Companies are demanding more accountability for every dollar. Seventy percent of marketers indicated that attribution is highly valuable in a recent survey conducted by Forrester. Although most of the survey participants also mentioned that they struggled to find the best attribution methods.
For decades, marketers have relied on simple last-click attribution model which allocates 100% credit to the last marketing channel before a purchase. Today, brands demand greater accountability and a scientific a scientific measurement of the value of each marketing contact that contributed to a desired outcome; this is a way for marketers to clearly understand what’s working and what’s not.
An attribution model uses a set of rules to determines how credit for sales and conversions is attributed to touchpoints in conversion paths.
For instance, last-click attribution model assigns 100% weightage to the last touchpoint that immediately precedes the sale or conversion.
Similarly, the first-click attribution model assigns 100% weightage to the first touch point.
As time passed marketers started believing that first-touch and last-touch attribution models were flawed. Multi-touch attribution model assigns a value to each touch point before the actual purchase or conversion. Time Decay & Linear are the most popular multi-touch attribution models.
The linear attribution model assigns equal weightage to each touch point before the purchase.
Custom attribution models can be also be created using rules that you specify. Using customer models is a great way of validating the set of assumptions across the conversion path data. You can create custom attribution models on Google analytics.
Market Mix Modelling
Marketing Mix Modelling uses an analytical approach that uses historical information, such as the syndicated point-of-sale data and companies internal data to quantify the sales impact of various marketing activities. Market mix models are extensively used by the consumer product, automotive, and hospitality companies.
Market Mix Models have a clear bias in favour of time-specific media (such as television commercial) versus less time-specific media (such as ads in monthly magazines). So if you using market mix models to evaluate sales impact from blog marketing it will not work.
Choosing An Attribution Model
If you are using both Multi-Touch Attribution & Market Mix Modelling you will find huge discrepancies. For companies which heavily invest in both digital and offline marketing, it makes sense to adopt a model that combines both Market Mix Modelling (MMM) & Muti-Touch Attribution (MTA).
Companies like Convertro, VisualIQ and C3 Metrics provide a unified model that combines the best of MMM and MTA. If your business gets revenue from both offline and online channels of purchase then a unified channel should be a great choice.
For companies that get revenue from online channels of purchase multi-touch attribution might be the best bet. But using a combination of attribution models will help you discover new insights on the touch points.
For instance, if you want to know which channel generates more prospects then the first-touch attribution should answer this question. If you want to know which channels act as good closing touch points then last-click attribution should be able to answer this. Leveraging multiple attribution models will unlock new insights to make better decisions.