Adam T. Sutton

SEO Metrics to Measure

February 23rd, 2010

Natural search marketers have been in a precarious position for the last few years. Much of the data they’re using is supplied by search engines, and some of that data is fuzzy at best.

Adam Audette, in a Search Engine Land post today, goes as far as calling some of the data unreliable and “downright misleading.” However, Audette astutely notes that marketers need the data even if they don’t completely trust it.

What’s a marketer to do? Here are Audette’s suggestions for the SEO metrics you should track:
o Percentage of overall site traffic from search
o Percentage share of each engine
o Free search traffic at the keyword level, clustering related terms
o Difference between branded and non-branded search traffic

Metrics that he implies are far less reliable:
o Ranking reports
o Indexed page counts
o Backlink counts
o Toolbar PageRank

For marketers, I would add conversion data to Audette’s list of primary metrics to measure — especially conversion data for non-branded keywords. If you’re a natural search marketer, any conversions you can prove came through non-branded keyword searches point directly to money you are bringing the company.

Branded search conversions are great, but they show that the searcher already knew your brand. The searcher has likely been reached by another marketing channel. A non-branded conversion implies that someone chose you over the competitors also listed in the results.

Which metrics do you consider vital? And how reliable are they?

Adam T. Sutton

About Adam T. Sutton

Adam T. Sutton, Senior Reporter, MarketingSherpa
Adam generates content for MarketingSherpa's Email and Inbound Marketing newsletters. His years of experience in interviewing marketers and conveying their insights has spanned topics such as search marketing, social media marketing, ecommerce, email and more. Adam previously powered the content behind MarketingSherpa's Search and Consumer-marketing newsletters and carries that experience into his new role. Today, in addition to writing articles, he contributes content to the MarketingExperiments and MarketingSherpa blogs, as well as MECLABS webinars, workshops and summits.

Prior to joining MarketingSherpa, Adam was the Managing Editor at the Mequoda group. There he created content and promotions for the company's daily email newsletter and managed its schedule.

Categories: Search Marketing Tags: , , , ,



  1. February 25th, 2010 at 09:12 | #1

    Totally agree that the less reliable metrics are, well, less reliable. However, they can be useful “directional” indicators of optimization success — especially when looked at as trends over time. In fact, trending all metrics over time is a best practice we recommend as it helps you see patterns emerge that can be useful in your overall marketing planning (e.g., seasonality factors, etc). Finally, if you are running a paid search campaign in addition to your natural search effort, it is important to separate the two when looking at your traffic and conversion metrics.

  2. February 25th, 2010 at 16:50 | #2

    Hi Marc — thank you for your input. I think you raise a valid point. I would suggest anyone doing analysis to give more weight to reliable metrics, and to place a large asterisk next to fuzzier data. But fuzzy or not, all the data has value.

  3. March 7th, 2010 at 18:10 | #3

    Adam – great point that if people came using the branded keywords, then they knew about the company or product already.

    To your point on the conversion, do you have any data points as to how well conversion helps in SEO? If people come to the web site and spend more time, and take meaningful action, would that by itself help in building credibility with the search engines and so the ranking?

    Thanks,

    -Deven

  4. August 10th, 2010 at 16:51 | #4

    Hi Adam,

    I just came upon your article while searching for SEO metrics. I have a question about measuring the conversion data for non-branded keywords. While I so agree that this an important metric, what do we do about the cases where the non-branded keywords brought initial visits, but the visitor doesn’t convert until two or more visits later? How do attribute that conversion?

We no longer accept comments on the MarketingSherpa blog, but we'd love to hear what you've learned about customer-first marketing. Send us a Letter to the Editor to share your story.