Complete Guide to Attribution Modeling

For any online business, the dream scenario is for a visitor to land on their website, see a product or service that tickles their fancy, and they immediately click that ‘Buy’ button. Sadly, conversions are rarely that straightforward. In this day and age, potential customers will land on the same site several times before deciding to make a purchase.  

This can make the tracking of marketing channels a tricky proposition for marketers. Say a customer visited a website 3 times via 3 different avenues. One time via a search engine query, another due to a social media post, and the third because of a retargeted ad. Which one ends up receiving the credit for the sale? 

With the customer journey featuring multiple touchpoints, answering the question can be tricky. This is where marketing attribution modeling can help.  

What are attribution models in marketing?  

Attribution models are used to investigate the touchpoints, aka marketing channels, that deserve credit for landing a conversionA conversion is a desirable result on a website that leads to an action such as completing an order, filling out a form, or simply clicking on a link.. There are various attribution models, and these each distribute a conversion’s value across the touchpoints differently. Here are the common types of attribution models we’ll focus on for this guide:  

More on those later.  

With a model comparison tool at your disposal, it’s possible to analyze a conversion and how its value is distributed across each marketing attribution model. 

As for which attribution model to use, there isn’t necessarily a right or wrong answer. This can depend on factors such as buying cycles and your business goals. As a result, there are certain models which are better than others for certain situations. It’s not uncommon for companies to select a primary attribution model they use for analysis and reporting.  

However, don’t feel you have to restrict yourself to a single model. To gain a full understanding of the importance of utilizing multiple touchpoints across the customer’s buying journey, it’s important to compare performance across different attribution models.  

The comparison process is necessary to understand how multiple marketing channels combine to generate conversions. It’s also required to assign conversion values for each individual channel you use. By doing so, you gain a more accurate picture of what is – and isn’t – working with your marketing efforts. 

To clear things up even further, let’s take a closer look at each attribution model: 

First Touch

The First Touch attribution is where 100% of all the credit for a conversion is given to the first interaction. Even if the customer used numerous touchpoints before they decided to make a purchase, only the initial one matters for this attribution model.  

As an example, say a customer first discovers your business via Twitter. In this case, Twitter gets all the credit for the eventual sale – no matter what happens in between. So even if they clicked on a blog post the following week, then a targeted ad the next, and found a specific product page via a Google search query, only Twitter is considered for the First Touch attribution.  

There are benefits to using First Touch. It’s straightforward and easy for everyone to follow. It can also be useful if your business operates in an industry with a short buying cycle, where customers tend to convert immediately. However, there are obvious limitations, such as ignoring the role other marketing channels play later in the sales funnel if you only use this attribution model.  

Last Touch 

As you may gather from the previous attribution model, Last Touch involves giving all of the credit to the very last interaction a customer has with your business before buying your product or service.  

For instance, a visitor could first land on your website via an organic search. Then a couple of weeks later, they’re reminded about your business from a Facebook Ad, and they click on this to learn more about what you have to offer. Then the next day, they directly visit your site to buy one of your products. In this situation, the direct traffic receives 100% of the credit for the purchase.  

Again, Last Touch is incredibly simple to implement and analyze. It also delivers a large slice of accuracy. Digital marketing might dance around all over the place these days, where people use multiple touchpoints, devices, and browsers. Still, there’s one aspect that cannot be confused: the last interaction before a conversion. The downside, of course, is this attribution model ignores everything that occurs prior to the final interaction.  

When using the Google Analytics attribution model, Last Touch is the default option. This means if you’re using Google Analytics to view standard conversion reports, you’re only seeing goals attributed to the Last Touch – aka the last interaction between the customer and your business. Make sure to change the attribution model for more details regarding conversions.  

Multi-Touch

This is another situation where you can gather what the attribution model covers based on the information mentioned above! The Multi-Touch attribution accounts for various marketing channels and campaigns responsible for a conversion.  

It doesn’t just add every marketing path a customer walks down, either. Multi-Touch incorporates the weight – aka importance – of each touchpoint during the customer’s journey. It provides more clarity about the entire customer journey, particularly when judged against both First Touch and Last Touch attribution models.   

However, the main negative of this is the added work it involves. There’s a lot more data that needs to be factored in and analyzed.  

Linear

What if there was a solution that involved less work than Multi-Touch but accounted for every interaction? That’s where the Linear attribution model comes to the forefront. This option covers all touchpoints by a customer with your business, yet it doesn’t assign a specific amount of credit for each one. Instead, it is split equally across every interaction.  

For example, say a customer clicks on an email link that directs them to your website. Then they find your business on Pinterest. Afterward, they visit your site after a Google search. Finally, a week later they visit your site directly and purchase one of your products.  

This situation involves 4 touchpoints in total. With a Linear attribution model, the 100% credit total is split equally between them. The result: each touchpoint is given 25% of the credit for the sale.  

The Linear approach supplies a balanced appearance of your entire marketing strategy. With that said, there is a disadvantage to assigning equal importance to each marketing strategy. After all, most businesses will benefit from certain promotional tactics more than others.  

Time-Decay 

The Time-Decay attribution model functions similarly to the Linear attribution. This is because the value of each conversion is spread across multiple events during the sales funnel. The difference, however, is that each touchpoint isn’t provided with an equal value. Instead, the value considers when the touchpoint occurred during the sales funnel.  

In that regard, greater value is attributed to interactions that happen closer to the purchase time. This means the first interaction receives the least credit, whereas the last interaction is handed the most.  

A Time-Decay model is suitable for various business types, including those that place a strong emphasis on relationship-building. It’s also recommended for a company that deals with long sales cycles, which is common for those dealing with expensive business-to-business purchases.  

Position-Based

The Position-Based model is based on splitting a conversion’s credit between the first and last interaction. Its other name, U-shaped, helps to illustrate the format of this attribution model.  

The first and last interactions are each handed 40% of the credit. What happens to the rest of the 20%? This is spread out across any other interactions that occur in between. For instance, if there are 6 interactions in total, the first receives 40%, the next 4 get 5% each, and the last interaction gets 40%.  

As it still delivers credit for every interaction, Position-Based attribution modeling is a top pick for many organizations. Alongside crediting every marketing element, it also places the strongest weight on the two most important elements of the sales funnel – the first and last interactions.  

Why does attribution modeling matter and how does it affect reporting? 

Attribution modeling is useful as it helps to highlight the prevalence of your marketing efforts. It can let you know which marketing channels are pulling their weight while also pointing out the ones that are not providing a suitable return on investment. With this type of added information for your reports, it helps your business to further refine its promotional efforts. Your money can be spent in better places, resulting in better success throughout your entire sales funnel.  

With that said, and despite their usefulness, it’s fair to say that attribution models are among the most complicated aspects of digital marketing. Are you still struggling to get your head around this topic? If so, don’t worry – you can Contact Us for assistance. Our marketing experts know how to select the right attribution models for your business type, report on it effectively, and make the most from this tool.