I talked to a couple friends who have retail stores on how to build a predictable marketing model for their local business. Here were my thoughts…
ADVERTISING — Impressions from radio, TV or newspaper. Keep a log of every day a radio ad, TV ad, or newspaper ad runs with a key to the # of impressions each vehicle should receive. SITE TRAFFIC — measure daily visits to the site, and visits to contact us and map page. CALL TRAFFIC — Attach a call logger to their phone system to measure the incoming and outgoing calls — caller id, time date stamp, length of call. They could even do lookups to see where the calls were coming from and going to…and find out how often they called back customers. STORE TRAFFIC — There’s gotta be one of the door entry bells that also records number of times it was rung…thereby a rough approximation of # of store visits. CONVERSION — easy…how many sales per day, size of orders, type of products bought.
With these metrics brought together into one spreadsheet, they can measure the relationship and correlations between them to determine which kind of advertising drives the most site traffic or call traffic, if web visits has a correlation to calls, if calls correlates to people coming in, and if they are getting better or worse at converting that store traffic.
With those numbers in hand they have a better handle on ROI for their marketing, know which measure needs improvement, and where there is leverage (ex: cheaper to drive web traffic than ads to get phone calls…which lead to store visits…which leads to sales).