An average website has a sales conversion of 1% to 2%. So it’d be fair to say that on most sites more than 95% of visitors don’t buy anything—especially on their first visit. So instead of trying to sell them right away, we should capture leads (i.e. emails) instead.
I firmly believe that the best way to sell something is to avoid it at first. You need to let people know you first, earn their trust, and then you might try to sell to them.
Sure, this somewhat depends on the industry, but it’s true for most businesses.
This is not the focus of this article, but in a nutshell, the formula for making a tons of sales online in this:
- Capture emails.
- Build and develop relationships via regular communication.
- Add tons of value (way before asking for anything in return).
- Repeat step #3 as much as possible.
- Sell great products.
That depends a lot on what you sell. As Exit Intelligence‘s CEO Matt Cimino recounted:
We had one client selling a high-end outdoor product. For them, offering a discount of 10% didn’t convert that well on a $500 purchase—it was a valuable offer, but the after-discount cost was still $450.
However, it was a product that you would use with your friends. An alternative offer of “Enter for a chance to win a package of 10 for your friends” worked far better.
Still, despite product-by-product and company-by-company variations, there’s one online lead generation technique that has produced far better results for me than anything else—tests and quizzes.
There’s something magnetic about tests and people are almost drawn to them. Some people are even willing to pay a fee in order to take a personality quiz.
I think it has to do with how people like themselves. Or maybe it’s nostalgia (thinking back to the teenage years and taking silly tests in magazines). It might be a subconscious hope to discover something magical.
Whatever the case, it works ridiculously well.
Examples of tests and quizzes
Example 1: T1Q
T1Q was a site for people who have no idea what to do with their lives. When I launched the site back in 2008, it had a typical email capture form on the site:
It got ~5 people per day join the list, out of ~500 daily visitors. 1% conversion rate. A couple of years later, I split tested it against an invite to take a test (both forms were on the home page):
The results: Now I got around 100 people per day joining the list. Overnight. That’s a huge difference (+1900%).
A later version was converting even better.
Why it works so well
Besides being a free test and having therefore innate attractiveness, it does one key thing: it solves a problem the visitors have (or to be exact, offers a promise to solve it).
Most visitors come via search and they are Googling variations of “find life purpose” or “what should I do with my life.” The keywords told me the problem they had, and this free test was built exactly to address these questions.
The lesson here is: don’t just build a random quiz. Your business and website solve a real problem. The test should, too.
If you don’t know what problems your customers need help solving, you’re not doing it right :) Talking to users should be one of the two key activities in your business.
Tip: Make your quiz relevant to your product
I could have also built “Do women find you sexy” quiz—and I’m pretty sure it would have been popular—but it wouldn’t attract my ideal customers. So I’d end up with a bunch of useless emails.
Ideally, your test would only be attractive to people who are also the target group for your products. In T1Q’s case, it’s ebooks, courses, and coaching programs related to finding your purpose and passion.
Example 2: ShopGemstones
ShopGemstones is a gemstone information site. One of the best in the world.
Their goal: build up an email list of gem nerds (my words, not theirs). Gemstone people are often somewhat esoteric and believe in the healing properties of gems and so on.
So a test was designed to attract these people:
A total of 11 question, and email asked at the end:
We did something else that was clever—we turned the test into a link building machine. Once they got their results, we showed them this among other things:
Hundreds of test takers added this to the sidebar of their blog. Once one person did it, her friends saw it on her blog, took the test, and put their results onto their blogs. ShopGemstones got hundreds of links thanks to this.
Example 3: Should I Quit
This “Should I Quit My Job?” test is technically a funnel to get people to sign up to courses like Live Off My Passion.
It targets a specific user and plays on the natural curiosity (“are you one of them?”) of people.
They ask for the email at the end. A score is given right away, but for the explanation you need to confirm your email (double opt-in).
I’m not in a position to disclose exact figures, but I can tell you that this resulted in thousands of opt-ins and nice sales down the funnel.
Can’t build a test? Build a tool
In B2B settings, it’s more difficult to come up with a good idea for a test, but a tool could be easy (although harder to put together).
HubSpot is killing it with Website Grader. Every business owner is interested how their website is doing, how they stack up against the competition, and giving your email for useful insights seems a small price to pay.
TweetCharts does something similar, letting you dig into Twitter data:
The main idea is the same: help your prospects solve a problem, and ask for their email.
Email up front or later?
There’s two ways of thinking here.
- Ask for the email before the test. Users see the test value proposition and they want the benefits. There is no other way to get to the goodies, but to give your email. So they do it. Also, a significant amount of users do not actually complete tests—so if they don’t reach the end, you also lost their email.
- Ask for the email at the end. Less friction up front—it’s easier to get going. At the end, you state that they will get the results by email (can reveal partial results right away), so they’re motivated to give it to you.
You have to choose which scenario is more likely to work in your case. Better yet, test it!