Marketing attribution is a well-known technique for measuring marketing impact.
When customers arrive at an eCommerce store, their path may involve multiple touch points across various channels. For example, initial interest might be captured via an organic post before the customer visits the shop, leaves, and is converted via a retargeting ad.
In these situations, it’s essential to know which touchpoints played a part in the customer’s journey and the role each one played. Though they work differently, marketing attribution and marketing mix modeling (MMM) are two important ways to measure marketing impact.
Marketing attribution aims to answer those questions while providing insight into what customer journeys yielded the best outcomes. In addition, it provides a means to track customers at the user level.
What Are the Benefits of Marketing Attribution?
Marketing attribution lets marketers understand touch points on a customer’s journey toward conversion. Around 76% of marketers can attribute marketing efforts already, but the truth is attribution is becoming tricker.
Why? Because marketing attribution primarily uses Google Analytics cookies to log information via UTM parameters. Many users deny cookies because they don’t want to share personally identifiable information (PII), and Apple, Google, etc., are now working on cookie-less tracking that doesn’t expose PII. Collecting just the traffic source and no PII from the cookie is also possible.
Google Analytics for Marketing Attribution
Google Analytics attributes referral sources for UTM parameters. Marketing attribution mostly uses UTM (Urchin Tracking Module) parameters, strings attached to URLs that provide data about the link’s purpose.
Google Analytics collects data from UTMs and makes them available for analysis. When the UTM parameters are specified on the URL, these get parsed and placed into a Google Analytics cookie.
You need to tag campaign URLs with UTM parameter data to get UTM data into Google Analytics.
The main tags are:
- Source: Where did the traffic come from?
- Medium: Which channel type (e.g., organic social, search, paid social, email) produced the traffic
- Term: Used for paid search tracking
- Content: For A/B testing, different types of content
- Campaign name: Name used to track the campaign
There are various best practices for tagging UTMs, including:
- UTM tags shouldn’t be used for any internal links – this will override the original visit source
- Be consistent and keep your UTM tracking procedure consistent
- UTM tags are case-sensitive, and you shouldn’t use spaces – use underscores instead
- Don’t use variables you don’t need, e.g., content type
- Keep the total URL short
- Don’t include personal information
- UTMs don’t affect SEO
Marketing Attribution Models
Once your UTM parameters are loaded into Google Analytics, you’ll start getting data to find out where your customers came from and what links they clicked. Smaller businesses may have just a few campaigns and limited links, whereas larger businesses might have a wide range of campaigns and associated UTMs.
The tricky thing about attribution is it won’t necessarily tell you which touch point had the most impact. To do this, you’ll need to use attribution models to attribute credit to different touch points.
The default model is typically the last click, which means all marketing credit is assigned to the last touch point. This assumes the touch point closest to conversion is the most important (not illogical) but ignores other touchpoints. For example, your last touch point may only convert when the customer clicks on two to three other channels.
Marketing attribution models assign extra credit to different touchpoints. This enables you to see how modeling attribution affects the conversion values of each touch point.
The six main touch points in Google Analytics are:
- Last click
- First click
- Time decay
- Algorithmic (Custom)
Here’s a little more about each:
1. Last click attribution
The most common form of attribution assigns 100% of conversion credit to the last click in the customer journey. Down funnel clicks, such as retargeting ads, receive all credit. Here, your entire conversion will be weighted where the customer clicks last.
2. First click attribution
In contrast to the last click, the first click attributes 100% of the conversion value to the first touch point.
This is useful for discovering the most valuable channels for attracting top-of-the-funnel interest.
3. Linear attribution
The most basic form of multi-touch attribution assigns equal weight to all touchpoints. Useful for comparing impact across a level playing field.
4. Time decay attribution
More weight is applied to events close to conversion. This helps gauge how conversion is assigned when the emphasis is placed on the bottom-of-the-funnel activity.
5. Position-based attribution
Also known as U-shaped attribution. Position-based attribution applies more weight to the start and end of the customer journey, with less weight to the middle touch points. There is also W-shaped attribution which assumes the middle touch point is also important.
6. Algorithmic attribution
After collecting enough data, you can allow Google Analytics to model the customer journey and assign weight to touch points most likely to convert. In theory, this should provide accurate data on what channels provide the highest conversion value as tailored to a business-specific customer journey.
Google Analytics has a Model Comparison Tool, which enables you to compare models side-by-side to investigate channel value for different attribution weighting.
The goal is to discover how customer journeys affect conversion value to discover high-converting journeys. This enables marketers to adjust spend or activity depending on what channels have more or less impact.
Limitations of Marketing Attribution
Marketing attribution is far from a perfect science. While it provides an excellent tried-and-tested way to investigate traffic and conversion sources, it neglects other factors outside of tracked channels. For example, maybe some external factors are affecting attribution.
Marketing mix modeling (MMM) is one way to account for marketing attribution’s shortcomings. It doesn’t provide the same user-level tracking but does take external factors into account, which can greatly enhance precision marketing.