Be An SEO Scientist
Thursday, March 29th, 2007 at 12:13 pm by Jeffrey Perren, AC Magazine
Has your traffic or conversion ratio changed dramatically? Many would look carefully if one or the other dropped. But, if it rises sharply, you should be equally interested to investigate. You want to know what you did wrong or right, so you can determine what works or what doesn’t.
The first step is to see whether you dropped off the SERPS. If you’re not there, people can’t find you (that way at least).
Next, try to correlate changes in traffic or conversions with any changes made to the site. But keep in mind there’s a time lag involved. You may make a change that isn’t indexed by the search engines for days, or weeks, sometimes longer.
It’s always tough to determine cause and effect when they’re far separated in time. (Ever wonder how humans first figured out what causes pregnancy?) The way to overcome that problem is to keep good data. Keep a detailed change log, and graphs of traffic and conversions. Then you can look for patterns and shift one graph backwards or forwards in time.
(That’s how scientists attempt to identify causes of climate change, for example. It takes decades for temperature changes in the atmosphere to affect the water dozens of meters deep. The time lag gives them useful information — a so-called ‘memory’ of changes.)
The specific changes to look for are usually obvious. Have you started (or ended) a new email campaign that (you hope) drives visitors to the site? Have you changed the look or navigation? Have you changed hosting companies? (That might affect page load speeds or percentage of successful connects.)
Run sample search queries that common sense tells you potential customers seeking you out would. Examine the results. Could you find you? Once you went to the site, do you find what the query result would lead you to suspect? I.e. is what’s on the page relevant to what you were looking for?
Is it attractive and interesting? Would you buy from yourself? A well-known Hollywood casting director once asked an audience of acting hopefuls at a seminar: Would you pay $8 to see you on the screen? Be honest with yourself.
As stage two of your research, you can check more technical measures, like: referring keywords, entry pages, and index counts, etc.
Sometimes the answers will pop out right away; other times, it requires a bit of digging and creative analysis. But the cause-effect relationship is always there. Sometimes the answer is unpleasant. Your potential customers may not be reacting to the changes you made as you’d hoped. But satisfying the market is what selling is all about. Good news or bad, it’s always better to know why.
Permalink
Post your comments »
Filed under: Affiliate Marketing
