The idea behind arbitrage is very alluring, if you can earn more from ads than it costs to buy traffic to your site, you can mint money. Is it actually possible and how does it work?
It is possible, but I would say it’s hard and only works with a certain type of website, the content-farm. I will leave it to the philosophers to decide whether content-farming is an ethical use of time and talent. Let’s just see how it works:
Revenue per Pageview < Cost per Pageview = WIN!
Where can you buy traffic? That’s what Taboola, Outbrain and other content recommendation widgets are for. As a publisher, you likely have these widgets on your site to earn money, but there is another party, often also a publisher, on the other side who is buying those ads.
If you decide to buy traffic to your website for $0.04 / per click. What RPM (revenue per thousand pageviews) do you need to earn to make the average visitor profitable?
$0.04/visit = $40.00 for 1,000 visitors
Now, a $40.00 RPM would be tough to reach. However, remember that this person is likely to visit more than one page of your website. So really, the equation is this:
Acquisition Cost < RPM * Pages-per-visit
This is a crucial factor. You paid $0.04 for this visitor. You need to keep them as long as possible. Suppose the average person visits 3 pages on your website. Things are looking a little brighter, now you only need a $13.33 RPM to be profitable. This factor explains a lot of what content-farms do: AB-testing shocking headlines for maximum clickability, splitting articles into multiple pages, top-20 lists featuring slideshows filled with ads. It’s all because this works to keep users on the page and clicking the next article. If the publisher is able to blend other sources of income such as affiliate revenue, the target RPM becomes achievable.
There are a few other factors to consider. Every time you buy a click, there is a chance that person will share your content, driving free traffic. This factor will vary, but suppose that one in ten people decides to share your content on Facebook, driving another person to the site. This effectively lowers your cost-per-visitor by 10%. While not strictly arbitrage, there is also a chance that each new visitor converts into a long-time user who frequents the site. But it’s unlikely that you are getting content or web hosting for free. These costs should be added to the cost of buying traffic.