By Evan on Jun 22, 2008 | Reply

Crappy affiliate traffic would be defined as clicks that don’t convert at all or convert at a 1 in 1000 or more rate. This typw traffic especially if the traffic is send in days can all but ruin a good EPC average.

By Dan on Jun 22, 2008 | Reply

Evan - so?

If an affiliate programme culled back to 1 or 2 cashback sites it would have the worlds best conversion rate and EPC but it would be the worlds worst affiliate programme.

If I were to start a PPC campaign and target some major terms its easy to send 1000 clicks in a day, but the click to sale time might be 7 days, or more. Would I be culled for ruining a network stat?

The above quote seems like your relying on a good EPC average to drive incremental sales for your programmes, recruit new affiliates and grow the programmes potential.

By Evan on Jun 22, 2008 | Reply

I manage thousands of affiliates across several programs, so culling back isnt the strategy in play. Sale cycle of course has to be factored in, but most of my programs have a relaively short sales cycle. New affiliate signups should be monitored for initial traffic levels. If a large amount of clicks come in witin a couple days with no conversions then the affiliate has to be contacted and let know they will be expired from the program if they cannot raise their conversions. I suggest ways of improving conversion rates and try to work with them. But depending on the method of traffic generation this sometimes can’t be improved.

When you run large affiliate programs the EPC average can make or break the long-term growth of your program since affiliates look at the average when deciding what programs to work on.

Let me know if you want to discuss futher…

By Matt Bailey on Jun 23, 2008 | Reply

There is another consideration to take into account here. Having experience both network and agency side, I am seeing that there is an increasing demand from merchants upon networks/agencies to protect their brand. This means regular site audits of all affiliates on the campaign.

I think it is a perfectly reasonable demand from a merchant that they know where their brand is promoted, but at present the auditing process is an extremely time consuming manual one. Therefore, sometimes we may be forced to remove affiliates from the campaign if they have consistently driven low volumes of clicks and sales.

This is a simple case of effective time management, but we would always contact these affiliates first and ask them if there is anything we can help them with.

By Stephen Pratley on Jun 23, 2008 | Reply

Absolute rubbish!! The EPC only affects the affiliate, not the merchant as they aren’t the ones paying for those clicks. There is no reason at all to cull affiliates on this basis.

http://www.stephenpratley.com/the-alternative-to-throwing-out-your-affiliates/

I appreciate Matt’s comments about protecting the brand, but there is no correlation between a site that is a ‘good brand environment’ and a high conversion rate. Having worked with a number of ‘high-end’ publishers and luxury brands on affiliate and advertising campaigns I’d say almost the opposite.

If you want to show a high EPC, you have two choices

1)Work closely with your affiliates on best converting offers, treat your site as a constant work-in-progress with constant testing where you can always improve conversion rates, prices and service levels which will benefit your business as a whole. Take criticism in a spirit of constructive feedback and work with affiliates who can prove incremental sales.

2)Just open your campaign to brand bidders and cashback sites and give out a non-expiring site-wide voucher code.

By Bruce Clayton on Jun 23, 2008 | Reply

I can see all sides of this one, but I would like to speak out in defence of the little guy for a moment. Whilst I appreciate fully the clients wish to protect their brand, there is a combined responsibility to ensure that the affiliate has the tools to promote the brand in the correct way, this means ensuring that they have correct creative and administering it in such a way that most affiliates will be displaying up to date banners. Similarly, make available merchant descriptions that an affiliate can use – this leaves less chance for affiliate creativity that may run counter to merchant wishes and less need for time consuming site audits.

In my experience it tends to be as much the “genre” of the affiliate that offends merchant sensibilities as much as the individual affiliate – such as cash-back or PPC. I advocate a broad approach to affiliates, there is a general focus on how all affiliates can promote the brand and generate sales, as well as the inevitable (and desirable) focus on those key sales driving affiliates (who to be honest I have far more merchant questions about than any others) – who are the key determinant of the EPC. I find it odd that at least one network that refuses to publish EPC’s should be one that takes the “periodic cull” approach. The EPC will always be artificially skewed by affiliates, however the actual is key is whether an individual affiliate can achieve a return on their time / SEO and PPC investment. If published EPC was the sole factor in supporting a programme then affiliate recruitment to new programmes would be next to impossible.

If an affiliate manager applies common sense and can see how successful affiliates generate their own EPC then they can apply this to other affiliates of a similar genre - and hopefully grow the programme without having to narrow their affiliate base which reduces the long-term health of a programme as well as putting up perceived barriers to recruitment.

On a practical level I know of a number of affiliates who promote seasonally or via newsletters that will fall foul of the network “inactivity” criteria – common sense dictates that it is surely better to be inclusive in this respect as it is causing negative perceptions of a number of networks to ripple out – which cannot be great for the industry as a whole.

By Azam Marketing on Jun 23, 2008 | Reply

Rather than culling affiliates, it is better to encourage, motivate and assist them. We have several thousand affiliates each on different programs and I would never think of culling the ones that haven’t been able to add their links yet. I know how busy affiliates are, and it can take them months if not years to get around to launching projects.

From the other side, in an affiliate capacity, we’ve belonged to affiliate programs where we’ve done zilch in sales for years and then, after belonging to the program for three or four years without generating a lead or sale, gone on to do significant volumes in sales for them.

As an Affiliate Manager, I actually much prefer working with the smaller guys as they are like sponge, keen to soak up as much as possible. A lot of the smaller publishers are incredibly passionate, and its rewarding seeing them grow.

Nadeem

By Evan on Jun 23, 2008 | Reply

Let me clarify…I have never advocated removing any affiliates for non-activity. EPC is a number that has to be maintained and improved upon when dealing with a CJ US for example. The majority of affiliates will look to this number when choosing programs to promote in a particualar category. Being ranked high in your category is vital to program growth. Of course medium to large affiliates don’t place as much credence on it but I do look to get the masses involved in promoting the offer. Anyone that knows me in the industry knows that my whole approach is one of enabling all affiliates to become producers. I pride myself on being ahead of the curve with affiliate management by providing affiliates: converting keyword files, unique merchant content, unique coupon codes and free web templates to host and use. We will soon be launching free templates with free hosting for our affiliates…which will be cutting edge and widely promoted by us. I always give out my direct cell number to affiliates that I manage in my emails to them and encourage them to call me. This brings in lots of novice phone calls but always some heavy-weights. My approach is getting to personally know as many of my affiliates as possible to form a bind and friendship with them. And hopefully coach/teach less experienced affiliates.

Lets face it there are some affiliates that use banner exchanges, pop-up traffic, contextual means, and other highly non-targeted traffic generation methods. These are usually the culprits who negatively skew down the EPC…and these are the affiliates that I usually look to expire.

By Stephen Pratley on Jun 23, 2008 | Reply

“banner exchanges, pop-up traffic, contextual means”
And your problem is with that?

If you were managing your merchant’s ad budget then fine, but you’re not, the conversion rate is only your concern if, like Chris suggests, you can sit down with the affiliate and try to improve it to increase revenues on both sides. At least these guys are actively trying to achieve reach and consequently incremental sales.

If your argument came from the side Chris is taking, that there is untapped potential in that affiliate, I’d applaud you wholeheartedly, but it doesn’t, it’s a superficial attempt to improve an irrelevant number.

By Evan on Jun 23, 2008 | Reply

Usually if the merchant has an inhouse program I will encourage the affiliate to join direct where conversion rate don’t matter. But where the EPC matters…affiliates that don’t convert at a $5 EPC or higher maybe expired depending on their traffic levels. You can call this approach whatever you want but I have been managing CJ accounts for 7 years and know that this method keeps the program growing healthily.

By Andy on Jun 23, 2008 | Reply

“You can call this approach whatever you want but I have been managing CJ accounts for 7 years ….”

Perhaps thats one of the reasons CJ UK is having such a bad time at the moment!

By Evan on Jun 23, 2008 | Reply

I’ve never managed anything on CJ UK, but would sure like to!

By Daniel Powel on Jun 24, 2008 | Reply

From a network’s perspective, EPC is an important benchmark for a merchant’s campaign. However there are many other ways to assess the profitability of a potential program. As suggested above, looking at the conversion of certain links/products can be an excellent place start.

In the end, there are many caveats rolled into the overall EPC for a given merchant. Different affiliate models will convert differently and the average should not be taken as representative of how your campaign could perform. Instead of removing affiliates, we encourage our merchants to stay in regular contact with their affiliates and provide them with tools, information, and guidance to deliver higher conversions.

Thank you,

Daniel Powel
Commission Junction
dpowel@cj.com
0208 785 5825

By Daniel Powel on Jun 24, 2008 | Reply

Hi Chris,

Thanks for having me aboard!

Can you clarify what you mean by “gaming”?

Thanks,
Dan

By Moose on Jun 24, 2008 | Reply

If it’s to do with culling affiliates based on EPC’s then its the top & tail discussion all over again which was discussed many moons ago. Say an affiliate was a top 10 affiliate but their EPC was not in the desired threshold, would the affiliate then be culled?

By Daniel Powel on Jun 24, 2008 | Reply

Gotcha,

It really depends on the level of vetting because the point can work both ways really. I’ve seen some affiliates absolutely destroy an EPC of an otherwise highly converting program. Sometimes the merchant might have to remove some affiliates if they are driving stacks of clicks, but little or no conversion.

However most affiliates don’t fall into this category. For the rest of affiliates, we would encourage the merchant to work with them to get them converting better.

At some point or another many merchants will look at their affiliate base to streamline its management or improve conversion by removing those affiliates that are not converting. For example, if an affiliate is sending clicks but has never driven as sale, then I can’t see the harm if the efforts I mentioned above are employed. Certainly the affiliate would have better places to send their traffic.

By Stephen Pratley on Jun 24, 2008 | Reply

“In the end, there are many caveats rolled into the overall EPC for a given merchant.”
I think larger affiliates know that and will have the nous to ask how their particular type of site converts.

Maybe instead of publishing EPC’s networks should just write “it depends” in that column

By Moose on Jun 24, 2008 | Reply

If the EPC is low, then perhaps the merchant has to look at their own short comings too. Maybe not having the site to convert traffic, perhaps it might be their own deeplink tracking not working.

How would you also factor in brand awareness via banners & site visitations.

The true litmus test is brand bidding which for a “recognised” brand provides an optimum epc for lesser known brands this would be product specific.

Pratically all programs are launched without knowing potential EPCs, ie brand via paid search as the litmus test. The affiliate ends up as being the guinea pig when programs are launched too soon without a proper incubation or nurturing period.

Infact merchants with lower than desirable vertical sector epc’s should have to pay a hybrid commission with an element of CPC. But networks wouldn’t enforce that instead they allow the cull & blame the affiliate.

Thus maybe merchants & networks should dig up their own backyard before foolishly & needlessly culling affiliates first.

When does a seed germinate into a crop or an acorn become an oak tree?

Again I refer to top & tail procedures rather than percentiles.

By Evan on Jun 24, 2008 | Reply

I always work with our clients to help improve the look and feel of their sites to improve the overall conversion rate. Everything related to expiring affiliates is done on a case by case basis. I would estimate 1-5 per month are expired per program. Of course this varies. It also matters who you let in the program in the first place. Affiliate will sometimes contact me to reapply and I let them back in. I do try to work with as many affiliates as possible. I would venture to say that 98 percent of affiliates don’t have this issue because their traffic is somewhat targeted and converts inline with the merchant’s overall site conversion rate. I think if an affiliates traffic is not converting well and not making them much money off of those efforts then maybe they may want to consider using that inventory for an offer that converts better for that type of traffic. This has been a really interesting discussion. I am always for affiliates prospering with the program and I have never had any affiliate get angry with me for expiring them with a little note explaining the reasons…

By whiterabbit on Jun 24, 2008 | Reply

Culling on EPC - by doing this isn’t there a danger of shutting out relevant sites that are new / just starting out and therefore have low volume of clicks or EPC initially, but may well improve both given time?

If so, seems a bit short-sighted to me.

By Keith on Jun 25, 2008 | Reply

Isn’t some of this though down to vetting affiliates before they join your programme. As an agency we vet every affiliate who joins one of our programmes - ok it takes time and effort but means we hopefully avoid a good number of the link exchangers joining our programmes in the first place - and saves the hassle of removing them later.

I would never cull an affiliate solely to boost the programme EPC, that is just being a slave to statistics imho - but some of the fault here must lie with the networks too - if more did like Tradedoubler do and topped and tailed the EPC to remove the top and bottom 10% then surely the effects of both high clicks and no sales at one end and rebate sites at the other are eliminated and you actually end up with a true EPC rather than a figure which in all reality is worth not even the paper it is written on and does nothing except massage an affiliate manager’s ego