An attribution model is a framework for analysing which touchpoints – or marketing channels – receive credit for a conversion.
The concept of an attribution model is not new. Marketing departments have been tracking and monitoring their advertising and promotions spend – trying to establish the lead source, an ROI or sales pipeline – for decades. What is new, is the use of AI and machine learning to deliver a more precise and reliable attribution model.
Whereas traditional attribution models rely on assumptions about customer behaviour, or use a more simplistic approach – machine learning attribution models continually improve and become more accurate as streaming data is collected and analysed to optimise business outcomes from an organisation’s marketing investment
When combined with more traditional marketing mix modelling (MMM), the attribution model gives a holistic view of customer behaviour. MMM is a top down approach providing high-level insights about what activities have the most impact and where marketing dollars could be invested at a macro level. Attribution modelling fills the gaps of MMM. It provides currency and granular insights – keeping pace with trends and changes as customer engagement shifts.
As part of a marketing and sales optimisation platform, an attribution model considers every possible action that can lead to an outcome – across every channel – and attributes each of these actions (if relevant) to each outcome (e.g. acquisition, engagement, retention).
Based on this information or the ‘outputs’, the optimisation platform may have an automated response or suggest a manual action.
An automated response may be to adjust promotional spend. For example, the platform automatically increases spend on Facebook video ads, targeting people aged 50-65 from 8:30-9:00am on Saturday as it has identified this leads to an increase in product enquiries.
Another automated action could be A/B content switching where – depending on the user’s demographics – the platform switches between different layouts on web pages to maximise leads. Or similarly, different prompts appear for call centre staff to use depending on the specific customer they’re interacting with.
The optimisation platform – and more specifically an attribution model – also provides insights for improving business strategy. It can inform customer journey maps and processes – giving organisations the ability to modify these to better align with their customers’ preferences – achieving improved business and customer outcomes.
These new machine learning driven attribution models are hugely beneficial to organisations – enabling them to maximise marketing spend, improve the customer experience and deliver high-quality sales leads. An attribution model delivering these improved business outcomes is clearly valuable to any business. In current times – with the impact of the pandemic on marketing and sales performance – such a model can be the difference between the success or failure of a project.
It is well understood that organisations are facing uncertainty due to COVID. The obvious being customers tightening their budgets, reducing spending and putting off purchase decisions. For marketing and sales departments, the impacts go well beyond this. Customers’ behaviours have also drastically changed.
Many no longer spend hours commuting – time previously spent scrolling socials. Others are at home isolating – using online activities to distract themselves. Events, workshops and information expos are only offered online with customers no longer needing to engage with organisations or their sales teams.
For organisations relying on traditional “rules based” attribution models, this could mean the information driving their current marketing and sales activities is no longer relevant. Organisations with a marketing and sales optimisation platform that delivers a data driven attribution model will be well placed to ensure their business continues to perform and thrive in any market.
With an attribution model that uses machine learning – where the model is continuously trained – organisations have the added assurance that their marketing investment decisions are based on current information – including the impacts of living through a pandemic.
Terry Donnelly leads the Data Services function at Novigi, and is based in the Sydney office.
For more information about anything you’ve read here, or if you have a more general inquiry, please contact us.