The way we purchase products and services has evolved dramatically in recent years. We are now faced with customer journeys that involve many different paths to purchase – and it’s been a challenge to keep up.
There are multiple marketing measurement and attribution tools that allow you to understand this constantly evolving landscape. Ultimately it’s about using technology to allow you to make better business and marketing decisions.
In this post I’m going to explain what attribution is, and look at one element we decipher to deliver results – correlation vs causation.
What is Attribution?
Attribution, in a nutshell, is the method of sharing sale value between the marketing activity, or touchpoints, that drove it.
Before a person purchases a product or service from you, they are often exposed to a multitude of marketing touchpoints. These cover a wide array of online channels, such as PPC and affiliates, and offline channels, such as direct mail and TV ads. Attribution assigns credit from the final sale to the marketing touchpoints that contributed to it.
What I want to focus on today is a challenge that we solve with our attribution modelling techniques – understanding the difference between causation and correlation. This is a common challenge for marketers I speak to today.
What is the difference between causation and correlation?
To illustrate the idea, let’s look at firemen and fires. Data shows that the more firemen that attend a fire, the more damage the fire causes. Statistically, this is accurate, but logically it is flawed.
If we decided to follow the statistics without applying a logical view, we would likely decide that the best course of action would be to reduce the amount of firefighters attending emergencies, which would be a big mistake.
Without looking at any other factors, we see that more firemen equates to more damage. Drilling deeper, and incorporating other factors, shows it is not the firemen causing the damage, it’s simply because the fire is bigger and more firemen are needed to fight it.
If we mistake correlation for causation, this may lead to us making incorrect decisions at the next fire we are called to.
Why does this matter to marketing?
If you know what touchpoints influenced a user, and how much they affected their decisions for each and every one of your sales – then you’ve found the marketers’ Holy Grail.
Let’s consider this scenario: you see a user on your site who abandons their basket, then returns ten minutes later and converts, after you’ve served them a retargeting ad. You could argue that person was always going to buy. But was the retargeting ad the catalyst that caused the user to come back and convert, or was it simply a happy coincidence?
It’s crucial to understand the relationships between your channels and know how different actions and touchpoints impact the path to purchase. Was the increase in conversion rate because of a new feature on your website, or was it the new content affiliate you’re working with? If you’re mistaking correlation for causation then you could be using your precious marketing budget on advertising that isn’t driving any uplift.
Identifying the difference will enable you to spend your budget on the activity that has a direct impact on consumers. This understanding needs to be built into any attribution model to ensure credit is given to channels according to their real influence on consumers.
Reducing the complexity and pinpointing the insight
Attribution modelling coupled with advanced reporting and a solid understanding of how a technology is working to provide your reports is the key to keeping up in today’s modern marketplace.
My company regularly hosts webinars with leading brands and technology innovators, click here to register your interest for our next one “The future of retail and emerging technology.”
Read More: Why “Last in” Affiliate Attribution Needs a Second Look
Comments on this article are closed.