Table 1 shows conversion rates for the landing pages displaying our client's 800 phone number versus their local number:

  Table 1 - Conversion Rates by Phone Number

City Conversion Rate 800 Number Conversion Rate Local Number  Oakland 8.8% 14.3%        San Francisco 15.6% 34.4%        San Jose 8.2% 26.2%        Seattle 9.9% 17.3%        Total Test 10.4% 21.9%

Table 2 shows the percentage of phone calls received from the landing pages testing the display of our client's 800 phone number versus their local number:

Table 2 - % Calls Received by Phone Number

City % Calls Received With 800 Phone Number % Calls Received With Local Phone Number  Oakland 38% 62%        San Francisco 31% 69%        San Jose 24% 76%        Seattle 36% 64%        Total Test 32% 68%

Based on this limited test, our team observed that a local number on landing pages converted more visitors to call at over twice the rate of the identical landing pages displaying the 800 phone number. 

Due to the nature of this client's business, they do not have an online form as a conversion option, so the only action that can be classified as a conversion is a phone call. Based on this, we were not able to determine what impact, if any, would accrue to online conversions. 

The takeaway here is that if you sell services in more than one area, you'll definitely want to setup an A/B test of phone numbers within geo-targeted campaigns. 

You may want to even consider performing a 3-way test: A page displaying the 800 number against an identical page displaying the local number against yet another identical page displaying both numbers. 

If you don't have a call tracking solution to measure the calls, you can learn more about Engine Ready's Call Analytics software at: http://www.engineready.com/products/callanalytics.html.