A/B testing. You have probably began to hear more about this popular newer technique for quickly increasing website sales, leads or subscribers from your existing visitors, and there are now tons of A/B testing case studies highlighting outstanding results you can get.
This technique helps you quickly test different elements and pages of your website to find variations for converting a much higher percentage of visitors into sales, leads or subscribers (known as increasing conversion rate – a really cost effective way of increasing sales).
To benefit from this, all you need is an A/B testing tool like Visual Website Optimizer or other test management tools, some good test ideas, and enough traffic for testing. Unfortunately for many blogs and small businesses this last part is often problematic – to get test results quickly you need a good amount of traffic to test each variation.
Therefore I’ve put together a guide to help you understand how much website traffic you really need to run A/B testing, and how to still benefit from it if you don’t have enough…
First of all, just how much traffic do you need to do A/B testing?
Let’s answer this important question first. Many experts agree that in the very least you will need 1,000 visitors each week to each page you want to run an A/B test on, otherwise you won’t get results from your A/B tests at all (or will take months to get a result). You need this good amount of traffic for the tool to serve each test variation to your visitors, and for it to find a statistically significant result.
To get results you also need many website conversions (usually your major website goal like sales, leads or subscribers) per week – at least 50. To get enough conversions to run A/B testing, consider making the conversion metric for your A/B tests be something that occurs much more often, like a click through to another page, or a click on a specific button.
You probably now realize that getting enough traffic/conversions is obviously a problem for many blogs and small businesses websites who don’t get that much traffic regularly. So…
What to do if you want to do A/B testing but don’t have enough traffic
You have two main options to consider – either direct a large amount of traffic in a short period of time to the page you want to test, or try some good alternatives to A/B testing:
Get more traffic to the page you want to test:
- A simple but more costly solution is to buy more visitors to increase traffic for the page you want to do A/B testing on. This way you hopefully get enough traffic to run an A/B test, and could be driven by using Google Adwords or Facebook Advertising (which I think is cheaper) or any online advertising method.
- If you have a large enough email subscriber list (over 1-2K subscribers), you could use that to drive enough traffic to do A/B testing. Just send an email campaign to them pointing specifically to the page you want to test, and then run the A/B test quickly. The bigger your email list, the more chance that you will get enough traffic for a test result.
Use great alternative methods to A/B testing:
And if you can’t get more traffic cheaply or easily, here are other great ways to get benefits similar to A/B testing. I actually suggest using these alternatives too even if you have enough traffic, as they give insightful feedback about proposed improvements on your website.
- Use Google Adwords or Facebooks Ads to do testing on titles and text in the ads. Then based on which versions you find gets highest click rates in the ads, use similar headlines or text on your key website pages. This is particularly effective for finding which versions of benefits or unique value proposition are most compelling to your visitors. Here is an example of 2 different styles of headlines and text being tested:
- Use simple low-cost website usability tools to get feedback from ‘testers’ on two different web page variations and see which is liked the most by them. Simply design a mockup image of a better page that you want to test, and use a tool like UsabilityHub.com to get quick high-level great feedback from testers on the mockup and the current page, and see which they prefer. Unfortunately you can’t ask specific questions, but it’s a great cheap way to get feedback on proposed page improvements.
- Get qualitative in-depth feedback from your target market about your current site and proposed improvements you want to test, using a service like UserTesting.com or WhatUsersDo.com. Both of these tools allow you to select testers from your exact target demographic, and ask many specific questions to them about which versions they prefer. You also get a useful recording of them giving feedback on your website or mockups. It’s like super insightful A/B testing for your content!
- Run a short survey on your website visitors using a tool like SurveyMonkey.com, show them proposed new improved page variations, and then ask them questions about which they prefer and the reasons why. You can also include questions about what else they would like to see improved on your key pages – great for creating further ideas for testing and improving your website.
Launch improvements after using these A/B testing alternatives
Go ahead and use a few of these alternative ways of doing A/B testing, then using what you learn from findings, make some improvements to your website and monitor your conversion rates for the next week. Hopefully they will bring you great results on increasing conversions, sales, leads or subscribers.
If the changes don’t seem to make much of an impact, come up with other more radical ideas for page variations, get more feedback on them, and then try launching improvements from what you learn. Keep on trying different versions until you succeed – never give up!
Hopefully that’s helped shed some light on how you can get the benefits of A/B testing if you don’t have enough traffic to your website.
Don’t forget that to get best results – higher conversion rates, sales, leads or subscribers – you need to come up with high-impact ideas for testing. And here are 12 great A/B test ideas to help kick start your efforts.
Now over to you – tried A/B testing yet? What’s worked best for improving your website?