What is multi-touch attribution?

Multi-touch attribution is a way to allocate credit to each marketing touch point that preceded a conversion within a customer journey. It provides a quantifiable evaluation of your marketing efforts and their direct and traceable effects on revenue, allowing marketers to report on activity and allocate budgets effectively.  

With so many choices, channels, and types of customers, it can be hard to know exactly where or how to focus your efforts. Or even where to start. 

So, in this guide we’d like to offer our evaluation of the types of MTA models available and reasoning as to how an effective solution can transform your life as a marketer. Read on.  

MTA Models

As with anything in marketing, there are a plethora options to choose from. Put simply, the difference between those available is the percentage of a sale they credit to varying steps within the customer journey. 

The Linear model for example, recognises that every touchpoint is equal. Whilst Time-decay credits more value to the interactions closer to conversion. On the other hand, both U- and W-shaped are position-based. U accredits the beginning and end more so (picture the shape of a U) whereas W recognises the first, middle and last stage as the most important – again the clue is in the name. 

The more advanced options are custom MTA models. For example, Cubed uses a proprietary (custom) MTA model built upon propensity modelling. This is based on a neural network (advanced machine learning model) which is constantly training, learning which user behaviours makes a person more likely to buy.  

Cubed gathers a robust amount of data about your customers – which pages a user visits, their previous purchasing history, their previous visits, how long they’ve been a customer, lag since last visit, duration on site, events such as checkout started to name a few. All of these data points about a user are then passed to the Cubed model which makes an informed prediction about the influence of each touchpoint. This fact alone makes this approach far superior to others on the market. 

MTA Measurement

Here’s an example to help illustrate how this type of MTA measurement works. 

The Cubed algorithm works on every visit it sees, providing a propensity score for the visit once it has been completed (as well as real time as content is viewed, but that’s less relevant here!). The Cubed MTA model then uses these propensity scores to give credit to the campaigns, keywords, channels and so on that drove that visit to site. As such, Cubed gives more credit to those channels which drove the biggest changes in a visitor’s willingness to buy, taking the concern away from the ‘converting’ visits and instead understanding the important touchpoints across the entire visitor lifecycle.  

Cubed’s advanced machine learning model considers all channels and the logic of the user journey. Generally, Cubed takes credit away from bottom of the funnel channels such as Direct, as typically users have already reached a significantly higher propensity by the time they are ready to purchase and hence incrementally, have much less of the credit. Whereas SEO and non-branded PPC generally sit at the top of the funnel and with Cubed they are accredited significant percentages of credit. This  means marketers efforts are fairly recognised for the work they do.  

With the adoption of Cubed, the challenge of knowing which channels work for your brand is eliminated. And suddenly, accurate reporting, forecasting and budgeting is possible.  

Now you need to make the case internally. 

Why does my organisation need MTA?

Give Credit

You have channels and content that are converting consumers – but what about the efforts that bring users to your website during acquisition phase? It’s good to think of your marketing efforts as a team sport, and conversions as the goal. It’s unlikely you’re going to score many goals with just a striker. You need your defense, midfielders and goalkeepers. Not every player is scoring goals, but that doesn’t mean they aren’t contributing in important ways. 

Reduce Wastage

“Half the money I spend on advertising is wasted; the trouble is I don’t know which half.” 

How many marketers have the capabilities to answer this conundrum with confidence? Multi-touch attribution helps you put a finger on what’s working, and just as important, what isn’t. Know with absolute certainty where you’re wasting time and money.  

Master Micro Conversions

This goes hand in hand with the idea of giving credit where it’s due. In today’s internet-driven world, marketers and need to keep in mind that all customers and prospects have tonnes of info at their fingertips; they’re doing their research on your product well before they interact directly with a sales rep, let alone buy. 

With that in mind, remember that your website visitors are coming for a number of reasons, at all stages of the buying journey. Your goals for this wide cross section of visitors need to be customised—you need to have micro-conversions in place. 

Multi-touch attribution models can help you track these micro-conversions and understand how they contribute to standard (or macro) conversions. Even better, you’ll learn a ton about your prospects, how they behave, and what they really want in the process. 

Capitalise on Data

Take a minute and think about all of the analytics platforms at your fingertips. Chances are you have a Google Analytics account, various social media accounts all with their own in-platform analytics, analytics attached to your email automation provider, paid advertising analytics, SEO numbers, some flavor-of-the-month free thing that was going to give you unprecedented marketing insights—you get our point. Marketers are far from short on data. The key is putting it to use. Multi-touch attribution provides a framework to help collate that data in a single platform of truth and actually make it meaningful. 

Make Your Case

Sure, you know that a cool new optimisation tool or increased budget for a content writer will be worth it for the overall success of your marketing efforts – but does your boss? With solid data to justify decisions, the entire marketing department can be confident in the strategic moves you make. 

Marketing attribution is essential for ensuring that your marketing efforts pay off. Employing software that can automate and systematise marketing attribution is critical for making the process reliable and efficient. 

Recognising this, our team developed Cubed.

Frustrated by the lack of robust MTA models and tools available for our clients, we decided to fill the void using enhanced machine learning algorithms. Cubed can connect the entire customer journey, both online and offline, and provides marketers with organised and genuinely actionable data. The brands who are currently using it? They’re performing miles ahead in their markets.