I had a chance to be part of a ClickTracks class a few weeks ago today called, “ClickTracks KISWA (Keep It Simple Web Analyst)” by Avinash Kaushik. Avinash compared the theme of this web analytic call to Occam’s razor: “All things being equal, the simplest solution tends to be the best one.” I think with all the options with ClickTracks and other tools out now, it is easy to get confused or over indulge in the numbers available to you. Sure, some of the stuff was pretty basic, but for those of us that can’t dive into the analytics daily/weekly, the highlights of the call can help you understand web analytics explained on a “macro” level.
Highlights of call:
Focus on Macro Analysis…See the forest for the trees.
- Visits.
a) How many visitors during given period, also called number of visits or number of sessions with website. Make sure it is accurate/correct.
b) Number of unique visitors during given period. Make sure it is accurate/correct.
c) Don’t dive into repeat visits, page views, conversion rates, and other metrics. - Visits from Where.
a) Understand referring URLs. Look for surprising URLs, friendly or unfriendly URLs. Understand short visits, bounce rates for certain URLs.
b) Inferring intent from search keywords. How from search engines, not from search engines. Understand short visits, bounce rates for certain keywords - What do you want from visitors
a) Why does your website exists?
b) What are top three web strategies?
c) What do you think should be happening on the website?
*Pick 3-5 key critical metrics only. Assign goals for those metrics.* - What are visitors doing
a) How did they get in? Top entry pages for each engine. Value of home page, first impression? Every page on your website is a actually a “Home Page.”
b) Top viewed pages. Is it the content we want them consuming or the content they actually want to consume? What content is valuable?
c. Site overlay report. Navigational elements: are they important/useful? Where people click on each page, CTR. Highly important. Also segment lick density. All traffic versus Google, for instance. Consider applying CT to intranet. - Segment reports
a) Funnel Reports: Website entry points–>CMTB (Convince Me To Buy) content–>Core Product content–>Cart–>Checkout. Each step can contain page or 100s in the funnel report. Don’t worry about path analysis.
b) Leverage long-tail report. Discover 1% cliff. Show business risk if top keyword rankings went away.
c) Why are your customers there? Web analytics tell you “what,” but not “why.” Use surveys on each page to explain the “why?”
Links from Avinash Kaushik:
Yahoo Web Analytics Discussion Group by Avinash Kaushik. I suggest everyone join it.
