3 ms·
Odd to see so many people focus on the slider. I don’t really care about the slider--meaning I'm neutral on it. What I do like is your copy for describing the s
by kevin 11y ago
Odd to see so many people focus on the slider. I don’t really care about the slider--meaning I'm neutral on it. What I do like is your copy for describing the sentiments. I threw the slider back and forth just to see all the options. Unlike other surveys, it felt like it more accurately resonated with me over other likert scales:
http://allpsych.com/wp-content/uploads/2014/08/likertscales.gif http://allpsych.com/wp-content/uploads/2014/08/likertscales....
I think that made the biggest difference on getting people to contribute. It felt like the data was going to represent me. This is probably a small experiment, but if you’re thinking about this mini-site as a form of lead generation for WorkShape…then your main goals should be around getting people’s credentials and having it spread to their coworkers.
I would highly recommend figuring out how to give people accounts without passwords needed up front. It should feel almost like a mailing list at the end of the survey…
“Thanks! Let us know your email and we’ll check back in on you next week.”
Bonus points if you explain clearly why tracking this weekly is good for them. Then in your confirmation email, you can say you generated a password and you can follow their stats over time through this login link.
Your secondary goal should be explicit as well (if it’s what I think it is).
“Want to track how your entire team is feeling this week? Send them a link.”
Repeat on telling them why that’s a great idea. Right now, everything seems a little too subtle. Of course, I know you were probably just whipping this up to get it out the door.
Again, the copy is the best part of this site. If you change anything on the interface, be sure to leave that alone. Good luck taking this further!
- GordyMD 11y agoAppreciate the comment and time taken to write this. We do quite like the labels too, but I do recognise, as pointed out by others, that their presence could create a form of bias. I believe that the choice of wording could divide the audience, with some people connecting with the sentiment of each and some people not. The concept of the survey is a neat idea, and could be a good thing for us try out. The suggestion of adding in a prompt for inviting team members is also good, if we decide to go down that route. Thanks again!