Periodically the media picks up on the story that a lot of the traffic people are paying for online is fake. It's been happening since the early days of CPC. Each time some businesses change their...
Periodically the media picks up on the story that a lot of the traffic people are paying for online is fake. It's been happening since the early days of CPC.
Each time some businesses change their advertising strategy, essentially by starting to pay attention, because if you pay any attention at all the fraud sources are usually obvious. The only way you're paying for large quantities of fake traffic is if you're not measuring ROI.
And advertising platforms start enforcing policies a little better for a while.
Then eventually things go back to normal until the next time. Behind the scenes the arms race between click fraud and detection continues.
If you're going to advertise online, especially using CPC or CPV models, you have to measure sales/acquisitions/ROI specific to traffic sources. It's not optional.
Don't trust the platforms. Don't trust their metrics, don't trust their advice. Just because Google recommends the partner network, or Facebook recommends the audience network, doesn't mean it's a good idea (it almost never is).
Have tracking code in your landing URLs and a metrics system. Blacklist bad sources of traffic, complaining to the network won't get you anywhere unless it's one of the little ones.
If you don't have the time/money/people to actively monitor your campaigns, just don't run campaigns.
It sounds like you know what you're talking about here. I provide business advice to my clients on the regular. but have shied away from advertising or marketing advice because it seems like...
It sounds like you know what you're talking about here. I provide business advice to my clients on the regular. but have shied away from advertising or marketing advice because it seems like there's a lot of fiddly details to making sure advertising is worth the money. Without going into depth (at which point I'd point them to a marketing firm) are there any common red flags beyond depending on a CPC model that I can make sure to warn people off of?
I don't think the parent is saying (and my experience would indicate the same) that CPC is inherently bad, just doing CPC without your own metrics is borderline useless. If you can't make a...
I don't think the parent is saying (and my experience would indicate the same) that CPC is inherently bad, just doing CPC without your own metrics is borderline useless. If you can't make a statement about how much a click from a certain source is worth to you at every part of your funnel, don't pay for them.
I wouldn't call CPC a red flag, it works great if you stay on top of it. I think I pretty much covered the broad strokes important stuff. The best advice for a small/medium outfit is: have someone...
I wouldn't call CPC a red flag, it works great if you stay on top of it.
I think I pretty much covered the broad strokes important stuff. The best advice for a small/medium outfit is: have someone dedicated to monitoring campaigns. Both at the ad network and locally. Dedicated meaning frequently and consistently, rather than full time, small companies don't have enough advertising to need that.
The challenging part is that a lot of outfits don't have anyone to set up in house metrics. That's not a dealbreaker in some cases (i.e. ads.google) because their metrics aren't terrible (for now). But it's vital to set up conversion tracking, which requires a little light teching in house. That might be where you come in :)
This could have some potentially scary ramifications. If businesses start losing confidence in online advertising and the money dries up it could drastically change the current "free but...
This could have some potentially scary ramifications. If businesses start losing confidence in online advertising and the money dries up it could drastically change the current "free but ad-supported" model of the internet.
I'm not quite sure if that would be a good thing or not.
Edit from the future: This would be a great thing.
The money won't dry up, but there will definitely be changes to demand better metrics. If all a company pays for is that a device click through to their site, that's what they'll get. No longer...
The money won't dry up, but there will definitely be changes to demand better metrics. If all a company pays for is that a device click through to their site, that's what they'll get. No longer will they pay for the shittiest of bots to arrive at their page for a moment, they'll get bots that show humanlike activity on the page for a minute, or five, or whatever they ask for.
Who runs the bots, anyway? Well, I guess the bots are part of the ad-fraud industry, but I don't see how a malicious ad network would benefit from their bots viewing ads from competing networks,...
Who runs the bots, anyway?
Well, I guess the bots are part of the ad-fraud industry, but I don't see how a malicious ad network would benefit from their bots viewing ads from competing networks, and it would be a major scandal if Facebook and Google were caught running their own ad-fraud bots.
The bots are mostly run by people gaming the advertising industry for easy cash. With few exceptions the ad networks aren't creating the bad traffic, they're just failing to catch it. Sometimes...
The bots are mostly run by people gaming the advertising industry for easy cash. With few exceptions the ad networks aren't creating the bad traffic, they're just failing to catch it.
Sometimes the sketchy ones know it's there and pretend not to, but in the case of the big guys they don't want that kind of traffic. The quality of the traffic they drive is their product.
Once upon a time Google in particular was really good at catching fraudulent traffic, and proactive about crediting advertisers who had paid for it. These days they're not so good at it but I'm reasonably sure it's not malicious (it would be incredibly stupid to risk their biggest source of revenue for short term cash). It's just the result of corporate bloat and the incompetence that goes with it.
The cynic in my says Facebook/Google/etc probably don't run their own, but they likely turn some semblance of blind eye towards them to keep their numbers up.
The cynic in my says Facebook/Google/etc probably don't run their own, but they likely turn some semblance of blind eye towards them to keep their numbers up.
It would be a tremendously positive force for the web since it would nudge people towards actually seeing the value in services again, paying for them instead. That's why headlines like this give...
I'm not quite sure if that would be a good thing or not.
It would be a tremendously positive force for the web since it would nudge people towards actually seeing the value in services again, paying for them instead. That's why headlines like this give me hope – immediately followed by skepticism, of course. If all this advertising really was pointless, people would have figured out quicker, right? And if I (who certainly isn't running a major advertising campaign) can find a headline like this, every larger company is aware of it, too. Yet there's still ads.
Anyway, I genuinely hope for a second major internet bubble that mostly affects the ad business. It does not seem sustainable (for society) and government intervention is inevitable. If a service like Gmail is still supported by ads in 2050, that would be a kind of depressing future for me.
Periodically the media picks up on the story that a lot of the traffic people are paying for online is fake. It's been happening since the early days of CPC.
Each time some businesses change their advertising strategy, essentially by starting to pay attention, because if you pay any attention at all the fraud sources are usually obvious. The only way you're paying for large quantities of fake traffic is if you're not measuring ROI.
And advertising platforms start enforcing policies a little better for a while.
Then eventually things go back to normal until the next time. Behind the scenes the arms race between click fraud and detection continues.
Which happens because, done right, online advertising works. Free money level works.
Some suggestions for internet business people:
If you're going to advertise online, especially using CPC or CPV models, you have to measure sales/acquisitions/ROI specific to traffic sources. It's not optional.
Don't trust the platforms. Don't trust their metrics, don't trust their advice. Just because Google recommends the partner network, or Facebook recommends the audience network, doesn't mean it's a good idea (it almost never is).
Have tracking code in your landing URLs and a metrics system. Blacklist bad sources of traffic, complaining to the network won't get you anywhere unless it's one of the little ones.
If you don't have the time/money/people to actively monitor your campaigns, just don't run campaigns.
It sounds like you know what you're talking about here. I provide business advice to my clients on the regular. but have shied away from advertising or marketing advice because it seems like there's a lot of fiddly details to making sure advertising is worth the money. Without going into depth (at which point I'd point them to a marketing firm) are there any common red flags beyond depending on a CPC model that I can make sure to warn people off of?
I don't think the parent is saying (and my experience would indicate the same) that CPC is inherently bad, just doing CPC without your own metrics is borderline useless. If you can't make a statement about how much a click from a certain source is worth to you at every part of your funnel, don't pay for them.
I wouldn't call CPC a red flag, it works great if you stay on top of it.
I think I pretty much covered the broad strokes important stuff. The best advice for a small/medium outfit is: have someone dedicated to monitoring campaigns. Both at the ad network and locally. Dedicated meaning frequently and consistently, rather than full time, small companies don't have enough advertising to need that.
The challenging part is that a lot of outfits don't have anyone to set up in house metrics. That's not a dealbreaker in some cases (i.e. ads.google) because their metrics aren't terrible (for now). But it's vital to set up conversion tracking, which requires a little light teching in house. That might be where you come in :)
Awesome, thank you.
This could have some potentially scary ramifications. If businesses start losing confidence in online advertising and the money dries up it could drastically change the current "free but ad-supported" model of the internet.
I'm not quite sure if that would be a good thing or not.
Edit from the future: This would be a great thing.
This isn't scary, this would be fantastic.
The money won't dry up, but there will definitely be changes to demand better metrics. If all a company pays for is that a device click through to their site, that's what they'll get. No longer will they pay for the shittiest of bots to arrive at their page for a moment, they'll get bots that show humanlike activity on the page for a minute, or five, or whatever they ask for.
Who runs the bots, anyway?
Well, I guess the bots are part of the ad-fraud industry, but I don't see how a malicious ad network would benefit from their bots viewing ads from competing networks, and it would be a major scandal if Facebook and Google were caught running their own ad-fraud bots.
The bots are mostly run by people gaming the advertising industry for easy cash. With few exceptions the ad networks aren't creating the bad traffic, they're just failing to catch it.
Sometimes the sketchy ones know it's there and pretend not to, but in the case of the big guys they don't want that kind of traffic. The quality of the traffic they drive is their product.
Once upon a time Google in particular was really good at catching fraudulent traffic, and proactive about crediting advertisers who had paid for it. These days they're not so good at it but I'm reasonably sure it's not malicious (it would be incredibly stupid to risk their biggest source of revenue for short term cash). It's just the result of corporate bloat and the incompetence that goes with it.
The cynic in my says Facebook/Google/etc probably don't run their own, but they likely turn some semblance of blind eye towards them to keep their numbers up.
It would be a tremendously positive force for the web since it would nudge people towards actually seeing the value in services again, paying for them instead. That's why headlines like this give me hope – immediately followed by skepticism, of course. If all this advertising really was pointless, people would have figured out quicker, right? And if I (who certainly isn't running a major advertising campaign) can find a headline like this, every larger company is aware of it, too. Yet there's still ads.
Anyway, I genuinely hope for a second major internet bubble that mostly affects the ad business. It does not seem sustainable (for society) and government intervention is inevitable. If a service like Gmail is still supported by ads in 2050, that would be a kind of depressing future for me.