Offline Conversion Tracking: 8 methods to measure offline sales and conversions from online marketing campaigns

In order to determine if an online marketing campaign is paying off you have to track (or at least approximate) the number conversions (leads, sales, important downloads, or whatever the goals of the campaign are) accurately enough to determine if the campaign is resulting in an acceptable ROI.

In general conversions that take place on a web site can be tracked accurately enough using conversion tracking code or site analytics, etc.

However it’s much more difficult to track conversions that are completed offline, such as sales that are completed on the phone or during a visit to a physical facility.  If there are significant numbers of offline conversions they need to be tracked to some level in order to estimate the number of conversions from the campaign.

Here are some methods that can be employed to measure or estimate offline conversions. Some of these methods are relatively easy to implement and others can be time consuming and costly. You’ll need to estimate the potential ROI to determine how much effort to spend on tracking.

The on/off test. This is a simple inexpensive way to estimate the offline conversions coming from a paid listing campaign such as Google AdWords. Turn the campaign off for one to four weeks. While the campaign is off count the approximate number of phone leads and/or visits to your facility. Then turn the campaign on with it set to the highest ad spend budget you can justify.  Compare the difference in the amount of phone leads or visits to your facility. It’s reasonable to assume that a large percentage of the difference came for the paid listing campaign. However you could do the on/off test one or more times to be surer. If you get similar results two or three times you can be pretty confident that the paid listing campaign is responsible.

Coupons. Include coupons that visitors can print and bring to a facility. You won’t be able to tell the source of the traffic using this method, only that they must have visited the web site. However you can go further with this method for more accurate tracking (see below).

Unique phone numbers. Use unique phone numbers on the web site that are not published anywhere else. As with the previous method, you won’t be able to tell the source of the traffic, only that the caller must have gotten the phone number on the web site. However you can roughly estimate the numbers from specific sources. See the method named “Estimate Traffic Sources using Analytics” below.

Ask How They Found You. Have the people who handle your incoming calls ask how the caller found you (“found you on the web”, “searched for you on the web”, etc) as best they can when dealing with the call. Then summarize the results. In most cases you won’t be able to tell the source of the traffic (AdWords or organic search or online yellow pages for example). However you can try to estimate the numbers form various traffic sources (see the next method).

Estimate Traffic Sources using Analytics. Use site analytics to determine the percentage of traffic from various sources (direct, organic search, various paid listing campaigns).  If possible go further and look at the numbers and percentages of some important conversions or other actions (such as views of your Contact Us page etc). Use this percentage to help approximate the number of offline conversions coming from, for example, the number of phone calls reported in the previous couple of methods.

Dynamic Phone Number Insertion. To track phone leads or sales there are offline tracking systems that dynamically insert unique phone numbers into the pages of a web site depending on the source of the traffic. Typically you add a tracking parameter to the incoming link which is read by the system. You can track to any level you wish such as the network level (MSN AdCenter vs. AdWords paid listings for example), which adgroup or ad, even to the keyword level. However the more granular you track the more phone lines are needed which raises the cost of these systems. These tracking systems can also be integrated with an internal call center / CRM system.

Live Chat & Click to Call. Buy using prominent Live Chat or Click to Call icons on the web site you can get a large percentage of visitors to use these features instead of picking up a regular phone line. These Live chat or Click to Call applications can usually be setup to track the traffic source.

Build Your Own You Internal Tracking System. You can build your own internal tracking system to just track to just about anything (including some of the previous methods) for example:

You could have coupons generated for visitors to bring to a facility that include a tacking code.

You can have a code inserted onto a page of the web site, such as Contact Us page that your call center people can ask for if the caller is still looking at the web site when they call.

An internal tracking system using first-party cookies can also track latent conversions (conversions that are not completed during the first session, but during a subsequent visit to the web site) much more accurately than the third-party cookie systems used by most PPC networks and they can track latent conversions beyond the typical 30 days most PPC networks can track conversions.

 
For More Information

Most people search online, but buy offline. It’s a few years old now, but still holds true

Measuring ROI from an online marketing campaign (SEO, PPC, or any online campaign). A example of how to measure ROI from an online marketing campaign (In this case offline conversions were not signicant).