When planning a research project, many people just start writing a questionnaire and organize their thinking on the fly. Perhaps there is some vague objective, like “measure the effectiveness of our ad.”
For those thinking “What’s wrong with that objective?!” I’ll have more to say about setting objectives in a future tip of the week.
Using this approach to planning a survey is like jumping in the car and driving away without knowing where it is you are trying to go.
So, what do you do? Start from the end.
By end, I mean start from a well defined objective. Think about what calculation you have to make or what piece of information you need to achieve the objective. Next, figure out what inputs you need to do that, and what inputs you need to get these inputs. Continue this until you work your way back to a single, specific data point.
An example: The objective is to see if your ad changed people’s opinion about the quality of your product. To do this, you need to compare the opinion of quality held by those who saw your ad and those who didn’t. To make that comparison, you need two data points:
- Did the person see the ad?
- What do they think about your product’s quality?
Depending on how you advertised, knowing who did and didn’t see the ad may be quite easy (email) or not so easy (broadcast, print, etc). In the case of email, if you tracked the campaign, you already have the answer to whether or not someone saw the ad.
Asking about an opinion of product quality is a fairly straightforward thing to do.
By starting from the end, we were able to do several things to get better research:
- We identified who we needed to talk to
- We identified exactly what pieces of data we needed from them
- We determined whether we had to ask a question or we could get the data from existing sources
- You are sure of getting exactly the information you need and your survey respondents are spared wasting time answering questions that you can’t use or don’t need.
Win – win.