SEM Arbitrage – A Dicey Proposition Explained
Published: March 21, 2013
Author: Spencer Fair
If you aren’t familiar with the term “arbitrage” except as a vague, slightly maligned concept, here’s a little background: it’s a term generally used in economics and finance and described a situation where a party purchases a product and immediately sells the product for a profit in another market. The crux is capitalizing on difference in market prices.
In the SEM sphere, arbitrage is when a party bids on low-value keywords and serves ads that direct the customer to another website/search engine that serves ads with higher CPCs. The first party benefits from the difference in the low-cost/high-payout ads and simply shepherds users from one to the other.
If/when the user clicks on the high-payout ads, the buying/selling party gets paid by Google a portion of the money that Google got for the high-paying click.
How does arbitrage work?
Let’s use an example.
– Customers search for the term “california insurance company ratings”
– They see:
– There is an Ask.com ad (top right) with the query in the ad that will redirect to Ask.com search results for that term and will serve the higher-CPC ads:
– When/if the user clicks on the high-position ads (high CPC), Google triggers that as a click and charges the business serving that ad (in this example, let’s say that’s AAA, which is serving the top ad). Google will then pay Ask a portion of that money for getting the user to click on the AAA ad through their website.
If you practice arbitrage
There are plenty of arguments for not practicing arbitrage, and we’ll get to those in a second. But if you’re intrigued by the idea, here are some best practices (note that they relate directly to the best practices for regular search ad copy):
– Customizing the original ad for the query promotes relevance and drives clicks (if the user thinks the ad is just going to redirect them to another search engine, they may not click on it)
– Having the query in the ads (original and second-click) improves the CTR with bolding and relevance
– The more the ad directly relates to the query, the more often the user will continue to click through on the second layer (high-CPC) ads
Why you might choose not to practice arbitrage
That said, there are a whole lot of considerations to take before jumping into arbitrage. These include:
– An industry consensus that arbitrage is bad and hurts the SEM field because of increased costs for businesses/clients on queries that used to cost less.
– The idea that arbitrage could change how users browse Google. Depending on how prevalent arbitrage becomes, users could lose interest in paid ads as a whole and avoid clicking on them to prevent being redirected to a page full of ads or a different search engine.
– It’s a risk for the buying/selling party, which must stay on top of the CTR and CVR (where a conversion = a high-value click) of the low-cost ads to make sure money isn’t being lost.
Google’s man-in-the-middle position
Complicating matters is that Google itself seems to have a conflicted relationship with arbitrage. Some reasons:
– Aside from harming Google’s image and search results quality, arbitrage nets Google a profit from the overall rise in costs and bids for queries, as well as the increased volume of users clicking on high-payout ads.
– Though Google could keep making money with arbitrage, it needs to keep clients/businesses happy with their ad placements and costs.
– If there are complaints about arbitraging ads, Google seems to protect clients/businesses much more than the arbitrager.
Search Arbitrage seems to be a very touchy topic among the SEM community. There are many who vehemently oppose it because it harms their clients and their own bottom line and raises costs as a whole. There very well could be a place for arbitrage to be done correctly and help fill some gaps that Google does not cover.
If enough time and energy were spent on building a well-made arbitrage system, it could be possible to great a strong user experience that made arbitrage less obvious to the average user. This could be done with custom-focused ads on specific proven terms that perform well with arbitrage and a clean well-made search engine/landing page which legitimately helps the user find what they are looking for better than Google.
For now? It’s still an enter-at-your-own-risk game.