After discussing the accuracy of your data collection methods in part 2 of this series, we learnt that it is pretty much impossible to track every single interaction your client has with your brand. This is a big deal, as the accuracy of your data has a direct impact on your attribution – and it gets more complicated, because accuracy isn’t the only factor.

In this article, we investigate the relevance of the data collected for attribution – including identifying relevant data, understanding its relevance and knowing what to do with it.

 

Single Activity Customer Journeys

Today, many marketing managers view attribution almost in a religious light. Typically, the greater the responsibility they have over budget, the greater is their desire for assurance – assurance that decisions they have taken were, in fact, the right ones. Also, since CMOs in today’s digital age are more IT-savvy to make use of technology to assist them, they are also more specific in their needs when it comes to working with technical solutions. Dynamically forward-looking multi-attribution, for example. Or self-learning game theory algorithm in real-time.

But many forget the basics. Before going into advanced attribution, it is also necessary to understand basic attribution. How do your attribution rules work? What proportion of your customer journeys can be applied to attribution rules? Or rather, how many percent of all customer journeys included have more than one marketing channel?

Why is it important to know how many Single Activity Customer Journeys you have?

Let’s say your reports shows that 30% of your customer journeys only experience one touchpoint. In other words, 30% is made up of Single Activity Customer Journeys. This means you don’t need a smart marketing technology to apply any form of attribution to award the winner for these 30%. Therefore, 30% of the output and sales are irrelevant for your attribution analysis.

This indicator should be known by every marketing manager. After all, the result of your attribution is only as good as the accuracy and relevance of your data combined. Accuracy here refers to the technically achievable coverage of all recorded transactions through marketing technology, whereas the relevance of your data refers to the percentage of total marketing that is covered by your attribution rules.

So if your attribution covers 70% of your marketing activities (because the other 30% is made up of Single Activity Customer Journeys, for instance) and your technology is capable of recording 85% of all your sales, then your data has:

70% relevance * 85% accuracy = 60% meaningfulness

This would mean that your attribution only covers 60% of your marketing activities.

What should you do with your Single Activity Customer Journeys?

If you do have a relatively large number of Single Activity Customer Journeys, your life isn’t over. There are many further analysis possibilities. We recommend the following:

  • Examine the percentage of Single Activity Customer Journeys for new customers and existing customers
    It is a plausible assumption that repeat customers buy through very similar ways, while the path for the first purchase from new customers is more chaotic. From our experience, it can be very different. There is also a high chance that new customers end up purchasing with just one touchpoint because of traffic sources like Google Shopping, and repeat clients may need reminders elsewhere to purchase again.
  • Identify how much of your paid channels, own channels, and earned channels are Single Activity Customer Journeys, and go deeper if you need to
    Marketing channels can be tricky to analyse, as they can contain hundreds of activities and advertising partners. If the affiliate channel has many Single Activity Customer Journeys, it cannot be concluded that attribution becomes irrelevant for affiliate as a whole. For such cases, it is crucial to dive deeper (category, subcategory, advertising partner) and study patterns from those customer journeys.
  • Evaluate the ability of your technology
    A great source of error is, of course, the tracking ability of touchpoints that led to conversions, as well as the technical ability of stringing together touchpoints to form customer journeys.

 

Reaching out to customers

One could imagine that repeat customers would have simpler purchase journeys through their own experiences, while new customers have more chaotic journeys. However, the analysis we conducted from our clients’ data proved to be different. This could be because as habitual creatures, customers enjoy reaching a destination on the same route that has been tried and tested. For as long as the journey had been pleasant, many would be happy to do the same thing over and over again.

Whether it is the former or the latter, the technology you have in place should be able to give you these insights – your proportion of single activity customer journeys, the patterns in customer journeys for recurrent customers, as well as those for new customers. With that, you are better equipped to adjust your strategy to reach out to more customers.

Stay tuned for the next instalment as we explain why the AIDA marketing model is rubbish, as understanding why will improve the accuracy of your attribution!

 

Here is part 3 of the masterclass by COO Robert Schneider. The masterclass was held in German.

Catch up on the rest of the Attribution in E-Commerce Series: Part 1 | Part 2 | Part 4 | Part 5