If you frequent any of the AdSense forums chances are you’ve come across the phrase ‘AdSense Arbitrage‘, while it’s been around for a while a lot of people don’t understand what it is, or how it works. While I’m not a big player in the arbitrage model I’ve been doing it profitably for a few months, so I can explain how I do it and share a few tips I’ve picked up. In part II AdSense Arbitrage: Keyword Selection I’ll take you through some real keyword options, and in part III AdSense Arbitrage: Automation and Button Pushing I’ll go over some ideas for automating the process I learned recently.
To start let’s take a high level view of what AdSense arbitrage is and how it works. The basic principle is to pay for a low price for keyword in an advertising program like AdWords, Overture Yahoo Search Marketing, or MSN AdCenter and direct users to a page with AdSense or another contextual advertising program on it that has higher payout. You can either bid low on the same term, looking for keywords with a large bid gap (more on that later) or you can bid on very specific keywords with little competiton and drive them to page with ads for a more general and competitive term.
Sounds pretty simple you bid $1 for a keyword, direct the traffic to a landing page with ads that pay $2 sit back and watch the profits roll in right? Well there’s a little more to it than that, it’s not rocket science but the more familiar you are with how both Adwords and AdSense works the more likely you are to succeed. Up until fairly recently most publishers were getting a 60% cut of the adwords price, $0.03 on a $0.05 bid. However now that AdSense has factored in smart pricing it’s a little more complicated. Here’s a decent explanation of smart pricing from Jensense (One poorly converting site can “smart price” an entire AdSense account - JenSense.com)
Google’s smart pricing feature automatically adjusts the cost of a keyword-targeted content click. So if our data shows that a click from a content page is less likely to turn into actionable business results - such as online sales, registrations, phone calls, or newsletter signups - we reduce the price you pay for that click.
The role smart pricing plays in your account varies from one account to the next and is an unknow variable. To keep the math here simple I’m not going to include it, but it is something you should factor into your computations. To keep things simple lets assume you are bidding $1.00 for a keyword. Again to keep things simple lets assume the advertiser is paying $2.00 a keyword, so you’ll get $1.20 a click ($2.00 * 0.6). So you pay $1.00 a click but get $1.20 a click so you make $0.20 every click, not a lot but you’re not really doing any work after you get it running so you’ll make money slowly. The problem is you aren’t going to get a 100% click through rate on your AdSense ads. Lets run some numbers using 100 users coming to your page
| Users | Cost Per Acuisition | Total Cost | $ Per Click | CTR | Income | Profit |
| 100 | $1.00 | $100 | $1.2 | 100% | $120 | $20 |
| 100 | $1.00 | $100 | $1.2 | 90% | $108 | $8 |
| 100 | $1.00 | $100 | $1.2 | 80% | $96. | -$4 |
| 100 | $1.00 | $100 | $1.2 | 70% | $84 | -$16 |
| 100 | $1.00 | $100 | $1.2 | 60% | $72 | -$28 |
| 100 | $1.00 | $100 | $1.2 | 50% | $60 | -$40 |
| 100 | $1.00 | $100 | $1.2 | 40% | $48 | -$52 |
| 100 | $1.00 | $100 | $1.2 | 30% | $36 | -$64 |
| 100 | $1.00 | $100 | $1.2 | 20% | $24 | -$76 |
| 100 | $1.00 | $100 | $1.2 | 10% | $12 | -$88 |
So what did our little experiment tell us, unless we have a 90% CTR we are going to lose money. While I’m not going to say you can’t achieve a 90% ctr, I am going to say if you did it would be pretty phenomenal. Let’s take a much more realistic CTR of say 30%. If you were able to get the cost down so you were paying $0.25 per click you would be in much better shape
100 leads @ $0.25 = $25
30 clicks at $1.2 = $36
$11 profit
$11 Profit per day isn’t a lot, in fact it’s probably going to cover lunch at the diner and not much else. So you’re going to have to find something with a lot of volume per day or look for a wider bid gap. A bid gap is what occurs when there is a large gap in bid prices for a particular keyword for example:
Bidder 1 - $10.00
Bidder 2 - $9.75
Bidder 3 - $7.00
Bidder 4 - $3.00
Bidder 5 - $1.25
There $4.00 difference between bidder 3 and 4 is a sizeable bid gap, and these are the opportunities you are looking for when you play the arbitrage game. Adwords isn’t much help in revealing bid in fact the best you can get is a kinda close guestimate, using the Overture Bid tool you can get much more accurate bid prices. Let’s assume you are going to bid 1 penny more than bidder number 5 so it will cost you $1.26 for each bid. The top bid is $10.00 so you will get $6.00 per click, so lets run some number again
100 leads @ $1.26 = $126
30 clicks @ $6.00 = $180
$54 dollars profit
Ok $54 dollars is more like it, now you can take your significant other out to dinner. But let’s slow down let’s say you are running a popular sized adsense block, good old 300 x 250. Well that size block displays up to 4 ads. If the user clicks on advertisement #3 forget going out to eat it’s left over mac and cheese for you. If they click on advertisement #4 you’ve lost money, so you better call up mom and see if you can come over for dinner.
The point to here is it may not always be in your best interest to display as many ads as possible, in many cases you can actually make more money by showing less ads, imagine that! You may even want to use single ads like 125 x 125, 180 x 180, 234 x 60 or the double 468 x 60, and 120 x 240. In addition you want to make sure you have the ads in prominant clickable spots (see Maximizing Profits With Website Design and Layout: Part II :). You may also find you have a lot more success using a landing page with no external navigation. By eliminating options the only choices are clicking the advertisements or using the back button. This means your content is going to have to “add value” along the way just be careful how you add value. Lastly if you use multiple ad blocks give each of them their own channel. If one block consistantly gets a higher CTR, have it appear first in the code so it gets the highest priced ad. Position it where ever you want using CSS. Also remember many publishers bid lower prices for contextual ads than search ads, so this could affect your ability to turn a profit, just again another factor to be aware of.
There are some other points to remember as well like cash management. Let’s say you spend $1000 on Adwords, and get $2000 from AdSense, $1000 profit so what’s the problem? You won’t be getting your AdSense check until the end of the month, and chances are your credit card bill will come before your AdSense check arrives, so make sure you can cover the expense. Letting your bill run up and giving the credit card company 18%-22% of your profit just isn’t cool. If you can’t float the money find a 0% interest credit card.
Getting your first campaign up, running and making a profit is definitely the hardest part. You probably are going to have to micromanage it for a week or two. You may even ask yourself is this worth the time? Well remember you’re just on the learning curve, once you’ve learned how to do it for one the second one is easier, the third even easier and by the fourth one hopefully you’ve found your groove. AdSense arbitrage isn’t rocket science but you do have to be on the ball, and it’s not for the faint of heart.
n part I AdSense Arbitrage: Tips, Tricks & Secrets we went over the basics of how AdSense arbitrage works. In this part we’re going to roll up our sleeves and look at some actual keywords and see if we can find some areas where we might like to try some arbitrage.
While high paying keyword lists may be somewhat useful it’s important to understand how those lists are generated and some of their limitations (see Contextual Advertising with AdSense | High-Paying Keywords, Part 2 (series) for a more in depth discussion). I like to start with the Overture bid tool, and if things look OK then go into Google Adwords tool. It’s been my experience that with the AdWords keyword tool there is quite often a substantial difference between what they estimate I will pay and what I actually end up paying for a keyword. Now this could have something to do with the “quality score” variable Google doesn’t share with you, or any number of other factors, but I don’t put too much faith in the Google estimated prices I use them more as a reality check for what I see from Overture. Let’s look at the keyword [mortgage]
/>
Click to Enlarge
Now depending on how close the time you are viewing this page to the time it was written there may be some variation in prices. At the time of writing the top 10 bids were ($4.58, $4.53, $4.10, $4.05, $4.05, $4.05, $4.04, $4.02, $4.01, $4.00) so there’s not a big bid gap there at all. Using Google we see the prices are estimated at $3.67 per click (I think you’d actually end up paying closer to the Overture price when all is said and done)
/>
Click to Enlarge
To get your pricing down to $2.00 or less you are going be bidder #31. According to the Overture keyword tool there are over 1 million searches for the term [mortgage] per month. So you may actually get some click throughs at spot #31 but I don’t think you are going to get too many, so you may make money just very little very slowly. What we really need is a way to get higher priced keywords. Lets look at prices for [new york mortgage]:
/>
Click to Enlarge
OK these prices are much better ($7.95, $7.91, $6.25 etc), however there still isn’t a bid gap. What we really need is a way to get the [new york mortgage] ads to appear and pay the [mortgage] prices and still be contextually relevant enough that the advertisements will get clicked on. Can you think of a way? Remember in part I that I said knowing how the system works can be to your advantage? Here’s one solution, select the word mortgage in your advertising, but only target customers from New York. Write your ad so it’s more appealing for people in New York, then on your landing page write about mortgages for New York. It doesn’t always work but you can use section targeting to help ( see Section Targeting for AdSense allows you to ignore on-page content - JenSense.com for more information).
While I’m sure some of you feel I “tricked” the system I disagree, I just used the way the system worked to my advantage. I also don’t feel the advertiser buying [new york mortgage] got the wrong customer, so nobody got “cheated”. This is just one way to use the system to make a profit, there are many more if you’re willing to spend some time and be a little creative. What you want to do take advantage of inefficiencies that exist, they could be gaps that exist in the market or just using poor searching habits to your advantage. The takeaway here is try to find a way to better pre-qualify the customer for the advertiser, that way you’ve added value somewhere along the way.
Now you can see how this process can be fairly time consuming, in part III AdSense Arbitrage: Automation and Button Pushing we’ll look at ways to search and identify keywords programatically.
