Posts Tagged ‘analytics’

Best Practices and SEO Don’t Mix Too Well!

Friday, January 22nd, 2010

I’m pretty sure I’ve used the phrase “best practices” here on the search engine optimization e-blog a time or two and now feel like kicking myself for doing so.

Intuitively, best practices don’t make any sense in the SEO profession.

Best practices grew out of organizational management techniques that became popular in the mid-to-late ‘90s. The idea was that certain groups who achieve “best practices” could then teach those practices to the entire organization.

From a practical point of view that makes sense…but in order for “best practices” to work in a strategic sense, the environment must be relatively static and homogenous…two characteristics online marketing & SEO don’t have at all.

Something that works today may be obsolete by the middle of next week in this business.

Instead of working under a static rule set that’s worked in the past for something (old) and has been popularized (is average), which is what best practices essentially are, use metrics to experiment with what works and what doesn’t.

It’s appealing to think there’s a magic formula of best practices you can employ and then just sit back and watch the sales come in.

But the Internet, you and your competitors are constantly changing.

In the end, “best practices” can offer the value of a good starting point to begin experimentation but nothing more. Establishing a good feedback loop and basing your decisions on results of your experimentation means you will go much farther than just re-hashing someone else’s “best practices” plan.

Focus on your business and be ready to constantly modify what works and what doesn’t.

Google Announces New Page Speed Report in Webmaster Tools

Friday, December 4th, 2009

On the heels of a recent announcement from Google explaining their newest addition to the search algorithm, the search engine giant has now added a site speed feature to Webmaster Tools.

You can now access the new feature through the “Site performance” reports under the “Labs” section of the tool. It basically shows how fast your site loads in comparison to other sites and also includes a nice graph that shows your site’s speed over time.

It’s a small step in their larger effort to make the web faster for their users. Studies show that speeding up a site leads to higher user retention and activity, along with lowering costs and increasing revenue.

In addition to general site speed, Google also details specific pages and their load time. They give tips on how to improve speed on a per-page basis.

This new tool can be very helpful to ensuring your pages load as quickly as possible. While we don’t know how much significance will be placed on the new ranking factor, you can be sure that having fast loading web pages will be beneficial to your site. However, since they are adding this new feature to Webmaster Tools, it may be safe to assume that page speed will become an important ranking factor.

See a screen shot of one the site performance reports below.

You Can’t Get Universal Truth from a Single Landing Page Test

Friday, July 10th, 2009

It’s sometimes hard to resist the temptation and good feelings that come from when you discover something that works tremendously well for a landing page. It’s easy to think it could be applied to every website you ever work on.

But the truth is, the Internet and by extension the world is a very complex place. So mass generalizing the latest popular search engine optimization technique is a bad idea. You simply cannot derive universal truths from a single landing page’s test results.

Any “learnings” from a single landing page are likely to be uncertain at best.

Why? It’s because of what science dubs as “false causality”, meaning “correlation does not imply causation.”

This phrase is referring to a common error people make, no matter if they’re thinking about a science experiment or a webpage. That is, they assume that since two effects are related or occur together, one causes the other. However, this assumption fails to recognize the fact there could be a third variable, or confounding variable, that causes the other two.

In landing page optimization, it’s dangerous to rationalize results after the test. Rather than focusing on elements of design that had nothing to do with improving performance, try to come up with diverse and interesting ideas for each landing page test you do. Be forward thinking and do not simply accept the most apparent analytic answer.

Read in more detail this concept along with some real world examples from this Search Engine Watch article from Tim Ash.

Why You Might Want to Check Out the Best of Web Analytics Guide

Friday, October 28th, 2005

An email from Web Trends today announced that they’ve compiled 12 of their best articles in a Best Of guide.

I was interested, since analytics are key to measuring success and refining your online marketing efforts, and those who specialize in analytics can clearly see results across the varied industries of their clients – surely they must have some worthwhile advice.

For example, 5 key areas the Best of Web Analytics Guide include how to:

  • Breakdown a Conversion Funnel
  • Calculate your Online Conversion Rates
  • Define Key Performance Indicators
  • Audit Before You Redesign
  • Create Landing Pages that Convert
  • You’ll Have to Sign In – But Consider It An Opportunity
    You’ll have to either sign in or sign up with yor contact information in order to get to the page where you can download the guide. Which means this guide is a lead generation vehicle for Web Analytics.

    This should be of interest to you as a marketer, even if you’re not considering using Web Trends. Although a nuisance for many who just want the free content, getting on the “lead lists” for companies like this gives you a firsthand look at how competitors or operators in certain industries carry out their campaigns.

    I may get a follow up call or email. And you can bet that I love to collect and analyze those, from all the companies in our industry. Designers, take note of layouts and typesetting methods used. Copywriters, see the approach the copy takes. You’ll notice trends and be able to spot the ones that are working and used more often.

    Go here to see the landing page from the promotional email: Best of Web Analytics Guide