Metrics Organization

Another great overview look at the metrics challenge under discussion arrived this week from Canada, where Ren&eacute van Diepen, from Phase 5, took the time to lay out another framework for approaching the problem. I’m a big fan of organization (mainly because I need to explain all these metrics to my clients), and my contribution would be as follows. In the Cutler & Sterne paper (which I think is great), I have taken these metrics further by placing them into various stages in the life cycle (reach, acquisition, conversion, retention, loyalty). Why? I find that I deal with clients who are at differing stages in their own lives, from emerging start-up focusing primarily on acquisition to established firm focusing primarily on retention. As a result, their CRM [customer relationship management] strategies are different, and therefore their performance measurement efforts vary.

In terms of actual metrics, I have added (of course, organized under headings):

Acquisition

  • Site exploration ratios (pages viewed/total pages available) calculated on a by-customer basis for the site as a whole, or more interestingly for sections of the site. This is a stickiness measure of sorts.
  • Most frequent abandonment location
  • Most frequently abandoned item

    Conversion

  • Session length by visit number (on the notion that this increases as one approaches a sale)

    Retention

  • Satisfaction and value ratings to survey responses Thanks, Ren&eacute, for the very helpful look at organizing the effort.

    None of these overviews totally solves the problem presented in the original request, but all the input helps others to get their arms around framing the problem and the components of a solution in a way that makes the ultimate effort of building solutions more complete.

    But what about the actual products that will allow you to track this sort of information on your site? And what will allow site owners to automate the process, rather than manually checking stats on hundreds or thousands of products to decide which to leave up, which to pull down, and which to change position?

    If you’ll recall the original query, our “client” here runs a specialty merchandise site with thousands of items, and the mechanical task of choosing which products to optimize and which to downplay is a major goal of the intended solution. Knowing what to measure, as pointed out above, is absolutely essential, but we also need to know how to make it happen effectively, efficiently, and, hopefully, effortlessly.

    We have not yet found a turnkey solution that does all of what is wanted here, but there are technologies that can help with significant aspects of the task at hand. The catch is in pulling them all together. Based on my own review of the materials piling up in my mailbox, that is no small challenge! Lots of vendors offer parts of the solution, but there is not yet a simple way to forge a total solution.

    I’m going to start looking at the parts that contribute to the whole solution, but at this point I have to warn you that there is no single technology that I’ve come across that does all of what our query requested.

    So, for those of you unwilling to wade through some confusion and piece together partial solutions to forge a more complete and customized fix, it seems to me that you’ll still have to wait. For the more technically daring among you, there are aspects of this challenge that can be addressed, and I’ll start on those in my next article.

    Related reading

    tencent_emily-ma_featured-image
    kenneth_ning_emarsys_featured-image
    bounce-370x229
    site search hp
  • <