I am a numbers guy as well, and I follow your logic 100%!
Data in and of itself is such a wonderful thing. Data makes no claims, it simply reports happenings. The numbers are pure and innocent. There is nothing to it one way or another. Interpretation of that data is where things get cloudy. So often, data interpretation can be skewed by the use of deductive reasoning and squeezing that data into logic equations in which it was not initially intended. The most common error in this manipulation is a small data pool. I have always struggled with this because I have never worked in an environment in which I could apply any data interpretation to a very large pool.
Sometimes data can give us tremendous insight. Sometimes it simply tells us nothing. I believe that this is what we are dealing with here. 1 in 3 people that identified themselves through chat bought within 90 days. So? "Bought"? Bought where? Bought from the dealer that they initiated chat with? How many dealers did they initiate chat with? What percentage of customers that picked up the phone and called a store purchased within 90 days? I personally believe that we are micro-analyzing a group of people that actually "will use" chat, not necessarily demand chat.
I dropped chat from my website about 60 days ago. I am certainly not saying that chat is bad. I am unquestionably saying that a chat icon was a waste of real estate for me. My customers just didn't click on that button. I sure wish they would have. Oh, and before all of you vendors start blowing up my phone, I tried 3rd party chat, mobile chat, pop up chat window with employee pictures and without, drop down chat windows, about 10 different color combinations, a small chat icon on all 4 corners as well as floating on the side, time delay between 5 seconds and 60 seconds on a page, as well as dealership initiated chat in which we manually invited customers to chat by live monitoring of the actual site visit. It didn't work.
