Spotlight on: How to read the ROI column

Wednesday, September 13, 2006 | 4:59 PM

One of the nice things about Google Analytics is how easy it is to see ROI for each of the keywords you buy on AdWords. But what do those ROI numbers in your reports really mean?

ROI is (Revenue - Cost)/Cost, expressed as a percentage.
-Revenue is taken from either the value you set as your goal value(s), or from e-commerce revenue values if you have set up e-commerce tracking.
-Cost is currently only derived from your AdWords CPC (cost-per-click) values imported from your AdWords account when you link your Analytics and AdWords accounts with auto-tagging turned on.

An ROI of 0% means that you earned in revenue the same amount of money you spent. An ROI of 100% means that you spent, say, $5, and made $10. In other words, you spent X and received 2X in revenue. A minus sign (-100%) indicates that you lost all of the money you spent. (If your numbers are all -100%, it's probably because you haven't configured e-commerce or defined values for your goals and therefore Google Analytics isn't registering any revenue.)

It's not uncommon to get an ROI percentage of several hundred or even several thousand. These kinds of ROIs simply indicate that your Revenue is many times greater than your Cost. Depending upon your business, you might need an ROI of 1000% just to break even. Let's say that you sell golf clubs online for $500 per set. You spend $100 on the keyword [beginner golf clubs], which results in 5 sales for a total of $2,500.

In this case, the AdWords Analysis Report will show that you've made a 2,400% ROI. But you need to factor in your operating expenses and your production costs to understand how much money you've actually made. For example, if your cost of purchasing or manufacturing the clubs is $350, you've really only made $150 per set x 5 = $750. That gives you an ROI of ($750 - $100)/$100 = 650%.

It's best to use ROI as a guide to your keyword spending instead of as the final answer on how much you've made. Those of you who want to learn more about keyword buying metrics might be interested in this post: what's a visit worth?.