Assortment Analysis: Using Customer Avatars

Merchandise appeals to people. While this is an obvious statement, applying it as the fundamental basis for creating focused assortments is not so obvious.

Some years ago, it became apparent to me that one of the most effective ways to create distinct focus within an assortment is to create "avatars" of segments of consumers. These avatars are complete descriptions of a consumer, designed to be representative of a larger group. The degree to which your merchants can describe these "avatars" is the degree to which the assortment can be focused on meeting their needs. Even if your technology doesn't seem to support creating merchandise segments based on lifestyle attributes, there may be ways to "fool" it into doing so. First, let's look at what an Avatar is.

An Avatar is a description of not only what a person looks like, but how they think, what they do, how they react, and where they go. A complete description includes what they do for fun and why. How they prioritize their disposable income. What type of home they buy, and why. What they eat, where they go out to eat, how much time they spend with their kids......and on, and on. Why is this worthwhile? Because people buy merchandise and they do it for reasons. The more we, as merchants, can identify who we are selling to, the more likely we are to be able to select merchandise which will appeal to them.

Within the context of limited open to buy and a recessionary climate, creating a focused assortment is a necessity. Limited open to buy means greater stress in creating breadth and managing the depth of the assortment. One of the most powerful ways to understand appropriate depth and breadth is based on consumer Avatars. If each item in the assortment is classified via product attributes on the basis of an Avatar it is expected to appeal to, it becomes easy to determine if there is overlap, inconsistency, price point gaping or any of the other elements of a less than optimal assortment plan.

Data from current and past merchandise assortments can also be retroactively assigned the same attributes. While time consuming, the result is a revision to historical data which can then be used to allocate open to buy......even at the store level. Any segmentation strategy which groups stores into like clusters will, in my experience, be less valuable than grouping merchandise into like clusters, and analyzing the results at the store level.

If you aren't using Avatars, explore them.