“The real power of this data is bringing it all together so you can start looking at campaign assessment in totality.”
In multichannel consumer product marketing, you need to be able to look at all the data inputs in an integrated way. What you’re trying to do is to get to the core of what makes something work, or what is not working. It’s beyond just whether something is meeting expectations. You need to know why it is or is not. To understand this, you get clues from a variety of data sources. You can look at how people
respond to one piece, or how one thing changes behavior, but you have to see this across channels, and it has to include less direct data inputs like social listening. This kind of analysis requires integrating a lot of data from different sources, and it is an iterative process. It is a discovery process.
During the planning phase, you are defining key metrics. Then during the create phase, you are using consumer feedback metrics and your knowledge base to guide creative. After a piece airs, you are tracking to see if it has the expected impact. You are using metrics to close the loop on the content cycle, and this is where you can piece together the elements of a multichannel campaign. If you are not tying these things together, you end up with isolated behavioral data, or claim data, or survey responses, or inmarket metrics. The real power of this data is bringing it all together so you can start looking at campaign assessment in totality. You need to look at campaign assessment across an integration of variables.
The goal is to have this integrated perspective and analysis for action. You need to look at how creative was received, what happened to sales, and what happened to brand metrics. You need to look at hard and soft metrics together. It’s a big data challenge that requires examining various
data sources and applying advanced technologies that include artificial intelligence and machine learning. In the end, though, it is an analysis for what you’re going to do differently, or similarly, the next time. There’s no point in having a metric if you’re not going to do anything with it.