Affiliate Sales Tracking Methods Report

Posted on 05. Oct, 2009 by in General

Paid On Results have decided to publish the Affiliate Sales Tracking Methods Report for all to see. We hope that a pie chart will make it visually clear just how big a piece of the pie Affiliates who are not using Paid On Results are missing out on when it comes to Affiliate Sales.


The pie chart represents all of the sales that happened during September 2009. You will note that we use the term Advanced Cookie as we like to make it clear that, although cookie based tracking is the default for every network in the world, some have better cookie set-ups than others. At Paid On Results we believe we have the very best set-up with regards to cookies therefore we have termed ours “advanced” so as to distinguish ourself from networks using inferior cookie tracking set-ups.

In September 2009, Paid On Results recorded 9.18% additional sales that affiliates using some other Networks would have missed out on. Check how much money you made from Affiliate Marketing in September and ask yourself if an extra 9.18% additional income would have been significant for you or not.

If we take 1,000 sales as base line (some of you guys make 10,000+ sales in a month) this is what it would work out at in real sales, I will be rounding up and down on the sales numbers.

  • Advanced Cookie – 908 sales
  • Flash Cookie – 62 sales
  • IP – 14 sales
  • Customer Base – 6 sales
  • Network – 3 sales
  • Other 1 – 6 sales
  • Other 2 – 1 sale

Consequently, in real terms, if the network you are with only uses cookie based tracking then you are missing out in 92 sales in every 1,000. Interesting when you think about it like that.

How Tracking Works at Paid On Results

The majority of sales are checked for tracking in the following order, Advanced Cookie then Flash Cookie, IP, Customer Base, Network, Other 1 and finally, Other 2. “Other” tracking methods are named as such as we do not wish to give away any secrets about our techincal capabilites. So while our Advanced Cookie accounts for the lions share of the sales tracked, if it didn’t exist then it doesn’t mean 90.82% of sales will disappear, but that the other tracking methods would step in and record the majority of these.

Thing You Should Note

All tracking methods can be blocked or cleared, but the most common to suffer from this is the standard cookie. Also, the type of sites people visit also determines how frequently they clear their tracks. If a Network is subject to cookie blocking software, for example, a standard cookie set-up the same across 2 networks could track differently depending on if one network is blocked by a popular software program or not. This is why it’s important for Networks to offer alternative tracking methods so that as many sales as possible are correctly tracked.

Please feel free to quote this research in any articles you maybe writing, but please link back to the source, thanks.

Tags: , ,

6 Responses to “Affiliate Sales Tracking Methods Report”

  1. BFG 9000

    05. Oct, 2009

    Hi Clarke,

    (Merchant Hat On)…..

    Do POR automatically switch to IP if there’s no cookie (‘advanced’ or flash) present?

    Do you have any stats (or guesses) as to what % of sales tracked via IP shouldn’t have tracked at all?



  2. Clarke

    05. Oct, 2009

    @BFG like everything in life you can do things the smart way, or the dumb way and claim they are the same thing. I have read your blog post and I am not passing any judgement on whoever it is your talking about but I can tell you how it works from Paid On Results.

    We only use IP Tracking if the others all turn up nothing, and it’s only for a period of 6 hours from the click. We secondly compare the browser agent, even if a whole office is using the same browser, there may be differences in the user agent such as third party toolbars/extensions, patch level etc. This prevents the majority of the situations you are talking about. Lastly any merchant can block their IPs from the Affiliate Network so if I was to guess based on all I have said above, maybe 1 sale in a blue moon but the other flip side is I could list you 20 ways merchants don’t have to pay for sales that should have tracked but didn’t and as you can see from our IP percentage it’s not a huge amount on it’s own.

  3. BFG 9000

    09. Oct, 2009

    Thanks Clarke,

    (merchant hat still on)

    I’m still not convinced though.
    Let’s take Merchant X as an example – I know they could block their own IP but – they can’t block mine?

    If I was to click on an affiliate link now – Merchant X would end up paying commission on every sale to any of the 10’s of thousands of users mentioned for the next 6 hours.

    This would happen because they all have identical browser agents, & are percieved to be at the same IP address.

    I would imagine that most very large networks would also have identical browser agents for all users too.

    (merchant hat off – affiliate hat on)
    This is great – it minimises leakage & I may even get commission I didn’t earn.



  4. Clarke

    09. Oct, 2009

    @BFG I have changed your post very slightly to read “Merchant X” because it might confuse people into thinking what your saying actually happens with that Merchant you used in your example, so a more general “could be any Merchant”, is better when talking theoretically about Merchants.

    Ok so let’s forget about this specific tracking method for just a moment and I explain what we do if we see multiple orders from the same IP Address or same Cookie (Flash or Regular cookie) within a short space of time (hours) then it will be red flagged to us anyway, as that can be a key indictor of possible fraud, which is then investigated.

    So if your situation did happen then it would be spotted by our automated fraud monitors, but going back to tracking using IP Address, the window is very small anyway, the maximum time we will track using just the IP Address is 6 hours from the click. And IP Address is a last resort to use for tracking and only comes in when other forms of tracking used turns up nothing, hence why it has a low usage % in our stats.

    The real fact of the matter is most people who use the web have cookies on at least on a per session basis before wiping them. The people who have cookies turned off fully will find it hard, to near impossible to order anything from many sites, and it’s these people who after clicking on a link, then enable cookies to order. That is one of the many scenarios this has been designed for, our goal is to accurately and fairly track all sales originating from Affiliate traffic to the benefit of all party’s involved and I feel as a network we do this rather well.

    There is a ton of ways Merchants get sales for free, and really on both sides of the fence I would estimate the percentage we are talking about to be so small that there is no point in worrying about it, otherwise we should all just shut up shop and do something else.

  5. BFG 9000

    09. Oct, 2009

    Hi Clarke,

    (BFG 9000 hat on)

    All very good points & very well made.
    I guess experience (not of POR) has made me slightly pessimistic…..

    I’ll crawl back in to my hole at BFG Towers now!



  6. […] Since then we have published statistics on the total number of sales across our whole Affiliate Network that have been tracked for Affiliates using our cookieless tracking technologies. […]