Conversion attribution needs to be multidimensional to track today's complex customer

Today's digital consumer is all over the place-busy, curious, distracted and distractible. In any given day, your customer may roam between multiple screens-tablet, smartphone, PC, and TV-in the work day and during leisure time. He or she will chat on instant messaging, Tweet, post status updates on Facebook, bank, shop, conduct searches, watch videos, and more across these different devices.

And at several of these points over the course of a day or even a couple of weeks, the customer might encounter your marketing message in a banner ad, in social media, through a search engine result or paid search ad, and more. A conversion might not be the result of a single interaction or exposure, but the product of a combination of encounters the customer had with your brand across a range of touch points.

For example, our data shows that 30% of search campaign conversions are assisted by media, while 70 percent of media campaign conversions are supported by additional media exposures. What's more, not all conversions happen on the same day, with our data showing that many customers might convert 14 days or more after their first exposure.

The picture becomes even more complicated when you're trying to track end-users as they move off the work PCs they use during office hours onto the smartphones and tablet computers they tend to use at home in the evenings and on weekends. We observe vastly different behaviours on tablets and traditional PCs, as well as during different times of the day. For example, response rates on tablet computers are higher than they are on PCs.

The long and winding road

That means that the pathway to conversion is no longer simple and linear-if indeed it ever was. For that reason, marketers should start using models that go beyond last-click attribution, where the customer's last exposure to the brand is given complete credit for the conversion.

Instead, you need to look at the data you have from a range of different channels and understand it in context. You should be visualising the customer journey across multiple timelines, and seek to understand the role that different end-user devices, touch points, times of the day and user contexts have on your campaigns.

Armed with the right data, you can start to go beyond tracking performance of individual channels towards optimising each of them to aid the customer's journey to conversion. Rather than understanding each channel in isolation, you can begin to see them as part of a complex machine where a change to one part can influence the performance of another.

For example, you might come to understand that media campaigns have a dramatic impact on search conversions and choose to keep your display spending steady rather than slashing it, even if it is not delivering the expected conversion rate in isolation. Or you might look at optimising the conversion process for mobile devices once you see just how many of your leads or new customers arrive via mobile search.

Readying the right infrastructure

Of course, tracking the performance of multiple channels in an integrated manner demands some hard work. It does mean that you need to invest in creating a single tracking platform that provides a single view of as many of your digital channels as possible.

Furthermore, your infrastructure should allow you to track users as they move from your Web site through the conversion funnel. You should establish clear, measurable KPIs, such as ecommerce sales, lead generation, and so on, so that you can map your digital marketing performance against business objectives.

This sort of infrastructure will equip you to do multichannel reporting and analysis that enables you to compare the performance of your various touchpoints and channels against one another, and to also start seeing how they impact on each other's performance. Armed with this rich data and the insights it yields, you can roll out a range of advanced digital marketing strategies.

Some examples include:

• Advanced targeting of ads, for example targeting ads by the time of day, behavioural data, the end-user's device or geographic location.
• Remarketing, using behavioural data to target users with more relevant messages that help move them down the conversion funnel.

These strategies can help boost customer conversions dramatically, translating into a better return on investment on your marketing and advertising budget across your digital channels.