.png)
In today's complex digital landscape, where customer journeys involve multiple touch points and interactions across various channels, attribution modelling has become an essential tool for marketers to accurately measure the effectiveness of their campaigns.
There are two main models: Multi-Touch Attribution and Marketing Mix Modelling.
1. Multi-Touch Attribution modelling (MTA)
In simple terms, MTA is the process of assigning credit or value to different marketing touch points that contribute to a conversion. It helps marketers understand which channels and campaigns are driving the most conversions and revenue.
For example, if a customer first discovers your brand through a Facebook ad, then later sees a Google ad and makes a purchase, attribution modelling would analyse the impact of both of these touch points in the customer's journey to conversion.
What channels can be included in MTA?
The main ones are:
● Social media ads
● Paid search (Google Ads)
● Display/banner ads
● Email marketing campaigns
● Organic search (SEO)
● Direct Mail
● Targeted letter box drops
Beyond these, it can even include customer interaction data collected by contact centres.
Why it’s important
Effective attribution modelling allows marketers to make data-driven media buying decisions and allocate their budget more efficiently. By understanding which channels and campaigns are most effective, marketers can optimise their strategies to drive better results.
Moreover, attribution modelling also helps in identifying and eliminating any ineffective or underperforming marketing efforts. This not only saves time and resources but also ensures a higher marketing ROI.
Plus, having a better understanding of how customers behave and interact with different touchpoints can also help in creating more targeted and personalised marketing campaigns.
By neglecting proper attribution modelling, marketers risk wasting valuable resources and missing out on potential conversions.
How it works
The multi-touch attribution model assigns weighted 'credit' to all the touch points leading up to a conversion.
Example:
A customer sees an Instagram ad, clicks on it but decides not to make a purchase. Later, they see and click on a Google ad and revisit the next day by directly typing in the website URL and making a purchase.
The multi-touch attribution model would credit all touch points involved in this customer journey - Instagram ad, Google ad and direct visit.
|
At UniFida, we use a time-based attribution model that assigns credit the closer a touch point is to the next touch point or to conversion. When combined with machine learning, this method produces near-real-time results, making it easier to optimise campaigns and budgets.
In short, multi-touch attribution allows for a more holistic understanding of the customer journey and how different touch points impact conversions.
2. Marketing Mix Modelling (MMM)
MMM is also known as Econometric modelling. Unlike the multi-touch model, it takes into account both indirect and direct marketing channels alongside external factors like seasonality and economic conditions. Indirect channels include things like TV, radio, print media and OOH.
It uses statistical techniques to analyse historical data and measure the impact of different marketing channels on sales.
Machine learning algorithms are also often incorporated to improve the accuracy and efficiency of MMM, much like in the UniFida MTA solution.
For example, MMM would consider not only the Instagram and Google ads from the previous example but also factors like TV commercials and seasonality to determine their impact on sales and conversions.
For large companies that use a combination of direct and indirect marketing channels, MMM can provide a more accurate and comprehensive understanding of their overall marketing performance.
How do other approaches stack up?
When we compare both MTA and MMM to traditional attribution models like first-touch or last-click, the problem becomes apparent – they only give credit to one touch point.
First-touch only considers the initial interaction a customer has with a brand, while last-click attributes all credit to the final touch point before conversion.
This simplistic approach fails to capture the complexity of today's customer journeys and often leads to overvaluing or undervaluing certain touch points.
Moreover, traditional attribution models do not take into account external factors like seasonality or offline marketing efforts, making them even less accurate.
While these may be somewhat appropriate for small startups with simple marketing strategies, they are not suitable for companies with diverse and complex customer journeys.
What’s the most accurate approach?
Therefore, the most accurate attribution model for marketing is a combination of MTA and MMM, a thoroughly data-driven approach.
By using machine learning and advanced analytics, this model takes into account all touch points and external factors to provide a more accurate understanding of marketing performance.
Our attribution solution is based on this approach, allowing businesses to accurately measure their marketing efforts and make data-driven decisions for better results.
Julian Berry and Jo Young from UniFida have recently written this authoritative 48-page guide to Marketing Attribution. To get your copy click the link below.
Comments