Q: We are in the process of gathering data about our customers. Bottom line, we want to know what products/services they are purchasing these days. After listening to a LERN speaker at a WeTrain conference last summer (sorry, his name escapes me, he is a futurist and has written a book), we decided that several short surveys of our top customers might be the place to start. He suggested imbedding the short (3-6 question) surveys into emails, but my IT folks are telling me this will make it tougher to compile and compare the results. They are suggesting we imbed a surveymonkey link into each email. I thought I understood that any additional steps will only make it “slower” for customers and they may not bother to take the extra steps, thereby impacting our response rate. Am I making sense? Is there a magic number of responses that are too difficult to calculate by hand? Also, how do I go about surveying people at organizations where we would like a presence, but are not currently working?
A: Lisa, LERN uses both strategies. I don't know how many responses you anticipate, but if it is 100 or fewer you have lots of options. We have had very good results with the email surveys. They are quick to prepare and easy to respond to. People can enter their answers right away and be done with it. Results are compiled manually, but it takes very little time. Another option to consider would be to create a form using Google Docs. Email the form to your respondents. They fill in their answers on the form and it is automatically uploaded to an excel spreadsheet in Google Docs. You don't have to do any manual compiling or anything else. Ask your IT folks to demonstrate how forms work for Google Docs if you are not already familiar with them. It is pretty slick. Since you are doing a simple survey, all you need are basic counts of people using a specific project. Even if you did this with tally marks on a sheet of paper, you could quickly and easily compile results for up to 100 respondents. You are only asking 3-5 questions and the responses could be tallied as they come in. Survey Monkey is also useful. I have found that we get good response when using it for some types of surveys. You might test the different methods and determine which works best for you. Given that this is a simple survey not requiring sophisticated data analysis, any of the approaches would work just fine.