There are endless ways to reach potential customers online, and companies invest more and more money in digital channels. But there’s one thing that all advertisers worry about: How do they know which of the advertising measures will be successful, and which campaigns won’t deliver the desired outcome and need to be optimized? These questions need answers, otherwise budgets cannot be implemented effectively. This is what marketing attribution is about.
WHAT IS MARKETING ATTRIBUTION?
Marketing attribution is an analyzing method which allocates conversions (i.e. marketers’ targeted objectives such as a sale of a product or a registration) to the responsible advertising measures. Each transaction is usually preceded by a series of touchpoints from various marketing channels. A customer is thus mostly reached multiple times through advertising before completing the conversion, for example via social media posts, newsletters or banner ads. Each of these touchpoints along the customer journey influences the purchase decision of the customer. Marketing attribution is designed to find out how much influence each of them has so that marketing decision makers can rate the efficiency of their digital advertising measures.
Further information on this topic can be found in our whitepaper.
TYPES OF ATTRIBUTION METHODS
Generally, we distinguish between rule-based and data-driven attribution.
Rule-based attribution is based on specified rules and weightings determined by the advertiser. One or more winner touchpoints are selected in advance from the customer journey and are attributed a certain value. That way, this method provides some initial answers about the efficiency of marketing campaigns, but the result is manipulated as it is dictated by fixed rules.
Data-driven attribution is based on objective modeling techniques in the form of mathematical algorithms. They analyze the value of each touchpoint throughout the buying process of a user and calculate, which of these touchpoints has a positive impact on the targeted conversion. This methodology is not only more accurate because it is based on the analysis of a variety of data. It records the current situation and automatically detects changes in the marketing efficiency. It allows advertisers to better plan and optimize marketing measures in the future.
RULE-BASED ATTRIBUTION MODELS
Last Touch: With a Last Touch model – also called Last Cookie Wins or Last Click Wins – 100% of the success is assigned to the chronologically last marketing touchpoint before the conversion. To date, it is still the most widely used attribution model in practice, therefore it is often used as a reference when comparing other attribution models.
First Touch: With a First Touch model – also called First Cookie Wins or First Click Wins – 100% of the success is assigned to the chronologically first marketing touchpoint before the conversion. This model is particularly suitable for evaluating branding and awareness campaigns because it measures which advertising media reaches the user in order to generate first recognition.
Linear: With a linear model, each interaction that contributed to a conversion is valued equally. It is assumed that all touchpoints build on one another and have the same value. This model is primarily suitable for advertisers whose campaigns promote brand building and awareness. The sales cycle is usually quite long and it is assumed that each touchpoint strengthens the brand awareness.
Time Decay: With the Time Decay model, touchpoints and their contribution to a conversion become more important the closer they are to the conversion. This model suits advertisers whose sales cycle only has a very short decision phase, and/or for campaigns that are designed to sell fast. With this model, touchpoints that are chronologically closer to the conversion receive a better rating than “older” touchpoints.
Position-Based: With a position-based model, also called U-shaped model, the first and last touchpoints throughout a customer journey are valued higher in terms of the conversion success than the remaining touchpoints. This model is a combination of the last-touch and first-touch model and ideal for advertisers who primarily want to rate their awareness campaigns and final sales.
Custom: With the custom model, value contributions can be determined individually. That way, advertisers can distribute conversion percentages in a way they think is best for their business.
The common denominator of all rule-based models is that the conversion shares are determined in advance by the advertiser. This means that the marketing efficiency, i.e. the value of the advertising, is set in advance and remains constant over time. Whether these fixed values match the actual impact of those touchpoints remains unanswered.
Data-driven attribution defines the impact of each touchpoint based on the collected customer journey data. It is based on complex mathematical algorithms that continuously analyze all of the user’s touchpoints and calculate which ones have the most impact on the conversion. With this method, all user interactions are considered, i.e. all successful customer journeys as well as unsuccessful journeys, also called user journeys. Advertising measures that didn’t drive the desired outcomes provide just as valuable information as successful sales. At the same time, the data-driven attribution model considers interactions and synergies between individual channels and devices. It also automatically includes changes in customer behavior and market conditions.
The great benefit of this model is how objectively it rates touchpoints throughout the customer journey. Unlike the rule-based method, there are no preset rules or weightings. According to the situation, it analyzes the actual value contribution of each touchpoint. Therefore, these models are also called dynamic attribution models.
Marketers can use this method to not only objectively measure their campaign performance. With these new insights, they can also optimize future marketing measures and distribute budgets more efficiently to all of their campaigns.
WHICH MODEL IS THE RIGHT MODEL?
A data-driven attribution model defines the value contribution and impact of each touchpoint to a conversion. This method undoubtedly provides more valid results about the advertising impact of campaigns and is therefore always the better approach.
But as convincing as the benefits of dynamic attribution may be – it is not an option for every advertiser. For a data-driven model to reach its full potential, there are certain requirements that need to be fulfilled, e.g. sufficient data volume and data quality. In our next blog post we will discuss these requirements in greater detail.
If these requirements cannot be fulfilled, advertisers should still use a rule-based attribution model. Because any attribution model is better than siloed channel view, in which marketing measures are evaluated separately and independently of one another. Each rule-based model offers insights into a customer journey. Ultimately, choosing the right attribution model depends on the advertiser’s goals and business requirements.