Web analytics as a clickstream focused discipline has been largely dead in the water for most people who are active in the area. As the continual march toward customer centricity gains momentum, we're seeing more and more of the big players pushing the concept.
Such was the content of the presentation by Greg Drew of WebTrends (a sponsor of the eMetrics Summit). Greg was focusing on the idea of moving marketers away from the 4 p's of marketing, which are product, place, price and promotion. Instead, Greg talked about the idea of engagement, something that requires a relationship with your customer. He posed the question, "How do you measure, foster and perpetuate that relationship?" His answer is the 4 R's of marketing:
- Reveal – understand your website visitors most current interests and identify common traits to build focused marketing segments
- Reward – reward behaviors with offers and special opportunities to further engage and delight. This falls under the umbrella of transferring the ownership of the brand form the company to the people. His example was Kettle chips customers getting to pick the upcoming flavor (think also of voting for M&M colors and so on.)
- Respect – highly relevant, targeted offers result in increased trust and respect of the customer toward the marketers brand. It's the concept of aligning your business with your audience's thoughts and aspirations
- Retain – stimulate interactions with the products through fun and imaginative extensions of the concept (ex: kettle-ictionary, potato gallery)
Being a vendor, he had to flog his latest product, WebTrends Score. It works by using interactions with the site to append an element to a person (cookie?) profile. Different values are assigned for different interactions and you can create different scoring groups. WebTrends can then pass this value to other elements of your marketing system (You like safety information about SUVs on the site? We have your mailing address? Great, here's a direct mail piece about safety). It sounds great in theory, but I'm a little skeptical about how it works in practice. Do you have any experience with ML2?