Making Sense of Your Affiliate Numbers: Part 1

It’s the end of another month. You’re not quite sure what to do with all the stats and charts provided by your affiliate network provider’s web site. After all, there are plenty of numbers to track: total affiliates, new affiliates, sales, clicks, impressions, CTR, CPM… The list could go on especially with all the variants that can be calculated and deduced from the raw data.

Where to start?

Different merchants have different goals for their affiliate programs. Some merchants are content to generate impressions, others traffic to their web site, still others are looking for leads or sales. Let’s assume you’re interested in driving leads or sales to your web site. To simplify, we’ll refer to them generically as transactions. To further simplify, let’s put off dealing with the issue of revenue per sale until a future column. In other words, assume all transactions are created equal.

This being the case, the first thing to look at is the total number of transactions. Obviously, you want this number to be growing month over month. Independent of volume increases or decreases, you should also consider your transaction rate, or conversion rate. This can help pinpoint where changes to your web site may lead to better throughput. In fact, learning from your affiliate program can drive better performance from all traffic to your site. Also on the throughput side, consider the effectiveness of your various linking options.

Using the hypothetical table below, here are a few observations. The click rate seems to be falling with time. To make sense of this, you should dig into where these impressions are being served. It could be that your program is being picked up by webmasters with fewer visitors that match your offerings. To see if this is the case, check whether a disproportionate share of the growing impression base is being served by new affiliates. If you discover that the majority of your impressions are being served by the same group of affiliates, the falling click-through rates are more likely a function of the need for new creative.

Moving to Convert Rates

You’ll notice that conversion rates make a nice rise between months one and three. Again, several factors could be at work. First, consider any changes that have been made to your web site, especially in the transaction pages and where visitors from affiliate sites land. If nothing has been changed here, again take a look at which affiliates are sending the traffic. If a new group of affiliates is sending the traffic, these new affiliates may be attracting visitors that, while less likely to click, are more likely to buy when they do click. Alternatively, if the traffic continues to come from the same set of webmaster sites, it could be that affiliates respond and convert only after being exposed to the message several times. If you start to notice a lag between the time a group of affiliates join your program and the time when their visitors start converting, this could confirm your assumption.

Month 1
Month 2
Month 3
Growth Rate
Convert Rate
Click Rate

Try limiting yourself to only this limited set of numbers. Look for patterns and oddities. By probing these areas, you’ll gain a better understanding of what’s happening with your program and you’ll go a long way toward making more intelligent decisions moving forward. Of course, there are many other numbers to consider.

Next week, we’ll discuss the ways to measure the health and well-being of your affiliates themselves. This includes a look at some data like total affiliates and new affiliates, along with some of my own metrics such as current affiliates, active affiliates and super affiliates.

Conference Update

The Online Affiliates conference will be held July 17-19 at the W Hotel in San Francisco. The three-day event has a full line-up. I’ll be moderating an afternoon panel focused on the trend toward context-centric product links featuring executives from SuperSig, DynamicTrade and Affinia.

Related reading