Use Case: Attribution modelling

, Posted by Adrian Kingwell in Case Study, Emily's posts

The challenge 

As a commercial manager, you need to understand how each channel contributes to the customer journey and its significance at every customer touchpoint. 

With this information, you want to have a true view of how channels all work together to drive customers towards conversion.  

As a result, you could then optimise marketing ROI across the mix of channels to achieve your sales targets quicker. 


What is the ideal solution? 

Through a combination of technology, statistical models and machine learning algorithms, you could understand at any point in time, how many customers are converting and how each marketing channel has contributed.  

The stuff that will get you really interested is if you can understand which customer segments are likely to convert in the future.  

You’ll be able to identify which customers are being reached by which marketing channels and whether they’re “desirable” or “non-desirable”.

With this information you can invest in marketing that targets the “desirable” segments. 

The icing on the cake would be to combine the view of desirable customers with the likelihood that they will convert.

This way you can target those desirable customers in the grey zone between those who will convert and those who are unlikely to convert with the right offers to encourage conversion.  


How will Mezzo Labs help? 

This is a common challenge among heads of media planning, digital and e-commerce and we have solved it many times. 

Firstly, we will hold workshops and interviews to explore and document your attribution needs.

We can use this information to audit your data collection and understand if your marketing channel data is being accurately collected.

We can then advise you on any fixes needed. 

Secondly, we will help you navigate the vast landscape of attribution tools by prioritising your needs and identifying a complementary technology solution. 

For example, simple attribution modelling can be carried out with the standard data and functionality available in Google Analytics and Adobe Analytics.  

Last interaction, first interaction, linear and time decay models are all available. 

More complex attribution modelling might require onboarding new technology and the use of machine-learning algorithms.  

CUBED, Syntasa and VisualIQ are all vendors who offer these solutions.  

We will work closely with you to make a commercially sensible decision, based on your detailed requirements, budget and the ease of implementation.  

Thirdly, depending on whether we’ve decided on a simpler attribution modelling approach or to use more complex technology, we will help you to set up the technology within your company’s IT security and compliance frameworks.  

Then, we’ll develop the best algorithms or calculations to identify touchpoints for each channel and their contribution to sales.  

We will produce a sustainable attribution model and where we can, train your staff to use it. 

Finally, we will make regular recommendations about marketing investment based on the attribution model’s data and our investigative analysis.  

We’ll also pinpoint the right customer segments to be pushed into your email, DMP or DSP systems for targeted advertising. 


What are the business benefits? 

The attribution model will allow you and your team to understand how your marketing channels interact and work together to drive customers’ conversion. 

Be ready for assumptions, (mis)perceptions and new insights to be surfaced.

These will stimulate discussions about how different touchpoints – and teams – could be credited for their sales contribution. 

Once these initial insights have been disseminated, staff can turn towards maximising the value of the advertising that is clearly working and focusing less on what isn’t working.

The insights and recommendations that Mezzo Labs will provide for you could drive in-channel and across channel tests.  

Perhaps the greatest benefit is that all stakeholders are aligned on a discussion about channel attribution and performance. 

Data transparency and a robust attribution model to monitor changes will drive smarter marketing investment.