With today’s ever-growing collection of digital marketing channel data, choosing the right multi-channel attribution model is becoming increasingly important.
The goal of multi-channel attribution modeling is to give credit where credit is due between all channels used along a consumer’s journey to conversion. As the consumer journey becomes more complex, it is important than ever to understand multi-channel effects.
MORE CHANNELS, MORE DATA FOR MULTI-CHANNEL ATTRIBUTION MODEL
Digital marketing today takes many forms. One product or service could have an advertising presence across several platforms, each with independent targeting and management. These channels work simultaneously to drive digital consumers towards a measureable conversion, and together gather a set of varied, incredibly detailed data that can be leveraged in analysis. With the right model you can:
IDENTIFY a channel’s strengths and weaknesses
OPTIMIZE channels for different positions along the consumer path
EMPHASIZE contribution from some channels more than others
ISOLATE channels or positions of the consumer journey for closer scrutiny
PRIORITIZE channels and positions more or less than others
STRENGTHS AND WEAKNESSES OF COMMON MULTI-CHANNEL Attribution MODELS
LAST NON-DIRECT CLICK ATTRIBUTION MODEL
All direct traffic is ignored and 100% of the credit for a sale goes to the last non-Direct channel the customer used before converting. Google Analytics uses this model by default when attributing conversion value in non-Multi-Channel Funnels reports.
This model intentionally understates the impact of direct traffic and undervalues other channels earlier in the journey. It does not credit brand recognition or brand value, which often plays an important role in driving direct traffic. Instead, it provides a simple picture of the most effective non-direct channels. As analytics becomes more sophisticated and tracking consumers through multiple channels becomes more reliable, it is critical to consider the potential impact of channels earlier in the consumer path.
FIRST INTERACTION ATTRIBUTION MODEL
The first interaction along a conversion path receives 100% of the credit.
This model ignores any channels used between the first interaction and a completed conversion. It does, however provide a sense of where consumers start on their consumer path, which can be a helpful perspective on digital marketing efforts.
LINEAR ATTRIBUTION MODEL
Each channel along the conversion path shares equal credit for the conversion.
By giving attribution to all channels, this model lessens the value of any particular channel or position along the conversion path. It attributes all channels that work, not just the channels that work well. Because of the universally-distributed attribution, channels with greater attribution overall are simply used more often than others.
TIME DECAY ATTRIBUTION MODEL
Credits all channels along a conversion path, but channels closer in time to the conversion receive more credit.
This model generates more detail through its association with time. Its time decay value is also adjustable, which gives flexibility for varied average time lag before conversions. Like the Linear model, it provides a composite channel value for all positions along the conversion path to analyze as well.
POSITION-BASED ATTRIBUTION MODEL
80% of the credit is equally split between the first and last interaction, and the remaining 20% is distributed equally between any middle channels.
Position-based attribution modeling can attribute first, middle, and last channels based on configurable attribution weights for each position. This allows for all channels to be attributed and given adjustable significance for each position along the conversion path.
MULTI-CHANNEL ATTRIBUTION IN WEB ANALYTICS
As tracking technologies continue to advance, adaptations must be made to fully utilize their capabilities. By utilizing alternative attribution models in the analysis of conversion traffic, additional perspectives can be used to interpret a customer’s conversion journey in new ways. Marketers can identify a channel’s strengths and weaknesses, and optimize them for different positions along the consumer path. It is possible to place emphasis on the contribution from some channels more than others or isolate them for closer scrutiny. Finally, marketers can prioritize digital marketing channels and conversion path positions moving forward.
MARKETERS’ NEXT STEPS WITH MULTi-CHANNEL ATTRIBUTION MODELS
To get further down the road to attribution, begin comparing alternative attribution models with data. If you are using Google Analytics, the Multi-Channel Attribution Model comparison tool is an easy way to start exploring data through different attribution models. Look for new insights in the contexts these models create, and see what multi-channel attribution can do for a marketer’s analysis.
INFORMATION SOURCES: Google, Occam’s Razor, Occam’s Razor 2
