Skip to content

Get the basics right in digital marketing first

De-Duplication Best Practice Guidelines


It still beggars belief that today in 2014 with all the analytics enterprises talking about digital media attribution and cross device platform tracking that so many of the BIG and little advertisers are still not getting the basics right.


This basic methodology that I am going to talk about today is simply de duplicating on line sales.


What is de-duplication?

De-duplication is the process of attributing a sale to one online marketing channel on predefined conditions, these conditions will decide which tracking tag is shown at the confirmation of sale page.


Dgm advocate all our advertisers to de duplicate sales the status quo  for this usually attributing sales on a last click wins basis.

Advertisers who do not act on this fundamental necessity  stand to lose up to 35-45% of their media budget in misallocation and data duplication.

We also advocate Advertisers perform de duplication at the point of sale in order to remove and further confusion.



Where an advertiser runs their programme on two networks, de-duplication is essential to prevent paying for the same sale twice.

This can occur when a customer has visited publisher sites from both networks, and ends up with two publisher network cookies on their computer.

Using automatic de-duplication, the advertiser only displays the tracking for the network that last referred the customer. Therefore, only one network receives notification of the sale, thus removing the need to decline duplicate sales manually at the validations stage.


What do I need to do to de-duplicate sales?

Industry standards recommends using a “Click Append” and setting a local cookie on the visitor’s machine when they enter the website.

A local cookie is one that is loaded onto the user’s computer by the advertiser. This can then be used to recognise the last referring channel and determine which tracking tag to display on the confirmation of sale page.

This is the cheapest, most efficient and accurate way to de-duplicate and it provides the advertiser with complete control over the logic employed.

This method is also referred to as “conditional tagging”.


Merchant X added a click append to their URL for each online channel to define where the customer had arrived from, a few of these are shown in red below: – dgm – Linkshare – Google PPC – Email campaign

They then wrote a script to place a local cookie on the user’s computer triggered by the presence of the click append.

They then established a rule that the click append source ID would be overwritten on a last referrer basis.

For a customer clicking through via a Linkshare publisher who then browses and returns via an dgm publisher the source id “?source=Linkshare” would be replaced with “?sourceid=dgm”.

The tracking tag logic at the sale confirmation page would recognise the “?sourceid=dgm” click append and show only the dgm tracking code, thus correctly awarding the sale to the last referring network.

All traffic must be tagged to ensure accurate cross-channel de-duplication. The last referrer wins method must be used across all channels to ensure fair de-duping.

Which online channels should I de-duplicate against?

De-duplicating sales reduces publisher conversion rates, and therefore their EPC/commissions also decline. Advertisers should therefore be cautious implementing de-duplication strategies as it could influence an publisher to promote a competitor.

We would recommend providing adequate notice to publisher s prior to the implementation of any new policy, especially one that may affect performance statistics. It is important to be specific about which channels will be de-duped and explain the reasons why. Every advertiser is different and policies vary considerably across the industry.

We have provided some example policies below and also some general rules that will ensure publisher s continue to support your programme.

Example Policy 1

Channels de-duplicated:

PPC Brand: No

PPC Generic: Yes

Price Comparison (advertiser’s own activity e.g/ Kelkoo): No

Display Advertising: No

Email Marketing: No

Direct publisher s/partnerships: No

Other publisher network: No, there are no other networks

De-duplication rule:

Last referrer is attributed the sale

Method used to de-duplicate:

Click append with local cookie

Note : For Example Policy 1 the advertiser has only one piece of CPA activity which is the Affiliate Window programme. They have a substantial in house PPC campaign with many generic and long tail keywords which they wish to attribute referring sales purely to this channel. For other channels including brand PPC the advertiser wishes to award the publisher recognising their influence in completing the sale.

Example Policy 2

Channels de-duplicated:

PPC Brand: No

PPC Generic: No

Price Comparison (advertiser’s own activity e.g/ Kelkoo): No

Display Advertising: No

Email Marketing: No

Direct publisher s/partnerships: Yes

Other publisher network: Yes

De-duplication rule:

Last referrer is attributed the sale

Method used to de-duplicate:

Doubleclick floodlight tracking tag

Note: For Example Policy 2 the advertiser uses a 3rd party solution to de-duplicate sales, they wish to reward the publisher s influence on the sale wherever possible. They are happy that a CPA rate be paid in addition to a CPC given the customer spent time on the publisher ’s site and subsequently browsed.

This advertiser however has another publisher programme as well as some direct partnerships. These are all also rewarded on a CPA model. The advertiser cannot pay the higher CPA cost twice, and therefore they must de-duplicate the sale to the last referrer.

Example Policy 3

Channels de-duplicated:

PPC Brand: Yes

PPC Generic: Yes

Price Comparison (advertiser’s own activity e.g/ Kelkoo): Yes

Display Advertising: Yes

Email Marketing: Yes

Direct publisher s/partnerships: Yes

Other publisher network: No, there are no other networks

De-duplication rule:

Last referrer is attributed the sale

Method used to de-duplicate:

Tagman container tag

Note: For Example Policy 3 the advertiser is using another alternative 3rd party solution to de-duplicate. They have decided to de-duplicate against all online channels.

They only ever want to attribute a sale to one channel. As a result of this they pay a higher commission to publisher s as they do not need to factor in duplication costs.


I have not read this yet, but like I said before attribution is the buzz word of the year for me


How to reward publishers for the right kind of consumer behaviour?

Whether we like it or not, 99.9% of digital marketers are using last click attribution models, and will probably doing so for the next year or so, at least.

So this leaves us with a challenge, how do we get the most of this ancient system?

We believe that is important to diversify or die in this industry and as such we are often forced to make compromises based on technical limitations not meeting the commercial models we predict possible in our minds.

Take for example this common problem that we see more and more.

  1. Consumer Jayden, interested in buying some fashion from a retailer online, clicks through to the site from a favourite content blog he reads every day.
  2. On the site they browse around and add a couple of items in the basket.
  3. They then go to checkout and see the coupon box in the checkout page.
  4. This triggers some form of thought, where is the coupon?
  5. Jayden then opens a new tab and does a Google search “Brand + Coupon,  and Jayden finds a coupon site not too dissimilar to,
  6. Jayden sees a code on this, usually clicking on the link to copy and opens the site.
  7. Jayden goes back to the original browser tab and completes the purchase using the coupon code copied from Retailmenot.

This is fairly common stuff.

What actually happens to this sale in last click attribution is the Coupon site has been attributed the sale based on a last click wins, even though Jayden was already in the check out.

So this goes back to the very title of the article “How to reward publishers for the right kind of consumer behaviour?”

We are now running last click attribution which works like this.

We work with the brand above to find their “mode average” time to complete a checkout (from beginning of checkout to Thankyou page).  Lets say it is 2.5 mins.

We then set a rule in place that deletes all sales that occur less than 2.5 mins from click to sale.

Note1: For Coupon sites about 30% of sales occur in less than 5 mins from click

Note2: For content sites about 5% of sales occur in less than 5 mins from click

The effect of this is that we are now only attributing sales to publishers where the journey time is greater than 2.5 mins.

The knock on effect for Brands is that coupon publisher ROI increased by 30%

The knock on effect for Coupon sites is that they really need to develop the content parts of their business so that they are adding more value for consumers and brands alike instead of focusing on “gaming” an old and antiquated attribution model as they are seemingly doing currently.

Australia spends per capita on eCommerce ranks in the top 2 globally contrary to BandT artcle Australia’s digital opportunity – Facts not sourced and incorrect

I was reading an article on B&T yesterday and found some of the information contained within to go against what I knew about digital marketing in Australia.

My main issue is specifically on this point below, I just don’t buy it.  According to all the data that I have compiled or studied Australia spends per capita on eCommercre are at least in the top 3 globally.

Quote taken directly from B&T

“Australia lies only 14th globally in terms of spend on ecommerce per capita, behind the Czech Republic and Luxembourg. (The UK comes in at #1). Although this situation is changing, as retailers get their heads (plus distribution channels and pricing) around the sheer vastness of the Australian continent and reflect relative costs in their prices. ”

What I really find amazing is that people can write articles pontificate facts and not source the data and get published in reputable publication.  In science this kind of reporting would be considered un-publishable. Oh but I forgot this is marketing after all!


Australia and Norway grabbed the No. 2 and 3 spots worldwide with average annual sales of $3,547 and $2,530, respectively, per person in 2012. The larger US market lags with just $2,293 in average sales per person last year.




Attribution modelling

Attribution modelling

This is fantastic article and currently I am writing something not too dissimilar.

IMO the industry will struggle to create a ubiquitous methodology for attribution as there are just such a large set of unknowns

 With your described position based attribution you fail to take into account the time on site each click delivers.

Through analysis one of our clients found out that for every additional minute a user spends on site that it is worth an additional amount of money in revenue and profit.

Therefore the time a user spends on site is critical in influencing the customer journey, as not each click to a site is created equal time on site therefore is the governing factor in executing attribution.


Attribution rules

If there is no click activity for a four day period prior to the sale then the sales is deemed organic and no channel is attributed the sale

If there is a click within a four day period of a sale being made then the following attribution model is instigated


5% of Sales Revenue = CPA = Analysis of click path { Channel A (time on siteA / Total time on site)+Channel B (time on siteB/ Total time on site) +Channel N (time on siteN/ Total time on site)}

It is worth noting there are significant known flaws with all current attribution models, under the below conditions attribution modelling falls down completely

  • When multiple users use the same computer, attribution tracking cannot identify which user agent is using the computer at any one time
  • When first and third party cookies are deleted are blocked by a browser
  • When a single user uses multiple devices to complete a purchase, currently attribution systems are not sophisticated in pairing multiple devices to a single user
  • If a user is a member of a cashback or loyalty points scheme any advertiser who wants to implement attribution and engage these type of audiences must have  a hierarchical structure that can supersede attribution in order to give a constant monetary value to these type of transactions

Every industry, brand, segment and sometimes even products have a different purchase paths and as such require their own customised attribution model.

And this is the problem with defining standard media attribution models. Because your businesses are unique, with unique products, strategies, competitors and environments, picking a cookie cutter (pardon the pun) attribution model based on an industry average can potentially be as misleading as last click attribution. 


The only way to guarantee attribution models are accurate and thus can be utilised for major strategic decisions such as media budget allocation between channels is to use a custom attribution model.



Don’t fall into the Coupon Trap, do it right or don’t do it all

This is a real hot potato in our office and so it seems across the globe.  Coupons we have written a great many articles about how to use Coupons in order to drive higher margin customers and we have published the findings here

And also first began this discussion way back in 2010

Using one of Australia’s largest online electronic retailers and measuring against a data set that ran into the several millions of dollars of revenue generated we found the following:

  1. 30% growth against PPC sales
  2. A comprehensive 21% growth against shopping comparison sites
  3. In addition coupon sites converted users 3.5 times more than PPC search

But there is a new discussion frothing over the edge of the coffee cup which is

Why show the coupon box on the final page of the checkout at all?

This article is certainly worth the read as it stimulates the need for debate on which side of the fence you are on.!

What I want to bring your attention to a really cool study performed by Red Letter Days Case Study whereby they tested the following

“The use of voucher codes has always been an issue at the forefront of the affiliate industry. Red Letter Days tackled this by developing a solution that displayed or hid the basket promo code box based on the type of referring affiliate. This ensured that customers driven from non-voucher code sites would not see the box (and therefore not be prompted to search for a voucher code), whereas visitors from a voucher code site would see the box.”

The logic is simple if the referring website was not a coupon site remove the coupon box from check out.  If it was keep it there.

This is not rocket science and can be happily A/B tested to see how conversion is affected.

3240 (1)

Read more:

Most of all we re iterate our recommendation on the seven rules of an excellent coupon strategy, if you stick to these you can not fail to play in the sweetspot of high margin sales from coupon sites, as opposed to the fast race to the bottom of site wide discounts coupons.

Making Coupons Work For You

We have been devising coupon strategies for more than two years, across a wide variety of categories, which has helped us to formulate a set of guidelines that, if adhered to, can make you money from your coupons:

  1. Use coupons on high margin products – don’t discount your volume drivers, use coupons to offer customers your higher ticket products or product bundles. This will ensure you never devalue your core products.
  2. Structure your offers such as this: “X% off when you spend $Y” where X% off gives you an increased nett margin compared to your average order margin, represented in the following formula: $Y Nett Margin – (X% of $Y) ≥ Average order nett margin
  3. Use coupons for upselling – always take the customer up the product range and thus improve their experience.
  4. Plan for your coupons to go viral – the very nature of coupons means that coupon sites are community driven and offer powerful word of mouth referrals. Good offers resonate, creating a ripple effect that can generate excellent user-generated content.
  5. Ensure your online checkout can reject inactive coupons. Most coupon sites do not remove out of date coupons – it’s your responsibility to make sure your checkout functionality has the ability to reject out of date coupons.
  6. Set expiry dates and unique identification to measure redemption. To create a call to action a coupon must have an expiry date and ideally a unique identification number so you can track back exactly where the coupon has come from.
  7. Always include stringent ‘Terms and Conditions’ for coupon use, enabling you to to protect yourself (e.g. ‘can only redeem one per household’, ‘only available to the first 50 participants’, etc).

Read more:

Email & Affiliate marketing delivers more sales than Facebook & Display put together



Good research from the US for 86 advertisers and 72M customers which shows how each channel stacks up in delivering new customers and life time value.  Email acquisition clearly is leading the way here and affiliate marketing holds up well too.



One thing that is not considered here is the actual Return On Advertising Spend or ROAS by Channel.  From dgm’s research in Australia of over $100M in sales delivered we can clearly see that affiliates delivers the most cost effective route to market to acquire customers, 800% more cost effective than display and 100% more cost effective than PPS Search




Get every new post delivered to your Inbox.

Join 309 other followers