a new approach to affiliate management

[kcheung]

Tweet

I've been mulling this over over the last few weeks, and i think it will be a good thing (theoretically)

it might not be totally revolutionary but i want to see what affiliates think.

problem -
tiered commission structures are too generic and only favor top affiliates, who then find it too easy to sit on the top tier.
tiered commission structures don't really factor in TYPE of affiliate or even the potential of each individual affiliate

what you end up with at most is a tier commission structure that -
splits affiliates in to incentive vs non-incentive
and sets tiers too far apart based on the top performing affiliate and the number of sales generated.

solution -
1. split affiliates based on promotion method Grouping 1
2. from that, split the affiliates into top, middle low - grouping 2
3. select the top 5 affiliates from each of the groups in grouping 2 - offer these bespoke deals e.g. bespoke tiered commissions, bonus commission for reaching targets etc.

the result - not only do you now have a tiered commission structure that addressed the needs of different types of promotion, but you are also setting tiers based on what is achievable within each type of promotional method.

you would be able to set different targets based on grouping 1
then set different tiers based on grouping 2
then with the top 5 from each in grouping 2 you would be able to arrange bespoke deals to improve performance

In theory this would mean that a massive voucher code site would be on different targets and tiered commission structure to a niche affiliate with a single site promoting 1 product. this would mean that it would be easier for a small affiliate to achieve the next tier up.

thoughts?

[mattb811]

How is this new?

[kcheung]

because its not a generic one size fits all tiered commission structure based on volume of sales generated.

it also factors in the type of affiliate and sets tiers according to that method.

[mattb811]

I think that this is generally done already on a lot of larger accounts Ken. Obviously it is not publicised but I'd say that it is standard practice, for us at i-level at least, to have differentiated payment levels for different affiliate types and bespoke rates for the higher performing affiliates.

[kcheung]

i realize its not revolutionary.

Matt, any idea of how affiliates respond to it?

[hpops]

Yeah, have to say Ken, we're doing this for a few folk.
I'd actually go a tad further and say that targets should be set for all affiliates on the program with incentives given for performance over and above their expected performance - tiers without targets is money for old rope.
A lot of the programs you see with low baseline commissions will have all sorts of "on-tops" related to sector/size/type etc.

[kcheung]

how have affiliates reacted to it?

I mean its pretty bog standard that we set a tiered commission based on targets that reaches out to all affiliates based on volume or value of sales. I've not seen a tier commission structure (as an affiliate and I'm signed up to all the big networks and most campaigns) publicly that splits affiliates by type, sets different commission tiers based on that and sets different targets based on that. so I'm very interested to see how affiliates would respond to that as a standard offering rather than one for select affiliates if that makes sense.

[hpops]

Pretty well, they aren't public tiers though, and I think you'll find that that is true for the majority of programs doing this. If the tiers are made public it can get all "political" for absolutely no good reason.
"My traffic's as good as anyone else's etc"

[mattb811]

The problem is that it is still very tricky to split affiliates out by sector on most campaigns/networks. We tend to focus on a few specific sectors, ie treating cashback and voucher very separately, and then segment the rest of the affiliates but the majority of content affiliates will go into the "other" pot who get offered standard commission rates.

Personally, I'm not convinced that the effort required to design bespoke target levels for a large number of small affiliates would justify the return, particularly on a big campaign like Orange, Lovefilm or Next. Would anyone disagree?

[kcheung]

i can see how its practical for VC and cashback sites, and thats quite common, i guess its that same old question...how do we structure the campaign so it also incentiveizes the smaller affiliates rather than gear towards only the bigger affiliates.

otherwise what you end up with is bespoke deals and tiers for 10 top affiliates and a generic tier structure for everyone else. - its everyone else that in a way needs that incentivisation. even something simple like tiering the commission structure as follows would make a difference to how smaller affiliates are motivated.

number of sales vs. commission tier - these tiers would be different based on the type of affiliate (e.g. PPC, True content, niche etc)
0-5 5%
5 - 10 6%
10 - 30 7%
30 - 100 8%
100+ 8.5%
then target based bonuses for anyone who is the top of each of those tiers.
eg.
affiliate making 5 sales a month, to get to 6% tier and get a bonus of £100 they need to make X more additional sales - this would be relative to the type of affilite they are eg. true content v.s PPC affilite.

hpops comment about it being political seems important to consider though...definitely food for thought.

I guess until all the attribution stuff comes up to speed it'll be difficult to position this.

as for your comments (matt) on effort required to design bespoke target levels - i guess there is something to that but if you design something relatively automated, it can be put in place by the networks and the whole process would run itself so to speak, i take your point that maybe the revenue might not be there to begin with, but that's kind of the idea, by making tiers for smaller affiliates more achievable if each small affiliate can make just 1 more sale each a month, that would already demonstrate growth in the campaign

[hpops]

Nope = wouldn't argue at all on a program of that size - segmentation is stupidly difficult anyway.
Of course - I might argue that those in the "other" pot are those who should be rewarded the highest levels....

[Hero]

iif you design something relatively automated, it can be put in place by the networks and the whole process would run itself so to speak

most networks have HUGE development queues so I would advise you to speak with your networks first before knocking on your merchants' doors with the idea, as changing how commissions are attributed is no small task.