How Much More Traffic Will I Get From SEO?

Note: This is the first in a series of “Free Advice” posts that I will start publishing in response to Twitter requests. Want your SEO or PPC question answered? Ask me on Twitter.

“How much more traffic will I get to my website if I optimize it for keyword X?”

It’s a very common question from prospective clients and was posed by Rick Whittington on Twitter. The answer should help build value for the SEO engagement by demonstrating the potential gains and how to measure them. In this example, we’ll focus on ACME, Inc., a Richmond-based widget reseller with operations throughout Virginia.

Step 1: Determine Your Current Search Traffic Segments

Segmentation is the key to deriving value from your website analytics package. In this case, I am talking about segments of keywords related to a particular theme. For ACME, the keyword segments can be defined as:

  • “Branded” – Company name, brand names, etc. (including common misspellings)
  • “Virginia” – Containing “VA” or “Virginia”
  • “Richmond” – Containing “Richmond” and surrounding towns
  • “Widgets” – Containing “widget” or “widgets”, i.e. “blue widgets” or “widget for sale”
  • “Doo Dads” – Other names for widgets and keywords consumers are also likely to use

In your Google Analytics Keywords Report, select only the keywords from each segment. For example, to select only “Virginia” OR “VA” keywords, use the | separator in the Filter Keyword box at the bottom:

Keyword Filter

Do this for each segment, noting the total traffic volume and number of unique keywords for non-paid search traffic:

Keyword Volume

Plot all of your keyword segments on a chart and you’ll end up with something like this:

Keyword Segments Chart

Now we know how many unique keywords we have and much traffic we are getting from each segment. Hold on to this chart for a minute.

Step 2: Gather Search Volume Estimates

Use the Google AdWords Keyword Tool to calculate rough estimates of search volume for particular keywords. To find it, click on the “Opportunities” tab in AdWords, then “Keyword Tool” in the Tools section on the left.

Type in your keyword segment seed keywords and make a table of the estimate local (U.S.) search volume for Phrase Match to include all combinations of your keywords. DO NOT use Broad Match, it will give you the wrong results for this analysis. You’ll end up with something like this:

Keyword Search Volume

Step 3: Look for Patterns and Gaps

Notice the pattern in the search volume data? “widgets” keywords (and many of the most popular variations) are searched more frequently than the same variations of “doo dads”, by anywhere from 25-100%. Example: “virginia widgets” and “virginia doo dads”.

Now, look back at your chart from step 1. ACME’s site is drawing a fair amount of traffic from a wide variety of “widgets”-related keywords but almost no traffic from “doo dads” keywords.

This is certainly not in proportion to the search volume that we KNOW exists for “doo dads” keywords. Hence, a gap in the SEO strategy and an opportunity to optimize the site for “doo dads” keywords (without taking your eye off of the “widgets”) keywords.

Step 4: Take it One Step Further

Want to be a true analysis ninja and really impress the client? Perform the same analysis but instead of looking at JUST organic search visits in Step 1, substitute conversion rates or conversion counts (leads, sales, revenue, or whatever) by keyword segment and demonstrate a tangible benefit to their business.

But This Doesn’t Answer the Question!

Before you say that this analysis doesn’t tell the client exactly how many more visits their site will receive, keep in mind that that is an unknowable number and no legitimate SEO consultant will guarantee particular rankings or traffic volumes. Clients that are only focused on traffic volume are usually focused on the wrong metrics. It’s all about finding the RIGHT traffic, not just the MOST traffic.

It’s not a perfect answer, but it’s an honest, fact-based assessment of how to improve the client’s business by segmenting data and looking for patterns and gaps. That should make any client happy. I know mine are.


  1. says


    Well done! I have a couple of follow-ups.

    1. Regarding the last two paragraphs of your post, is there any guidance as to what percentage of the average search traffic you would get if either optimizing for or bidding on particular keywords? For example, could I reasonably say I’d capture 1-2% of the traffic for a keyword if I were to achieve first page results or a reasonable PPC campaign?

    2. If I were to only look at Google Analytics for keywords that I’m already getting a low amount of traffic from, aren’t I missing popular terms that I might not be optimized for?

    Love your methodology, by the way!

    • says

      Thanks Rick! Good follow-ups. Here are my thoughts:

      1. It varies by query type (some are navigational, research-driven or transactional) and nobody but the engines know for sure, but most people seem to agree that 70-80% of the clicks on a SERP are on the organic results, the other 20-30% are on sponsored results. Additionally, 90%+ of searchers don’t look past the first couple pages of search results, or even past the top 5-7 results for that matter (more here, here, and here (PDF)). So we can guestimate that roughly 50% of the clicks on a SERP are going to occur on the top 3-5 results and beneath that you are just fighting for scraps. PPC results are even easier to estimate, just run a test campaign and rotate your ads at various positions to see the differences in CTR. Divide that by your share of voice and you get a pretty reasonable estimate of the PPC traffic potential.

      2. That’s where it pays to be focused on conversions, not just top-level traffic. Even if you only get some long-tail traffic in different keyword segments, you should be able to compare your conversion rates to other segments and figure out where to focus your SEO or PPC efforts. And, if there are popular keywords out there that you are missing out on completely, chances are your keyword research would uncover them and you can run a quick PPC test to determine if they are worth optimizing for or not.

      Hope that helps.

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>