In Analytics

In a previous blog post I described how to use Google Analytics to track down user experience problems, and argued that those problems, generally speaking, fall into one of two site optimization categories:

  • broader, site-wide design challenges.
  • single-page tactical issues.

In the case of site-wide issues, you’re probably due to look into a redesign or a information architecture reboot. But the majority of work that you can do on your site is usually the other kind: small, tactical improvements to individual pages on the site. Yet most site owners miss this opportunity, and basically leave their site unmaintained after an initial large redesign.

Why does this happen? It has a lot to do with people’s definition of “done”. TUG generally groups redesign activities into market research, content management, and page optimization. Marketing effort drops off as understanding and messaging for the market segments stabilizes, then content creation settles into a more stable pattern after a big burst following the marketing research effort. Once those two categories have been hammered out, it’s a lot easier to optimize the site pages, so that effort should go up until it’s a primary focus.


Most web owners don’t follow this cycle, but instead stop at about the point when all the content for the existing pages is “done”; that is, there are no more “blank spots” on the site. Often, at this point, maintenance activity on the site drops to almost zero. Drops to zero, at least, until some problem is identified that is too large to ignore, creating a frantic and often ad-hoc redesign cycle. This “fire drill” dynamic means that an organization with urgent digital needs doesn’t have a process or an infrastructure for dealing with the issues that have come up, increasing overall expenses and reducing the effectiveness of the effort.


What a missed opportunity! Even with the expansion of content into multiple technology contexts, page optimization in modern CMS tools is simple enough that a page doesn’t have to generate much at all in additional value to justify the effort to plan, build, and test it. The secret is to make the change from optimization as a fire drill to optimization as a practice.

Like most successful efforts, a site optimization practice starts with a goal, is fleshed out in a plan, and is then crystallized in a schedule.

In this case your goal should be to CREATE three landing pages using the schedule outlined below, with the expectation that ONE of the three performs well. This is new to you, so you may need a few tries to work the kinks out.

The plan should be structured with the following elements:

  • a justification based on expected effort.
  • a way to track and test your outcomes. Below is an example of an outcome matrix that tracks issues for optimization.


The main dependencies for the schedule are

  • how many person-hours you can devote to the effort per week
  • how long you need to measure the results of the test.

Overall, the schedule should be an iteration that tests any given hypothesis in sequence and allows enough time to assess the results and create a new hypothesis or choose a new page to optimize.

Outcomes can be tracked pretty easily through whatever project management techniques you have internally. The only recommendation is that you be sure to have a hypothesis and a well understood test that you design to prove the hypothesis. There are a large number of great articles about custom reports you can use to find pages to test, particularly landing pages.

Measurable outcomes can be pretty easy to prove statistically. I recommend either split testing, or, if you have a high-traffic site and a responsive IT team, The Google Experiments tool included in Google Analytics.

So make a commitment to optimize your site between redesign cycles! Develop a practice which will damp down costs and reduce the frantic effort of fixing urgent, major design issues.

Recent Posts