Every marketer encounters them from time to time, measurement differences or other deviations in-between platforms that collect data. Let's agree on this first: no two web analytics services are the same, so a 100% match is rare. However, with a few simple tips it is possible to match all your (analytics) platforms as optimally as possible.
In addition, it is especially important to know why there are differences between different platforms. There are an enormous number of measurement and attribution methods, with many more variations.
There are several causes for measurement differences. It is important to realize that there will always be measurement differences between different online platforms. Measuring in Google Analytics is certainly advisable, but it should not be experienced as leading. Common causes are for example:
- Cross-device tracking 1
- Missing referral information 2, 3
- Blocked Analytics scripts / Adblockers 4,
- Duration of cookies
- Difference in attribution
Below we explain how to get your statistics as correct as possible (at least in Google Analytics) and why the statistics differ between platforms.
What impact does a misinterpretation of the statistics have on my campaign?
One of the most common reasons for high rejection is that advertisers do the conversion attribution based on a Web Analytics service such as Google Analytics or Adobe Analytics (the former Omniture) and keep the source of the last point of contact. In doing so, they are assuming that the rejection option within the Daisycon system is intended to deduplicate across all channels. Unfortunately this deduplication possibility should only be applied for rejecting invalid (e.g. returned) orders and, at most, deduplicating with other affiliate networks.
In this last case, a Last Cookie Count script is the preferred solution, which prevents duplicate transactions before they are measured, automatic network deduplication is also recommended by the IAB.
Affiliate marketing is a separate branch of online marketing. Based on an extensive and relatively complex measurement solution, where different (anonymized) methods are used to match a conversion to a publisher from the campaign.
Web Analytics services serve a totally different purpose than affiliate tracking software. The measurement and attribution algorithms of the former are far too primitive and generic to be used as an affiliate dashboard.
They are reporting tools, not management tools. The difference in technique and purpose is usually the cause of measurement differences.
Web Analytics services
Tools such as Google Analytics were originally designed to track and optimise traffic to and on a website. It is the way to optimise the customer journey and sales funnels within the website and make strategic decisions about the use of, for example, advertisements to attract new visitors. Put simply, they are reporting tools for visitor flows.
Another use case is the optimisation of Google Ads campaigns. By linking Google Ads and Analytics, it becomes possible to track Google Ads visits with greater precision. By juxtaposing the data from both platforms, the webmaster/marketer has a lot of data available for optimising SEO/SEA texts and keywords.
This is also the reason why Google Analytics can measure and attribute conversions. It is therefore logical that the attribution models are based on less performance based channels such as Google Ads.
Although a large part of the advertisers/marketers we deal with are often familiar with the platform, Google Analytics training does not or hardly discuss comparing statistics with other platforms. This is exactly where it goes wrong: there is little knowledge about the functioning of special channels such as affiliate marketing.
Publishers can forward visitors who are at different points in the purchase process. So before they proceed to a transaction, visitors have often already had several interactions with the website via various online channels. Almost all advertising channels are paid on the basis of a CPC or CPM fee. The affiliate channel, on the other hand, is the only channel where a performance based payment model applies (CPS / CPA / CPL).
Publishers in the beginning of the conversion path, a blogger or content website for example, will in many cases not be the last source for a conversion. Contrary to other online channels, the publisher will not receive any compensation for their proven contribution when they are judged on the basis of the latest source.
With Google Analytics' Model Comparison Tool, it is possible to view the influence of the different attribution models on the different marketing channels.
The default measurement duration (overview period / look-back window / cookie time) in Google Analytics is 30 days. This can be adjusted to a maximum of 90 days. It is important to match this as closely as possible to the measurement duration of your campaign within Daisycon.
The standard session duration in Google Analytics is 30 minutes. This can also be adjusted to a maximum of 4 hours. Chances are that a transaction with a 'click interval' of more than 30 minutes in Google Analytics will be assigned a different source.
If we would only rely on the last source in Google Analytics, it would become less and less interesting for many publishers in the beginning of the conversion tunnel to use promotion.
In the Daisycon system it is possible to view the Engagement Mapping per transaction. The Engagement Mapping provides insight into not only the latest, but also the previous ads interactions (publishers) within the affiliate channel that have influenced the final purchase decision.
Also in Google Analytics it is possible to analyse the contribution of different channels in the Multi-Channels funnel reports. Depending on the UTM tags added to the campaign URLs, if you choose 'Standard channel grouping' as the primary dimension, you will see the value of the affiliate channel in this report. You can adjust the overview period up to 90 days before the conversion.
A visualisation of the entire conversion path can be found in the report 'Top conversion paths'.
Affiliate statistics in practice
Advertisers generally look purely at their Analytics environment, usually due to a lack of knowledge about other platforms. As mentioned before, these statistics (even after correctly setting up the affiliate marketing channel in Analytics) will seldom or never fully match Daisycon.
Many people interpret affiliate marketing as "you only pay if it delivers a sale", without further discussion of the measurement method and what "delivering a sale" exactly means. If it is subsequently explained that this is based on a certain measurement duration, one often feels misled.
It is important that the negative image is put into perspective. With affiliate marketing there are few paid actions and the risk is low. Publishers should receive a reward for the contribution they make, this certainly applies to lower funnel publishers who rarely deliver the converting click but do produce good quality content about your brand or product.
Make sure that they can receive that reward, so that they will continue to promote you. Even if the reward has to be adjusted downwards. In that case, the compensation of the campaign is simply miscalculated.
The discussion that ensues is therefore often to the disadvantage of the publisher from the affiliate campaign. What people don't think about is that less-performing channels such as Google Ads or Facebook Ads are usually also not "deduplicated/rejected" if the last-click is the affiliate channel. There is no such possibility.
In that case, the costs are justified as a contribution to the final conversion, just as it should be done with affiliate marketing.
Due to the unique operation of the tracking system, the only reliable reporting tool is the Daisycon dashboard. The click path is usually fully transparent and traceable, the page of the pixel call is also recorded and can therefore be proven.
Together this should provide conclusive proof of the contribution of Daisycon publishers.
We have learned that there can be a variety of reasons why another tool does not assign this transaction or assign it differently to the affiliate channel. Different criteria are used. However, this does not mean that Daisycon is wrong.
There was a click from a publisher website, which led to a transaction where the Daisycon pixel on the conversion page of your website registered a transaction. Barring exceptions, there is basically no reason to deny the contribution of this publisher.
Affiliate marketing is a form of marketing that involves very low risk. After all, one only pays if a sale actually results from a click, the clicks themselves are free of charge. Rejecting a transaction because it cannot be found in Google Analytics does not benefit the campaign. If, conversely, the affiliate channel is last-click and Google Ads appears in previous clicks, those clicks are paid for.
Use the information in this article to make sure your dashboard and information are as well aligned as possible. In this way you can limit the measurement difference enormously. A basis of correct, complete information that is interpreted correctly is extremely important for a successful affiliate campaign.
Are your acquisition costs too high? Then don't disapprove transactions, but adjust your payout. Ideally, you will earn from your conversion and the publisher will receive a reward for doing so. Even if the payout has to be slightly reduced.
- Daisycon measurement platform is leading.
- Web Analytics services have a different way of measuring and attributing
- Rejection is intended for cancelled, erroneous or returned orders/transactions.
- Google Ads will always be paid even if they are not the source of the last-click.
- Provide appropriate compensation to avoid the rejection of a legitimate contribution.