Segmentation is dead, the interest graph killed it.

This post was originally published on socialmediaclub.org for the SMC Editorial Team.

During a recent discussion at work (Sailthru) my colleague Amy declared “segmentation is dead”. It’s something that struck me as a really important statement because I think she’s right, and it’s the interest graph that killed it.

We’re living in an age where every click we make, every site we visit can be tracked and often personalized. Our relationships, clicks and likes are creating an individual map of our interests that can be leveraged to deliver us a highly targeted stream of recommendations. You have become a “segment of one”.

Traditional segmentation allocates a customer to one market segment based on characteristics such as gender, interests, location, income etc. Often employing continuous refinements to group people into smaller and smaller groups so that they can be targeted and moved through a sales funnel that is wide at the top and narrows toward a conversion goal. But, what if we’re looking at this process backwards?

A segment of one

All the data that is accessible to us via tracking tools provides us with a way to dynamically target our messaging in a way that skips the funnel altogether. When browsing Amazon I get recommendations on what books and products to buy based on previous purchases, browsing history and a collection of other data.

Amazon doesn’t segment, it use algorithms to put products in front of me they know I’ll be interested in and more likely to buy. I’d bet money that a marketer in Seattle isn’t creating customer segments that I neatly fit, they’re using real data collected from clicks and purchases and algorithms to present information to me.

An individual customer that receives content or product recommendations specifically tailored to their interests that’s highly likely to incite an action is a valuable asset.

Amazon’s not the only company doing this, publishers and e-commerce providers around the web are employing techniques to market directly to the individual. It simply makes sense to target a “segment of one” ME.

We’re in a time where an e-commerce store can literally present a product to me across any number of devices and know there’s a high likelihood of a purchase. Publishers can present a piece of content targeted to me knowing I am likely to read it.

Where it gets really interesting is when we add my interest graph and social graph to the equation. The similarities and differences in my connections start to create patterns that are highly charged with actionable data. When I share content that has been targeted specifically to me via a Facebook like, or re-pin it to Pinterest or even share it via email with a friend I am implicitly recommending it to people who will likely share common interests or relationships. (it’s important to note that an implicit recommendation is not always the reason for a share)

Those actions are repeatable and if the interest in the content shared matches he interests in my network it will gradually reach a wider and wider audience until it naturally fizzles out due to lack of engagement. It’s the reverse of the traditional segmentation model. We start at an individual and work our way up to a larger group.

If you can dynamically target your messaging and content on an individual level it’s simply more valuable. Amazon knows this and I suspect you do too. The data is there it’s up to you to leverage it for the benefit of your customers and your business.

As my boss Michael always says “When you look at snow it’s a mass of white, but when you examine it in detail each snowflake is unique”. That’s how you should treat your customers they are an individual with a unique interest graph…they are a segment of one.