Measuring SEO Success

Posted by: william in SEO on Aug 25, 2009

Measuring the increased traffic and sales from a search-engine-optimization initiative is among the most challenging issues site owners face. Frequently, when I’m starting a new SEO program with a client, the seemingly simple question of “What’s your goal?” is met with a pause. It’s not that these clients are SEO neophytes or that the companies aren’t disciplined and committed to their online presence. It’s that determining cause and effect is confusing work in the fuzzy area that is SEO. If you don’t know how to measure where you are today for SEO, it’s pretty tricky to define a measurable, realistic goal for the future.

Make SEO Goals Specific, Measurable and Timely

The goal cannot be simply "Increase search traffic" or “Rank better for footwear.” There is no way to succeed at these goals, or to fail. Instead, one should ask the following.

• Increase search traffic for what valuable keywords and phrases?
• By what date?
• How much increase is desired?
• By search traffic do you mean natural search only or does PPC count also?
• Is it really traffic you're after, or a conversion of some sort, such as revenue?
• Is it success to improve rankings without increasing sales?

A more measurable goal would be stated like this: "By December 2009, increase monthly non-branded natural search referred revenue by 50 percent over the prior year." This example goal specifies a time period, a baseline of comparison to measure the success, and a specific numeric value to achieve. Everyone involved should be clear on the achievement or failure to achieve the goal, resulting in easier discussions about course corrections and resetting expectations (whether higher or lower).

What Contributes to a Reasonable SEO Goal?

To create a realistic goal, one must establish dates in which the goal should be met, a baseline to measure the goal, and values in which the goal should be measured, as follows.

• Date to Achieve.
  First and foremost, a date needs to be based on the frequency with which a site is crawled. If your site is crawled every two weeks, achieving measurable results in one month is unrealistic. Even if your site is crawled hourly, changes like content optimization and internal linking structures can take 30 to 90 days to drive results. External links tend to register more quickly.
• Baseline of Measurement.
  Make sure that the time period you choose to measure results is representative. Setting the baseline as the top of a seasonal sales peak will make it appear as if SEO efforts are hurting results when in reality the site is experiencing a typical seasonal downturn. Measuring year over year results or an average of multiple months will ensure that the baseline is more representative.
• Value to Achieve.
  This is the question most often asked: How much growth can you realistically expect? That depends entirely on how well the site is doing today. Factors like opportunity for growth in a keyword market, competition, strategies targeted and ability to implement all play a role in setting expectations.

Measuring the SEO Goal

After the goal is set, establish how it will be measured. Which data pulled from which tools will illustrate performance to goal? How frequently should it be measured to gauge progress? What is the expected trend line – smooth and easy, hockey stick and plateau, decrease before increase? What milestones need to be in place to ensure that course corrections happen if the expected trend line isn’t emerging?

At a minimum, you’ll need to determine if the goal has been met, rather than why it was or wasn’t met. Measuring SEO performance to determine the why a trend has appeared is an entire discussion in itself. But measuring if a goal has been achieved is relatively simple (assuming the goal is specific, measurable and timely).

For example, a tracking spreadsheet for the goal "By December 2009, increase monthly non-branded natural search referred revenue by 50 percent over the prior year" could contain cells to track (a) monthly, (b) non-branded, (c) natural search referred, (d) revenue for every month from January 2008 to the last complete month. If the goal is specific to those four measurement elements (a through d), then the measurement of the goal must be as well.

Showing historic data before the baseline period, from January 2008 to today in this example, gives one the ability to chart a full year of seasonality trends. Understanding seasonality is critical to predicting upcoming trends and determining if improvement or decline is likely due to SEO changes or other expected factors. Using this data in addition to the agreed upon year over year of other goal measurements will paint a clearer picture of success.