The real key to successful testing is traffic volume.

The lower the traffic numbers, the larger the difference in conversion rate needs to be to decide on a winner. Statistical significance depends on numerous factors but the most important is driving enough traffic to the test to have confidence that the test showing the highest conversion is genuinely a 'winner'

Quality of traffic is next. If you direct thousands of untargeted visitors to a test (visitors who are unlikely to ever become subscribers or customers) then you'll see corrupted results. Traffic should always be targeted, visitors who will be interested in what you sell or offer.

Finally, Test Smart:  We recommend you limit each tests to a single element on your pages, determine a winner and then create additional tests for other elements. (Elements are things like your headlines, the order buttons, page images, colors, call to actions etc.) The more variations you have in a single test the longer it will take (or the more traffic you will need) to determine what's working.

This actually depends on the goal of your page...

We all know it's good marketing practice that a single page should have a specific, singular goal with a desired result, but what that result is, is really up to you.  It could be an email opt-in, a product sale or an affiliate link click etc.

What We Test And Why:  The most important factor is always the offer itself. (This is what you're providing to the visitor) The offer is the most important consideration in the mind of your visitor. What solution are you providing?

So first test any element that could make your offer more compelling or elements that explain why you provide the solution your visitors are looking for.  Q:  Can you add more persuasive elements to your page to increase the desire for your solution? - A:  ALWAYS!

When you get the offer right you're 50%
of the way to achieving your desired goal.

Secondly, (and also HUGELY important) is your Headline. Testing headlines can have dramatic results. The right headline can deliver 200%, 300% and in some of our tests even 500% increases in conversion rate. So test your headlines (especially on landing pages and sales pages) and ensure you're using good sales copy practices.

Combine a great headline with a great offer and you're
practically guaranteed to achieve your page goal.

Beyond these two key elements we're entering "testing geek" territory: Font styles, images, page or background colors, buy buttons, using credit card icons, guarantee seals, calls to action, sub headlines, bullets and post scripts, pop-ups and more... And while each can impact your conversions rate (thus worthy of testing), changes here generally produce only minor improvements.  For instance an alternative style or color buy button style may deliver a 5% lift in conversion. But your main priority should be far bigger lifts. So make these test once you have the big ones in the bag (the offer, the headline).

Can I Still Split Test With Low Traffic?

Yes you can! And you should - Even if your site gets low volume traffic having a split test in place can be beneficial (especially so if you set your split test using Split Test Monkey's 'Autopilot' feature) - It might take a while to deliver a result with the right level of confidence but it will eventually come.

Even better, if your site gets an unexpected traffic boost (maybe from an affiliate/JV partner or influential tweet/social share) then the data you gather will be genuinely valuable.

Consider making BIG changes rather than minor ones - You won't need as much traffic to get a reliable (significant) result - To do this change the offer itself, your page layout, switch out the entire sales copy (yes multiple changes on a page are fine) but test only two variants (page a vs. page b, not a/b/c/d/e/f/ etc.)

Once you have a winner use that page as your control and continue testing but this time against fewer individual elements.  

Also consider that traffic alone might not be your only hold up...

Very low conversions can prolong testing results, but you also can work around this if you need to. Don't test for sale conversions, instead test alternative metrics such as visitors clicking the order button through to your checkout page. This way you can test and optimize your page even with less actual sales.

The key point here is that if you don't test, your conversions will never improve, nothing will change and you'll be constantly stuck in a cycle of guess/fail/guess/fail.