Nowadays, while investing in multi-channel online marketing strategies, marketers face the obvious consequences of this large focus spread: namely, the complexity of the customer journey and the marketing attribution dilemma. 

In addition, new ePrivacy-related regulations that will soon eliminate third-party cookies certainly do not make life easier. This upcoming change puts the functionality of the existing attribution models in question. 

As a result, the question of how attribution modeling can be done in the future remains the elephant in the room for many marketers. 

Well, it’s time to face the elephant and figure out what to do next to best prepare yourself for the upcoming change.

Let’s take a deep dive into the topic.

Table of contents: 

  1. What is marketing attribution, and why do marketers need it?
  2. The seven most common attribution models
  3. The future of marketing attribution in the cookie-less world

Get a free recording of our Webinar about the future of attribution modeling! 

What is marketing attribution, and why do marketers need it?

First, let us briefly remind ourselves what marketing attribution is and what it does.

Simply put, marketing attribution is a way advertisers define the contribution of various marketing tactics to sales or conversions. In other words, it’s a model that determines the return on investment (ROI) that each channel generates. 

Let’s make an easy example to understand the core idea. 

Imagine that you need to buy a new sofa. 

It’s very unlikely that you will make your purchase after visiting one website. Most probably, you’ll scroll through several websites, read a couple of articles, check out a few Pinterest looks, subscribe to a couple of Instagram accounts, view a certain number of targeted ads and visit a couple of stores before making the decision. 

As a user, you go through many touchpoints and channels before deciding to buy a specific brand. All these touchpoints are interconnected. Acting together, they help marketers achieve their goal: convert a customer.

But how do we define which channel or a touchpoint contributed the most to the sale? Or in other words, which channels are worth investing in?

That’s where attribution models come into play. 

Marketing attribution helps marketers to give the right credit to the right touchpoint. 

While having the advantage to nurture a potential client on multiple channels instead of just one, applying multi-channel strategies make it harder for marketers to estimate the overall impact of a particular channel on marketing ROI.

Attribution models are designed to solve these riddles and help marketers understand the estimated contribution of each channel to the overall return on investment. In other words, by using attribution models, marketers can better understand which tactics bring the most value.

The seven most common attribution models

Let us quickly go through the most common attribution models and their core principles. 

Generally, depending on the level of complexity, there are two types of attribution models: 

  1. Single source marketing attribution: where all the credit goes to only one touchpoint or interaction. 
  2. Multi-source marketing attribution: where the model attributes the credit across multiple touchpoints. 

Let’s have a quick look at the most popular marketing attribution models.

Single-source marketing attribution models:

The most common single-source attribution models are first-touch and last-touch. 

First-touch attribution – This model gives all credit to the first touchpoint the lead engaged with. 

Last-touch attribution – In contrast to first-touch attribution, this model gives all credit to the last touchpoint that led to a conversion. 

Already here, we see how much information we lose when focusing only on the first and last touchpoints. Both these models basically ignore everything that happens in between and, as a result, obscure all the other marketing efforts. 

Therefore, many marketers consider them inaccurate and outdated, especially in the context  of multi-channel strategies.

Multi-source marketing attribution models:

Unlike single-source models, multi-source marketing attribution models take into account more than one touchpoint while attributing the conversion. 

Linear – this model equally distributes the credit to each touchpoint that led to a purchase. 

U-shaped – This model got its name due to the weight distribution amongst the touchpoints. If you graphically imagine the distribution – it will form a U-shape. This model attributes 40% of the credit to the two key touchpoints: the first touchpoint and the touchpoint that created the lead. The remaining 20% is divided between any interactions that occurred in the middle.

W-shaped – Similar to U-shaped, this model also prioritizes specific touchpoints, but instead of two, it prioritizes three major touchpoints. It attributes 30% of the credit to the first touchpoint, 30% to the interaction that created the contact (lead-conversion), and 30% to the last interaction that created the deal (deal-conversions). The remaining 10% are distributed evenly across all interactions between the first interaction and deal creation.

Full-path – Full-path model goes one (more) level deeper. This model adds one more step. It gives 22.5% to four key touchpoints: first interaction, touchpoint that led to contact generation, last interaction that created the deal, and interaction that closed the deal. The remaining 10% of the credit is distributed among all other in-between touchpoints. 

Time-decay – This model gives more credit to the most recent interactions. In other words, it assumes that later touchpoints had a bigger impact on the sale or conversion. The time-decay attribution models are best suited for lengthier sales cycles. Therefore, it is pretty popular in B2B companies.

These models are typically used as a default by many companies. However, there are, of course, also Custom attribution models that are tailored to a specific business case. Custom attribution models are simply models where you can assign your own attribution weights to various touchpoints throughout the customer journey.

All in all, multi-source attribution models are most suitable for more complex customer journeys. Therefore, such multi-source attribution models are prevalent in B2B marketing environments, where customer journeys are longer and more complex. 

Despite its many benefits, the future of attribution modeling doesn’t look so bright. 

With the ever decreasing possibilities of cookie-based tracking, mainly due to privacy-compliance regulations, marketing attribution as we know it today might soon become a relic of marketing history. 

What are cookies?

In digital marketing, there are two types of cookies: first-party and third-party cookies.

First-party cookies

First-party cookies are cookies that are used to improve your browsing experience. They are directly stored by the website you visit. Thanks to first-party cookies, the website can provide an autofill function that saves your time when filling out another form, remembering language settings, and collecting analytics data. When we talk about first-party cookies, only the user and the website have access to this cookie – external advertising platforms cannot access this information. 

Third-party cookies

Unlike first-party cookies, third-party cookies are those tracked by websites other than the one you are currently visiting. As you might have guessed, advertisers are the primary users of third-party cookies. These cookies are helping advertisers get an in-depth view of users’ activity. When a user accesses a website, both parties obtain a cookie. These days after the user gives consent for third-party cookie tracking, additional cookies are created for the advertisers which help them identify and track the users along the journey. 

As data privacy issues become more relevant for many users, digital marketers need to prepare themselves to live without third-party cookies. 

In a cookie-less future, marketers will see a reduction in conversions for much of their campaigns. It will become significantly harder to attribute conversions without opt-in from users. Consequently, it will be much harder to measure the effectiveness of their advertising campaigns. 

This means that many analytics tools that heavily rely on third-party data, such as Google Analytics, will have difficulties attributing conversions and providing a clear view on cross-channel performance.

On the other hand, solutions that guarantee 100% data ownership and a full GDPR-compliance, such as Matomo and Plausible Analytics, will become a necessity for digital marketers. Find out more about Google Analytics alternatives in our previous article.

Marketing attribution technologies of the future

So, how do you attribute conversions in the future? Without a doubt, the market will need next-generation attribution technologies. 

Web fingerprinting can be considered as one of the alternatives to cookie-based tracking. However, this method does look more invasive compared to the classical cookie-sharing approach. Moreover, with the growing data privacy concerns, such alternatives will retire pretty soon. 

Statistical models such as causal Impact analysis and Marketing Mix Modeling, on the other hand, have all the chances to become a GDPR-friendly alternative for cookie-driven attribution modeling. These methods would help marketers get more insights into their customer journey without tracking user ids and cookies. By leveraging statistics, marketers can discover patterns and learn more about their data while remaining privacy-focused. 

However, to be effective, these models require: 

  • Vast volumes of data
  • Having multi-channel data in one place 

Therefore, there will be more demand for solutions that aggregate data from multiple marketing platforms and provide a clear overview across channels while ensuring data safety. 

Solutions such as Nexoya provide complete control over your data across channels. Nexoya helps you to grasp the holistic view and use AI-based models to optimize your performance across multiple channels. Using statistical models such as adstock modeling, cross-correlation, and pattern analysis, we allow you to attribute your conversions without tracking cookies and get more insights into your data.

As a result, Nexoya helps you to allocate your budget more effectively, ensuring the highest possible return on investment. 

The cookie-less world is here to stay. And therefore, the earlier you start adjusting to these changes, the smoother the transition will be. 

How should companies prepare themselves to work without this data in their strategies?

Find out more in our Webinar about the future of attribution modeling!