𝗡𝗼𝘁 𝗮𝗹𝗹 𝗿𝗲𝘀𝗲𝗮𝗿𝗰𝗵 𝗶𝘀 𝗴𝗼𝗼𝗱 𝗿𝗲𝘀𝗲𝗮𝗿𝗰𝗵

You’ve heard me banging on about how important it is to understand your customers, their desires and then how to adapt your product development to meet these needs. I haven’t changed my tune, but there is a right way and a wrong way to do it. If done poorly, using the wrong signals, you may end up building a worse product despite your best intentions.

I always review surveys to understand how questions are being asked and what I could learn. In the attached survey, not much.
Here are the major issues:
-𝗟𝗶𝗺𝗶𝘁𝗲𝗱 𝗼𝗽𝘁𝗶𝗼𝗻𝘀: they survey has only pre-selected options, forcing you to choose from them. e.g. if one of my reasons for choosing was because of great range, but its not an options then that insight is lost (which is why I prefer an open question, which AI can easily help you classify)
– 𝗢𝘂𝘁𝗱𝗮𝘁𝗲𝗱 𝗺𝗲𝗺𝗼𝗿𝗶𝗲𝘀: I bought something from them 18 months ago, the chances of me recalling the difference between a FB ad or post is small. People want to help in a survey (especially if there is a chance to win something) so they’ll give you a (possibly incorrect) answer. You don’t want to base ad spend on misremembering of 18 months ago.
– 𝗔𝘁𝘁𝗿𝗶𝗯𝘂𝘁𝗶𝗼𝗻 𝗶𝘀𝘀𝘂𝗲𝘀: if you want attribution on marketing spend, use analytics not survey feedback. You’ve got enough of a war on analytics of whether the first, last or any subsequent click was more important, but at least you can get a revenue to channel match.

In short: Ask more open-ended questions, avoid leading respondents, and use the right tools for the insights you’re seeking.