you are viewing a single comment's thread.

view the rest of the comments →

[–]magnora7 3 insightful - 1 fun3 insightful - 0 fun4 insightful - 1 fun -  (3 children)

Interesting take. A lot of the valuation of the internet entirely depends on advertising click-through rates. Which they're finding more and more are faked, to make more money. Or just improperly count bot traffic (from search engines cataloging and stuff) as human traffic.

I don't think most advertisers really understand this yet. I wouldn't be surprised if this caused a partial collapse in the internet advertising industry because a lot of the numbers are arbitrarily fluffed up. Also a lot of over-leveraged internet companies are now beginning to collapse, like facebook. It doesn't bode well for reddit's 2020 IPO stock opening.

I think if you combine all that with the student loan crisis, and another real-estate crisis, and a stock market bubble... we could easy see another 2008 crash or worse. I think it's something everyone should be prepared for over the next few years.

[–]danuker 1 insightful - 1 fun1 insightful - 0 fun2 insightful - 1 fun -  (2 children)

I don't think most advertisers really understand this yet I think people serious about advertising will start small and pay attention to their ROI.

"How many $Y did customers give me thanks to $X spent on advertising?"

This, together with advertising on multiple platforms, will let any merchant discover the better platforms.

[–]magnora7 1 insightful - 1 fun1 insightful - 0 fun2 insightful - 1 fun -  (1 child)

"How many $Y did customers give me thanks to $X spent on advertising?"

The problem is, it can be very hard to isolate which people came because of the ad and which came naturally.

[–]danuker 2 insightful - 1 fun2 insightful - 0 fun3 insightful - 1 fun -  (0 children)

It might be hard to implement, but if users are clicking on ads, you can have an URL parameter like "?from=saidit", then track the users via IPs, cookies and/or user agent until they either leave or order something.

Figuring out where the users "leak" away is called "funnel analysis".