How to Increase Your Conversion Rate If Your Traffic is Too Low for A/B Testing

Frustrated because you want to do A/B testing to improve your website conversion rate and revenue, but you don’t have enough traffic?

You are not alone. One of the questions I receive most from businesses is how to improve their website if their traffic is too low for A/B testing. So what can you do?

The simple answer is yes, you can still use A/B testing, or an equivalent, no matter how much traffic your website has. That is because there are techniques to use that make your A/B tests more efficient on a lower level of traffic, and if your traffic levels are too low for that, there are even similar techniques to A/B testing that you use to improve your conversion rate.

First, check if your traffic is too low to run an A/B test

The most important first step is to understand how much traffic you need for the website page you want to A/B test.

How much traffic do you need to run an A/B test on a single page?

To put it simply, you need at least 5,000 visitors per week to the page you want to run an A/B test on. If you don’t have this much traffic, it will take a long time (often months or never) for your A/B testing tool to gather enough data to find a statistically significant result. Use this handy A/B test length calculator to give you an idea of how many days it will take you to get a result. Also, the more page variations you test, the more traffic you will need to get a result. But you need something else too…

It isn’t just about traffic though, you also need enough conversions

Even more importantly, you also need enough ‘conversions’ on your website to run an A/B test. This is because to run a test, you need to tell the A/B testing tool what determines success, and this is usually a major goal like a purchase, a sign-up or a form completion. And the less conversions your website gets per week, the longer it will take the testing tool to find a winning result. As a guideline, your website needs at least 500 conversions per week for a simple A/B test (250 per test version).

Techniques to turn low traffic ideas into ones you can run

Don’t have enough traffic? Don’t give up hope about trying to improve your website conversion rate and revenue! You can still use some techniques to increase the chances of you being able to run an A/B test, or use some alternative approaches to give you similar insights and website improvement results. Here are some great techniques to try:

Use the add-to-cart goal instead of orders
As mentioned earlier, having enough conversions is even more important than traffic for getting a testing result. And to increase the number of conversions you get (and increase your chances of a successful result) use the add-to-cart goal, as that happens much more than orders, yet still has a strong correlation with orders.

If using add-to-cart still doesn’t give you enough conversions to run an A/B test, you can also consider using an engagement goal, like clicks to the next page in your funnel. These are often called micro-conversions, and which ones you can use depends on the location of your A/B test. For example

• Clicks from category page to a product page (for ecommerce websites)
• Clicks from your cart page to your checkout (for ecommerce websites)
• Clicks from homepage to your sign-up or pricing page (SaaS websites)

When changing your goal from orders to add-to-cart, don’t forget to check what your add-to-cart numbers are, and use them in your A/B test calculator instead of orders.

Engagement metrics to avoid: I don’t recommend using time on page or scroll depth goals though, as they often don’t correlate with a positive website experience, and can cause false positive results. And if you launch the winner based on one of those goals, it may actually have a negative impact on your conversion rate or revenue.

Only A/B test a single variation instead of multiple variations

It’s common and exciting to want to test multiple different variations in an A/B test, as you may have multiple ideas for it. However, the more variations you want to test, the more traffic and conversions you will need to gain a statistically significant result.

So if your A/B test duration calculator says you won’t have enough traffic or conversions for your A/B test, then you will need to limit your A/B test to only one variation. Then the chances are much greater that you will have enough traffic and conversions to run your A/B test.

Here is where the improved version comes in. This doesn’t mean you can’t test other variations, you simply need to run follow-up tests with the other variations you want to test. So if the initial version you wanted to test wins, then you would test that against one of the other variations you want to test.

This iterative style of A/B testing is always recommended anyway, regardless of your traffic levels, as the cumulative effects will often result in larger gains in conversion rates, rather than just testing a few variations of one element and then moving on to test something else.

Make your A/B test variation bolder and increase the minimum detectable effect
When using an A/B test duration calculator to determine if you have enough traffic, there is one thing you can change that will mean you need less traffic or conversions. It’s called the called minimum detectable effect (MDE), and the higher you make this percentage, the less traffic your tool will need to detect and determine the winner.

This is often at only 5-20% by default in most test duration calculators, and you can increase this to 30% or more.

Here are the MDE’s that the most common calculators use, which you can increase higher:

• VWO – 20% MDE
• Optimizely – 20% MDE
• Evan Miller – 5% MDE
• AB Test Guide – 15% MDE

To increase the MDE to 30% or higher, it is essential to create a really strong A/B test variation that will get detected faster by your website users (and therefore also your tool). You should use either of these techniques:

• Change multiple elements on one page. For example, on your product pages, at the same time, change which product image is shown first, make your free shipping or returns messaging more prominent, and add benefit icons in your header.
• Change elements on multiple pages. For example, changing your homepage hero buttons, navigation, and category page filters at the same time. You would need to use a multi-page A/B test for running this.
• Make bolder variation changes and make sure it stands out as much as possible. For example, using a colour that contrasts better, a very different style of imagery, a larger font size, or even using animated elements.

To make use of this technique, it also depends on what your current conversion rate is. If you already have a high conversion rate, like over 5%, then it will be much harder to gain a 30% increase on that, compared to if you have a much lower conversion rate of 1%.

What to watch out for: Don’t be tempted to increase this MDE if you are testing something very small, as your tool won’t detect lower differences, and you won’t get a result. And the higher you make the MDE, the less chance your tool will be able to detect a result, so I don’t recommend going much higher than 30%. If you do either of those, it will also increase the chances of wasting your all important time and resources for A/B testing.

Techniques if you still don’t have enough traffic or conversions

Do my ‘User-Focused A/B Testing’ technique which doesn’t require any traffic 

My new technique called ‘User-Focused A/B testing’ helps you discover ‘why’ users prefer different page versions, not just ‘what’ version they prefer. And it doesn’t require lots of traffic like traditional A/B testing does. You can even do it on design prototypes.

And more importantly, it gives you great feedback and insights for creating high-impact website improvements, meaning a much better chance of increasing your website conversion rate and sales. And A/B testing can never give you this feedback either, so I suggest you also use this method even if you do have enough traffic.  Read full details about my new technique.

Use Google Ads experiments to split test and find high-converting wording
With this technique, instead of creating an A/B test in a testing tool, you use Google Ads experiments to find which ad variations get most clicks on wording relating to the page you want to improve. This is ideal for testing wording for headlines, benefits, call-to-actions and other important words on your website.

To do this simply create a Google Ads experiment using a few different ads that emphasize different headlines and descriptions, let it run and see which has the highest conversion rate (see below for an example of an experiments report).  Then once you have found the winning ad, replicate that winning ad text on the page you wanted to improve, and watch your conversions and sales grow.

Launch your improvements and then monitor impact on key metrics
This technique is where you simply launch the new improved page version for one week and check for the impact on key metrics like website conversion rate and orders. Then determine if there is a good improvement in comparison to the previous version of the page (you will need to create a good benchmark of metrics to compare this to – at least a month ideally). Then if the new page performs better, you keep the improvement live. If it performs worse, then you roll back to the previous version of the page.

Although you don’t get statistical significance with this method, it’s better than just launching new pages and hoping for the best. To reduce the chance of issues while you are monitoring new improvement launches, try not to make any major change in traffic source or any other major website changes that might impact your key metrics.

When doing this type of website improvement launches, the quality of your improvement idea is really important for determining success. Rather than just guessing, to increase the chances of success, I suggest you get website improvement recommendations from CRO experts like myself.

You can still do CRO if you don’t have enough traffic for A/B testing

You may think you can’t do conversion rate optimization (CRO) to improve your website if you don’t have enough traffic for A/B testing. Many people even think that A/B testing and CRO are the same thing. Fortunately these thoughts are both wrong. But why?

It is because A/B testing isn’t essential for improving websites and is only one of the four main elements of CRO. Conversion research, user experience (UX), and website persuasion are the other key elements.

Learn about these other elements of CRO to help improve your website converion rate and revenue.