Achieving a data-driven attribution model: the three ‘must-knows’
With the increasing complexity of the customer journey, traditional single-touch attribution models are no longer sufficient. In this article, I'll look at the 'must-knows' on how to achieve a data-driven attribution model and tailor it to the needs of your brand.
Businesses know they need to focus on attribution. But as the customer journey gets more and more complex, it’s harder and harder to know what that focus should look like.
In a recent webinar on mastering the art of data-driven attribution, held by ClickZ in partnership with Fospha, close to 100 marketers were polled on their use of various attribution models. 49% of respondents were still using a last-click approach to attribution, while the next most popular model was first-click, represented by 13% of those taking part.
A single-touch attribution model is better than nothing. But single-touch models won’t show you relative impact across different channels, and without that, you can’t get an accurate measure of which channels contribute to which conversions. Acquisition costs are rising, and now is the time to focus on getting an accurate, bespoke, data-driven model in place for your business.
So how do you move on?
In the first two parts of this series, we looked at the challenges of multi-channel attribution and the barriers to successful attribution. Here, we’ll look at the three ‘must-knows’ to achieve a data-driven attribution model and tailor it to the needs of your brand.
Content produced in association with Fospha.
Take that first step
The complexity of the customer journey means that many marketers simply don’t know how to achieve data-driven attribution. Either they settle for an off-the-shelf attribution model, or they de-prioritise attribution because they won’t be able to realise any revenue benefit ‘in-year’.
Our webinar poll data showed that simple models like last and first-click attribution are certainly better than nothing. If you’ve sidelined attribution until now, start simple and aim to build sophistication incrementally. Having an attribution model in place is the first step towards effective data-driven marketing.
Bring it all together
To build a tailor-made attribution model that’s fit for your brand, you need the right data. Simple models are a good start, but they will always fall short. Assigning all the credit to the first or last click means you don’t know how other touchpoints have contributed to the customer journey. Unless you understand multi-touch, cross-channel behaviour, you can’t unlock the full potential of your marketing spend.
A Customer Data Platform (CDP) breaks down silos and unifies data from a range of channels and devices. CDPs are fast becoming the go-to standard for effective data management. So what does a CDP do for you? Gartner’s defines a CDP as “An integrated customer database managed by marketers that unifies a company’s customer data from marketing, sales and service channels”. With unified data from a CDP, you can drive conversion, increase lifetime value and manage cost versus revenue.
Rinse and repeat
Attribution will help your business grow. And as it grows, your attribution model has to evolve. Attribution is not a one-time activity.
Better data means better attribution. A data-driven approach also allows you to carry out predictive modeling, so that your brand can experiment quickly without real-life impacts.
Marketing-channel attribution (MCA) may seem complicated, but that’s because the customer journey is complicated. Your customer is living a ‘nonline’ life – non-linear, online, offline and over multiple devices – and that means you should be living one too. Yes, it’s complicated, but you don’t have to do it alone. The impact of not getting it right is too big to ignore, and that means there’s never been a better time to move to data-driven attribution.
This is Part 3 in a series of three articles about how data-driven attribution can work for your marketing campaigns.