- The Adventures of SEO Boy® - http://www.seoboy.com -

Learn How to Project Traffic and Leads Using Google’s Insights for Search

Posted By John On February 12, 2009 @ 8:56 pm In Advanced SEO Tips | 2 Comments

This week I found myself in an interesting predicament. I had to project monthly lead generation totals for the remainder of 2009 for one of my biggest SEO clients. My problem: I only have 6 full months of performance data, and the site has only been live since May of ’07. The obvious solution was to take lead totals from the past 6 months and come up with an average. However, this method doesn’t take into account continuous improvement in the SEO campaign (and other SEM efforts) and seasonality. I was able to come up with a workable projection model using Google’s Insights for Search [1].

First up, enter your primary keywords into the Insights for Search tool. The maximum is 5 keywords at a time. However, the tool allows for you to download the data, so you can replicate the report for as many keywords as necessary. Once your keywords are entered, you can choose the location – worldwide, or by specific country, state, city, etc. Then, you choose the span of time for the keyword data. I chose 2 years, 2007-2008. But the data goes back to 2004.

The primary stat that is pulled is “Interest Over Time.” Defined, this means Google is comparing search volume [2] for your keywords to the total number of searches on Google over time. Interest Over Time is a normalized statistic, and is presented as a number between 0-100. Because you can download the data (numerically), it becomes quite simple to manipulate the Interest stats to show interesting trends for your SEO keywords.

To create the projection model, I pulled the Interest Over Time stats into Excel, and pulled average Interest stats per month. I pulled my report for 2 years, so there were 2 Interest stats to average for each month. Repeat this process for each keyword you’re researching. Here are the Interest stats I calculated for 3 keywords (that I chose as a “cross section” of my targeted keywords):

KW 1 KW 2 KW 3
January

72

54

73

February

61

51

60

March

68

53

68

April

66

58

68

May

64

50

70

June

65

47

63

July

66

57

62

August

66

46

67

September

66

57

74

October

57

43

64

November

54

47

64

December

50

45

52

After finding the monthly averages for each month, the next step is to calculate an overall Interest average. So, for each month, calculate the average across each keyword (kw1+kw2+kw3 / 3). This statistic shows me average Interest per month for the past 2 years. This data can be used to graph out seasonality throughout the year:

Now, it’s time to calculate the percent of change from month-to-month, and use that percent of change to calculate how traffic or leads will be affected by seasonality (or Interest) throughout the year. In this example, I’m looking to project monthly lead estimates. I already have January’s absolute statistics, and I have a solid projection for February (based on performance in the first 11 days). These 2 numbers will act as the anchor to build out the rest of the projection model. This shot shows the Excel formula to determine the percent of change from month-to-month (% Deviation):

Next, it’s time to calculate how the % Deviation will affect the proceeding month’s performance. This shot shows the formula for determining the estimated leads for March (February Leads X % Deviation + February Leads = March Leads):

Copying the formulas down for both the % Deviation and Est. Leads columns will give you a decent feel for your projected leads for the rest of the year. It’s important to note that this does not take into effect changes you may make to the account and any fluctuation in your conversion rates. You can play with the set up of this projection model to incorporate conversion rate factors per month (just realize that if you update a month with increased conversions, it will have a ripple effect through the rest of the projection model). I actually took this to the next level and added factors for keywords that have yet to reach page 1 of the search results and the anticipated increase in lead flow. This involved the application of PPC click-through rate and conversion rate data to estimate conversion data… but I’ll save that for another day.

This is a pretty effective method for projecting traffic or leads [3] when you have a limited amount of data on hand. And this is yet another use for Google’s Insights for Search [4], which continues to grow on me as a useful search marketing tool!

What useful tools have you found for predicting SEO traffic and leads? Do you have a strategy you’d like to share?


Article printed from The Adventures of SEO Boy®: http://www.seoboy.com

URL to article: http://www.seoboy.com/learn-how-to-project-traffic-and-leads-using-googles-insights-for-search/

URLs in this post:

[1] Google’s Insights for Search: http://www.google.com/insights/search/

[2] comparing search volume: http://www.google.com/support/insights/bin/answer.py?answer=87285

[3] projecting traffic or leads: http://www.seoconsult.co.uk/SEOBlog/search-engine-optimisation/is-it-possible-to-predict-seo-traffic-levels.html

[4] yet another use for Google’s Insights for Search: http://www.ppchero.com/how-do-you-anticipate-ppc-traffic-changes-if-you-have-no-account-history/

[5] Tracking Organic Traffic and Leads for Free Using Google Analytics: http://www.seoboy.com/tracking-organic-traffic-and-leads-for-free-using-google-analytics/

[6] 3 Performance Metrics That Can Indicate The Quality Of Traffic to Your Site: http://www.seoboy.com/3-performance-metrics-that-can-indicate-the-quality-of-traffic-to-your-site/

[7] Learn How to Harness the Power of Links to Improve Your Site’s Internal Linking Structure: http://www.seoboy.com/learn-how-to-harness-the-power-of-links-to-improve-your-site%e2%80%99s-internal-linking-structure/

Copyright © 2008 The Adventures of SEO Boy. All rights reserved.