Analyzing Conversion Paths with Google Attribution

Earlier this month Google announced that the beta of their new analytics suite product, Google Attribution is being opened up to more advertisers following positive early results since its launch back in April.

The newest addition to the G-Suite boasts some exciting new features – TV Attribution pairs ad spot data from broadcasters with relevant search query data to identify website traffic spikes and conversions attributable to specific broadcast ad placements. Cross-Device Tracking pieces together individual user’s unique interactions with a brand across different devices to map out the customer journey in unprecedented detail.

The most exciting aspect of Google Attribution is its use of machine learning to calculate the contribution of each component of a brand’s marketing mix to conversions. While multi-event attribution models aren’t a new concept, researching, implementing, testing and evaluating the success of more complex attribution models has proven too time or labour intensive for most marketers to adopt.

For this reason, the more rigid first-click, last-click or even-distribution attribution models are most commonly used, which help generate a level of insight into customer behaviour, but doesn’t provide the granularity necessary to quickly and confidently optimize media spend for maximum ROI.

The incorporation of machine learning to analyze conversion paths for unique business objectives and identify what works and what doesn’t will allow marketers to understand better than ever how to put the right message in front of the right person, on the right channel at the right time.