I use FiveSecondTest.com to test landing pages before I build them. It shows a mockup of your page to random people for 5 seconds, then lets you ask them questions to learn what they recalled and understood. This is very valuable to make sure the message you intended is getting communicated, which is important to rule out if your smoke testing a value proposition.
It answers the question: Is this doing poorly because the offer is bad or the design is bad?
My typical questions are:
You'll notice that these questions aren't leading in any way they give no clues about the design except for the last question, where I tell them there's a button.
If I don't get the answers I expect to these questions, I know I need to improve the design. You'd be surprised how differently people think from you. I iterate on FiveSecondTest until people clearly get the point and can answer those questions to my satisfaction. Only then does the page go live.
Sometimes, this is a frustrating process. People are super lazy and don't really read stuff or take a proper look at your page, so you really have spoon feed the information. But sometimes, thanks to the fact that FiveSecondTest has a decent number of smart-ass designers on it, I get answers that crack me up. Its nice when your work makes you laugh out loud.
What are the reasons the page gives you to click the button? A pink arrow. Gee, thanks!
Btw, here's a later version that consistently got the right answers to those questions:
One more item and bolder text did it! Go figure.
A tough experience triggered me to reconsider by goals, so switched from educational services and NGOs to a monastic, creative life.
Mentor Impact - from the startup mentors that get results.
Decision Hacks - early-stage startup decisions distilled.
Peer Learning Is - education for fast-changing topics.