overhead) / ($1 CPM x 5 ads per page) = 12 million pageviews per month. You cannot get 12M pageviews per month in any local city, much less the ~100M it would take to have a small staff and an editor. Hence, all the local journalists are gone.

pc86 7y ago
This is exactly why lower tier local newspaper "publications" started splitting already short articles into 3-5 pages, and why 10-item lists require 11 or 12 pageviews to read where pages 4 and 9 are just an ad. For context, to get 12 million pageviews for a local newspaper website, if you're in a small metro of ~250k people with 5% readership, that means every single reader would need to generate nearly 1,000 pageviews a month each. Even at 2 pageviews per article, that's approaching 20 articles a day M-F.
- edmundsauto 7y ago
  What I don't understand is this -- when an article is split into 5 pages, sure there are 5x more ads loaded. But wouldn't they be worth 5x less money per impression, due to overloading the user? Maybe it's a nonlinear scaling -- 5x the ads is 4x the value. Or is it because these are "branding"* ads, and so nobody really measures the ROI on them? * vs direct response
  - pc86 7y ago
    How would "overloading the user" be quantified when buying the ads, though? Sure maybe you could calculate ROI from the ads like you mentioned, or track some other engagement metric, but by then you've already paid for the ads.
    - edmundsauto 7y ago
      It's a recurring purchase, not a one-time thing. So each month/quarter/year, there's a decision whether to continue investing in each channel. (These decisions are often not made well, but that's another issue.)
pizzaparty2 7y ago
Does it even make sense to advertise online? I know that sounds like a dumb question but despite being told that ads DO work on me I don't think I've ever bought a single thing online because of an ad. If I want to advertise my product or service I'm afraid it will just be me dropping a couple thow for ads that no one will ever click on. Maybe then the ad company would blame me for not having a compelling enough ad and I'll have just thrown away my money. I wish advertising were more formulaic so I had a better idea of how much I'll get out given how much I put in. Alternatively, maybe online advertising can work if you think outside the box. But I have serious concerns about buying $2000 worth of customers from Facebook.
- yellow_postit 7y ago
  You don’t have to buy it online, just raising awareness or influencing an offline purchase are also goals for online advertisements. Across large groups it can approach being more formulaic where FB et al. will show you predicted # of conversions based on your inputs before you start a campaign. Of course you are trusting their estimation process then.
- Kaiyou 7y ago
  Those ads aren't meant to make you buy something. They work in a more subtle way by shaping culture. It's a little difficult to explain what the purpose of that is and how it benefits the company. I recommend trying to read up on it, as it's pretty interesting, but will easily take you a few hours.
  - pizzaparty2 7y ago
    So if they are a macro tool does it make sense for me to use them when hocking my widgets?
- squeaky-clean 7y ago
  They don't necessarily have to work on you, they just have to work in general. HN itself is a bit of a bubble, for anyone on here it is likely they want to intentionally avoid internet ads more than the average person. I work in ads for the airline space, I'm on the engineering side so I can't give great financial numbers offhand, but we usually do trial runs on just subsets of routes an airline offers, and for every customer so far the increased revenue on those routes heavily outweighed the new ad spend. Of course you can make bad ads, just like you're saying. We have departments dedicated to that though, I myself couldn't tell you too much about making a good ad that a few SEO blog posts wouldn't cover better. But it can be formulaic to a degree. If you're an airline, give us some of your metrics and we can get a pretty good estimate of your ad performance at different levels. Got a SaaS product, video game, new snack? We're useless, but there's likely some company of analysts out there who have you covered. I'm sure it's extremely variable by market though. Air travel has huuuge volume and lots of customers from every tech-proficiency level. I often get YouTube ads for B2B SaaS services and other extremely targeted business things, and I wonder who is signing a 20mm contract that started from a YouTube ad?? Maybe some other commenter will tell me it works though. Edit: Actually, I signed up for Robert C Martin's (Uncle Bob) new software training video site recently because his youtube ads were so silly. It helps that I knew his name long before though, and I've yet to pay for one of the non-free ones.
0xffff2 7y ago
To put the math further in the red, isn't 33% crazy low overhead? The number I've generally heard is 100% overhead to pay for all benefits and support staff.
soared 7y ago
12M isn't too far out of reach for a medium sized city, but definitely is for a small one. If 75,000 people in your town read the paper, 5 page loads per day, every morning, that is 7.5MM pageviews per month. If a few articles get read by people one town over, you have a few direct ad deals with local businesses, or your cpm is slightly above average, then you can make money. Most local websites don't sell ads on the open market with cpm pricing, but do direct deals with local companies that cost something like $5,000/mo.
- cagenut 7y ago
  sorry but that 75k and that 5 are at least 2x too optimistic. also, direct sales deals require direct sales people who then expect direct deposited bonuses. its kindof like the tyranny of the rocket equation, you can't afford sales people because they would have to take 95% of the revenue they generated just to pay their own (meager regional medium city ad sales) salary.
dialtone 7y ago
That’s an average price and mostly driven down by lack of data on users browsing the site and the mix of campaigns that bought, given retargeting campaigns can easily pay averages in the $5-10 CPM ranges depending on the campaign and users and data density. So the bad persistence of cookies is what largely drives this price to be low. Obviously anyone can choose how to take that.

pc86 7y ago

This is exactly why lower tier local newspaper "publications" started splitting already short articles into 3-5 pages, and why 10-item lists require 11 or 12 pageviews to read where pages 4 and 9 are just an ad. For context, to get 12 million pageviews for a local newspaper website, if you're in a small metro of ~250k people with 5% readership, that means every single reader would need to generate nearly 1,000 pageviews a month each. Even at 2 pageviews per article, that's approaching 20 articles a day M-F.

edmundsauto 7y ago
What I don't understand is this -- when an article is split into 5 pages, sure there are 5x more ads loaded. But wouldn't they be worth 5x less money per impression, due to overloading the user? Maybe it's a nonlinear scaling -- 5x the ads is 4x the value. Or is it because these are "branding"* ads, and so nobody really measures the ROI on them? * vs direct response
- pc86 7y ago
  How would "overloading the user" be quantified when buying the ads, though? Sure maybe you could calculate ROI from the ads like you mentioned, or track some other engagement metric, but by then you've already paid for the ads.
  - edmundsauto 7y ago
    It's a recurring purchase, not a one-time thing. So each month/quarter/year, there's a decision whether to continue investing in each channel. (These decisions are often not made well, but that's another issue.)

pizzaparty2 7y ago

Does it even make sense to advertise online? I know that sounds like a dumb question but despite being told that ads DO work on me I don't think I've ever bought a single thing online because of an ad. If I want to advertise my product or service I'm afraid it will just be me dropping a couple thow for ads that no one will ever click on. Maybe then the ad company would blame me for not having a compelling enough ad and I'll have just thrown away my money. I wish advertising were more formulaic so I had a better idea of how much I'll get out given how much I put in. Alternatively, maybe online advertising can work if you think outside the box. But I have serious concerns about buying $2000 worth of customers from Facebook.

yellow_postit 7y ago
You don’t have to buy it online, just raising awareness or influencing an offline purchase are also goals for online advertisements. Across large groups it can approach being more formulaic where FB et al. will show you predicted # of conversions based on your inputs before you start a campaign. Of course you are trusting their estimation process then.
Kaiyou 7y ago
Those ads aren't meant to make you buy something. They work in a more subtle way by shaping culture. It's a little difficult to explain what the purpose of that is and how it benefits the company. I recommend trying to read up on it, as it's pretty interesting, but will easily take you a few hours.
- pizzaparty2 7y ago
  So if they are a macro tool does it make sense for me to use them when hocking my widgets?
squeaky-clean 7y ago
They don't necessarily have to work on you, they just have to work in general. HN itself is a bit of a bubble, for anyone on here it is likely they want to intentionally avoid internet ads more than the average person. I work in ads for the airline space, I'm on the engineering side so I can't give great financial numbers offhand, but we usually do trial runs on just subsets of routes an airline offers, and for every customer so far the increased revenue on those routes heavily outweighed the new ad spend. Of course you can make bad ads, just like you're saying. We have departments dedicated to that though, I myself couldn't tell you too much about making a good ad that a few SEO blog posts wouldn't cover better. But it can be formulaic to a degree. If you're an airline, give us some of your metrics and we can get a pretty good estimate of your ad performance at different levels. Got a SaaS product, video game, new snack? We're useless, but there's likely some company of analysts out there who have you covered. I'm sure it's extremely variable by market though. Air travel has huuuge volume and lots of customers from every tech-proficiency level. I often get YouTube ads for B2B SaaS services and other extremely targeted business things, and I wonder who is signing a 20mm contract that started from a YouTube ad?? Maybe some other commenter will tell me it works though. Edit: Actually, I signed up for Robert C Martin's (Uncle Bob) new software training video site recently because his youtube ads were so silly. It helps that I knew his name long before though, and I've yet to pay for one of the non-free ones.

0xffff2 7y ago

To put the math further in the red, isn't 33% crazy low overhead? The number I've generally heard is 100% overhead to pay for all benefits and support staff.

soared 7y ago

12M isn't too far out of reach for a medium sized city, but definitely is for a small one. If 75,000 people in your town read the paper, 5 page loads per day, every morning, that is 7.5MM pageviews per month. If a few articles get read by people one town over, you have a few direct ad deals with local businesses, or your cpm is slightly above average, then you can make money. Most local websites don't sell ads on the open market with cpm pricing, but do direct deals with local companies that cost something like $5,000/mo.

cagenut 7y ago
sorry but that 75k and that 5 are at least 2x too optimistic. also, direct sales deals require direct sales people who then expect direct deposited bonuses. its kindof like the tyranny of the rocket equation, you can't afford sales people because they would have to take 95% of the revenue they generated just to pay their own (meager regional medium city ad sales) salary.

dialtone 7y ago

That’s an average price and mostly driven down by lack of data on users browsing the site and the mix of campaigns that bought, given retargeting campaigns can easily pay averages in the $5-10 CPM ranges depending on the campaign and users and data density. So the bad persistence of cookies is what largely drives this price to be low. Obviously anyone can choose how to take that.