Attribution Modelling

Where do my customers hear about my business?

This is the question everyone in marketing should be asking. If you know where your customers first hear about your product, you can maximise your exposure there and scale your business efficiently. Sounds simple, right?

Unfortunately, when it comes to measurement, the world of digital marketing is extremely complex.

Combine this with the fact that the industry is rapidly moving away from cookie based tracking on an individual level, it is now more important than ever to be thinking about how to develop your own method of attributing value to your marketing without having to rely on 3rd party tracking tools such as the Facebook Pixel and the Google Conversion Tag.

Below we explain why attribution is so important to the growth of your business for the long term and how we go about helping our clients answer some of the hardest questions in marketing, such as:

> How does my revenue scale in relation to my marketing spend?

> Where do my customers hear about me?

> Which of my marketing channels are most efficient?


The Customer Journey

Getting to grips with the complexities of online purchase behaviour is at the heart of starting to develop an attribution methodology for your business.

Every customer takes a different journey on their path to purchase. It can start with a conversation with a friend who has recently purchased your product, through a quick Google search, or through something they saw while browsing Instagram.

From that point onwards, they may visit your website multiple times, read reviews online (either on your website or elsewhere), speak to people they know, look at your competitors, making sure that they are fully informed before committing to spend their hard earned money with you.

Most eCommerce businesses advertise on both Facebook and Google, supplemented by other channels such as Organic Search (SEO), email (CRM), smaller social networks (Tik Tok, Pinterest, Snapchat), affiliates, programmatic, direct mail and even TV ads. With ad budget and resources being spent in so many different places, it’s very difficult to clearly attribute value to each of these different channels

Below we break down the process of how we have helped our clients to create an attribution framework that helps them to identify the value of their different marketing channels and evaluate how their business scales with increased marketing spend.


Our Attribution Methodology

We strive to build a realistic understanding of the value of different marketing channels by using and comparing multiple approaches to measurement. There is no silver bullet analysis method, but by using and comparing different approaches it is possible to build a solid understanding of how to drive growth.


> Statistical analysis of overall marketing ROAS

The first step is to evaluate how well the business as a whole is scaling with marketing spend, while accounting for variables such as demand, seasonality, discounting and promotions and your customer lifetime value.

This allows us to create a revenue prediction model that caters for the impact of the external factors that can have a significant impact on performance when looking at short term comparisons, such as week on week or month on month.

Many marketers cannot relate changes in ROAS to the changes in the amount they are spending on marketing. By building a revenue prediction model, we can take an objective view to performance which enables us to plan strategically and not make rash decisions based on short term declines or improvements in performance which are driven by seasonality or macro-economic trends.


> Click-based attribution models

Click based reporting in tools such as Google Analytics is very useful, but limited due to the last-click model that many of them use.

We can take your session data and create first click, last click and mixed models of your customer behaviour, we can find answers to specific questions about your marketing mix and where the value of each of your marketing channels.


> Volunteered Attribution

Click based models can be very useful, allowing you to be certain that a user has taken action after seeing one of your ads online, but it doesn’t capture all of the nuances of the higher funnel channels, such as Facebook, Youtube, Display, OOH and TV.

By surveying your customers at checkout you can add an extra layer to your attribution model, with your customers telling you where they first heard about you.

While this is relatively subjective, it can be a good indicator of where your ads are most memorable and is a valuable addition to your marketing data set.


> Experiments

With a solid knowledge of what to expect from past performance and a model which predicts what performance is likely to be in the future, you can carry out experiments to identify the uplift in performance from doing things differently.

This allows you to evaluate the performance of your marketing without having to rely on the measurement quality of your ad platforms.

After changing your strategy and giving it enough time to reach significance, are we now exceeding our existing model’s expectations of performance?


> Attribution frameworks

Building on top of a basic, click based attribution model, we can create a more advanced framework that gathers evidence from click based models, customer surveys and experiments we’ve run to build a list of assumptions that can be used to build a custom view of what value channels are driving. 

Comparing these custom frameworks to other methods allows us to constantly check the validity of our results and is a part of the ongoing, evolutionary nature of this work.


Get in touch

If you are looking to build a custom attribution model for your business or want to find out more about what we can offer, please get in touch via our contact form and we will get back to you to book in a call with one of our founders and our Data and Analytics Lead.

Once we have a clear understanding of your needs, we can propose a solution which will be able to answer the main questions you have around your marketing mix and the scalability of your marketing channels.