Comments on: Will Our Conversion Rate Ruin Your Affiliate Program? http://www.onelittleduck.co.uk/affiliate-marketing/will-our-conversion-rate-ruin-your-affiliate-program-1381.htm Online blog of Jason Dale, co-director of Loquax. My views on affiliate marketing, running a website and anything else that seems appropriate! Mon, 05 Sep 2011 14:55:23 +0000 hourly 1 http://wordpress.org/?v=3.1.2 By: hero http://www.onelittleduck.co.uk/affiliate-marketing/will-our-conversion-rate-ruin-your-affiliate-program-1381.htm/comment-page-1/#comment-52431 hero Sun, 31 Jan 2010 09:46:45 +0000 http://www.onelittleduck.co.uk/?p=1381#comment-52431 This is very frequent with merchants that have employed enthousiastic but inexperienced and untrained staff to run their affiliate program. On top of that, they usually are on a self managed solution with their network - or they have an inexperienced network AM too, whose advice leaves a lot to be desired. If this specific AM also has the popular KPI of active affiliates vs total affiliates (active being one that generates sales), then you're pretty much screwed, you'll be kicked out of the program in a short while. One way of improving an affiliate program is to improve its stats. And one way of improving its stats is by kicking out the affiliates whose stats you don't like. The other way is of course to help the affiliate improve their stats. Guess which one's harder to do. There's plenty of retail programs which offer the normal CPA and a hiddden CPL for selected affiliates. There's also retail programs that actually do manage to convert their leads into sales, and for the original affiliate. There is a popular misconception that a program's visible stats (those published by the network) for conversion & EPC will make or break the program. Affiliates who know what they are doing don't pay much attention to these 2 things - they are a)average on the overall performance b)some times adjusted to remove non performing affiliates c)relay no info on how these are achieved (ppc, vouchers, cashback etc). Is any affiliate seriously expecting to get the same figures (if they're high) or decides not to join a merchant because the figures are too low? Nop. The important visible figures are approval rate (or reversal rate), click to sale time, and how fast commissions are released. Anything else should be taken with a large pinch of salt. This is very frequent with merchants that have employed enthousiastic but inexperienced and untrained staff to run their affiliate program. On top of that, they usually are on a self managed solution with their network – or they have an inexperienced network AM too, whose advice leaves a lot to be desired.

If this specific AM also has the popular KPI of active affiliates vs total affiliates (active being one that generates sales), then you’re pretty much screwed, you’ll be kicked out of the program in a short while.

One way of improving an affiliate program is to improve its stats. And one way of improving its stats is by kicking out the affiliates whose stats you don’t like. The other way is of course to help the affiliate improve their stats. Guess which one’s harder to do.

There’s plenty of retail programs which offer the normal CPA and a hiddden CPL for selected affiliates. There’s also retail programs that actually do manage to convert their leads into sales, and for the original affiliate.

There is a popular misconception that a program’s visible stats (those published by the network) for conversion & EPC will make or break the program. Affiliates who know what they are doing don’t pay much attention to these 2 things – they are a)average on the overall performance b)some times adjusted to remove non performing affiliates c)relay no info on how these are achieved (ppc, vouchers, cashback etc). Is any affiliate seriously expecting to get the same figures (if they’re high) or decides not to join a merchant because the figures are too low? Nop.
The important visible figures are approval rate (or reversal rate), click to sale time, and how fast commissions are released. Anything else should be taken with a large pinch of salt.

]]> By: HelenMarie http://www.onelittleduck.co.uk/affiliate-marketing/will-our-conversion-rate-ruin-your-affiliate-program-1381.htm/comment-page-1/#comment-52423 HelenMarie Thu, 28 Jan 2010 12:13:36 +0000 http://www.onelittleduck.co.uk/?p=1381#comment-52423 Interesting point Jason. Being on the dark side (?) I can see why this could be an issue for the Client. Clients that are focused on efficiency and acquisition will scrutinise their conversion rate, and lets face it that is most clients in this sector. Affiliate as a channel will generally always have a much lower conversion rate than other channels due to the variety of ways that affiliate promote clients. We have one affiliate on our network that has a football streaming site, everytime a big game is on he completely kills our conversion rate and throws all of our stats out! A pain, yes, but we wouldn't get rid of them as they are valuable as an acquisition strategy. I do think however it is the responsibility of the Client to give affiliates the tools to improve conversion and better understanding of how to convert customers, I rarely see this done very well. One for us all to work on ... Interesting point Jason. Being on the dark side (?) I can see why this could be an issue for the Client. Clients that are focused on efficiency and acquisition will scrutinise their conversion rate, and lets face it that is most clients in this sector. Affiliate as a channel will generally always have a much lower conversion rate than other channels due to the variety of ways that affiliate promote clients. We have one affiliate on our network that has a football streaming site, everytime a big game is on he completely kills our conversion rate and throws all of our stats out! A pain, yes, but we wouldn’t get rid of them as they are valuable as an acquisition strategy. I do think however it is the responsibility of the Client to give affiliates the tools to improve conversion and better understanding of how to convert customers, I rarely see this done very well. One for us all to work on …

]]> By: Richard http://www.onelittleduck.co.uk/affiliate-marketing/will-our-conversion-rate-ruin-your-affiliate-program-1381.htm/comment-page-1/#comment-52422 Richard Thu, 28 Jan 2010 10:40:05 +0000 http://www.onelittleduck.co.uk/?p=1381#comment-52422 I agree you should appear as a statistical anomaly in publisher's stats but I bet the various affiliate networks do nothing other than take a simple average, probably as sum of conversions/sales value over sum of traffic. If this is so then as an outlier with high traffic but low conversion rate you would not only bring their statistics down, you would have a disproportionate effect because of that high traffic. For example, a merchant with 5 publishers each of whom bring 100 visitors for 10 conversions each would have an average of 50/500 = 10% conversion. Add in one extra merchant with 1 conversion for 500 visitors and they now have 51/1000 = 5% conversion and as far as they are concerned that's your fault! The problem, of course, is not that you're sending them this traffic, its that in the presence of data which does not follow a nice distribution (such as this situation), simple averages are a completely unsuitable, if not misleading metric. We seem to be living in an age of unsuitable metrics (SATS, NHS targets etc) and I have no doubt that our grandchildren will read about these decades as a time when major organisations were happy to compromise their performance in pursuit of ranking highly on some unsuitably constructed league table. I agree you should appear as a statistical anomaly in publisher’s stats but I bet the various affiliate networks do nothing other than take a simple average, probably as sum of conversions/sales value over sum of traffic. If this is so then as an outlier with high traffic but low conversion rate you would not only bring their statistics down, you would have a disproportionate effect because of that high traffic.

For example, a merchant with 5 publishers each of whom bring 100 visitors for 10 conversions each would have an average of 50/500 = 10% conversion. Add in one extra merchant with 1 conversion for 500 visitors and they now have 51/1000 = 5% conversion and as far as they are concerned that’s your fault!

The problem, of course, is not that you’re sending them this traffic, its that in the presence of data which does not follow a nice distribution (such as this situation), simple averages are a completely unsuitable, if not misleading metric.

We seem to be living in an age of unsuitable metrics (SATS, NHS targets etc) and I have no doubt that our grandchildren will read about these decades as a time when major organisations were happy to compromise their performance in pursuit of ranking highly on some unsuitably constructed league table.

]]>