Q: From an Affiliate Manager standpoint, whether in-house or outsourced, how is it possible to realistically forecast monthly sales and conversion? I mean, you can do promotions to affiliates up to the rafters, recruit hundreds of content-related sites as affiliates, provide a wide selection of banners, links and creative, establish a streamlined and user-friendly checkout, and yet, at the end of the day, estimating how many sales are going to be driven through the affiliate channel has consistently been crystalballing at best. Sometimes a comparison to year-on-year sales is a reference point, but still, using past performance as a model for the present is not always practical, especially if you are doing promotions, recruiting and providing incentives to current affiliates on a constant basis! We affiliate managers never see the customers, we have no idea when they will really convert, and we are completely responsible for the affiliate channel bottom line! Granted, affiliate management is mainly about sending quality click-through traffic to the merchant, as it is up to the merchant site itself to convert it, I would like your thoughts and insight about how affiliate managers can make accurate monthly sales forecasts.
A: That’s a great question, and one that was covered in depth recently by Beth Kirsch at the Affiliate Manager Boot Camp.
One thing I’d like to point out is that there is no one answer to this question, because mature affiliate programs are going to have a whole different set of metrics than newly launched affiliate programs.
In general, a mature program can be expected to grow at the same rate as the company, while a new program is more erratic for the first 18 months.
Also, the rate of growth will vary from channel to channel. That said, I’ll borrow some from Beth’s presentation to get you on the right path.
The consensus among affiliate managers is that affiliate programs drive between 10% and 35% of sales for merchants.
That said, the percent of sales attributed to a given affiliate program will depend on the level of activity within other marketing channels in your company.
For instance, if your company does not engage in search engine marketing, then you are essentially outsourcing this duty to affiliates, and some numbers that would be attributed to search engines otherwise are being credited to the affiliate program.
Beth explained that for New Program Projections, in order to determine you growth, you need to understand your key variables.
- Affiliates generating traffic
- CTR (click through rate)
- CTB (click to buy)
- Commission paid
The indicator variables change from program to program. However, growth trajectories remain similar for all programs.
The process is a bit more complicated for Mature Program Predictions, according to Beth. In order to find key indicators, you should use regression analysis.
For those of us you that slept through this in college, a regression analysis is a statistical technique that can be used to estimate relationships between variables.
As an example, does the Good Humor Man (are those guys still around?)sell more ice cream on hot days. Of course he does, and he knows this intuitively. But sometimes, the picture is not as clear as that example.
Beth continued that in order to use regression analysis to predict program performance, you will need to determine which variables you want to test in relation to one another.
Try starting with active affiliates, CTR, CTB rate, and commission paid.
You can plot these variable regressions in Excel.
Then, have Excel convert the data into a scatter plot graph and add trend line.
So what did you learn from that?
There is a very strong relationship between orders and commissions (no surprise): R2 = 0.9713
The formula for predicting growth: y = 0.1455x + 7805.4
Good luck with your projections.