I have to disagree a little. You shouldn’t start anything with page level analysis. You should start with a business objective. I know that this sounds obvious, but how many of your teams really start here, really? If for example, let’s say, Zappos wants to sell Nike better on this page (they could be great at it already, I don’t know, this is just an example):

http://zeta.zappos.com/brand/111/Nike

The starting point then would be to ask what can we do here to make people more likely to buy Nike products? Then you would need to meet with marketing, design and user experience people to develop ideas which you will then test to see which one creates the most lift and converts visitors into Nike product customers.

Obviously, having an understanding of how visitors click and interact with the page is important, but the entire process of coming up with your business objective and testing never really required giving an arbitrary dollar amount to a page before you started testing.

Instead of giving a dollar amount to a page, which is very easy with about any analytics solution, ask yourself this one question:

How effective is the “X” page at actually selling “X?”

Your analytics solutions will probably have a more difficult time answering this very basic question. In fact you might need to do some kind of segmentation to find a conversion for product “X” where visits contain a view of the “X” product page.

Refer to my last post (http://www.jasonegan.net/2008/09/09/more-on-pages-not-being-worth-anything/) for a little more clarification on what I meant. Basically, I am saying that we are monetizing the wrong thing. We need to be monetizing our marketing efforts through testing and optimization.

I’ll admit the title was part truth and part bait . The point that I’m trying to make is that the reason things like dashboards and regularly delivered reports are at some point in their lives continually and ALWAYS ignored is because things rarely change. For example, think about revenue trended for an entire site over time.

While revenue over time is an important number, worrying about why it changes from day to day or week to week is somewhat wasted effort if you’re not really doing anything to make that number change. So revenue was up 10% from yesterday. Is that really information that you can leverage for anything if nothing on your site or any marketing efforts changed? In this case it falls in that useless category of “good to know” information. How much of our analytics solutions is really just “good to know” information. Do we really need to track everything on the planet if we can only really change a few things?

This is exactly the reason dashboards are ignored and rarely used by business owners and decision makers. A number just moving over time is easily ignored, unless there is something going on, giving context to the information. If you are constantly testing and optimizing your site, then the numbers mean a LOT more on a daily basis.

I’m quite agree with your main idea,that we should keep optimizing pages even though they “do not” make any revenue(may be I’m wrong). You know, when saying this, I feel it’s quite funny to make such a “assertion” because it’s just like saying a shopping mall doesn’t make revenue for a e-commerce site, or a exposition doesn’t make money for a content site.

Finally, may be you try to say marketing doesn’t worth anything but the improvement of marketing make revenue- if we take all things before “payment” as marketing. Still feel funny about your title, but no evil.