There comes a point in every CRO’s experience when he or she thinks, “That’s it. We’re done. My conversions won’t get any higher. I’ve reached a conversion threshold.”
What do you do? Admit that you’re done? Stop optimizing? Watch cat videos instead of setting up a new A/B test?
I’ve encountered this situation many times with clients, with myself, and in the general literature on conversion rate optimization. I’ve given this issue some serious thought, some more serious research, and have produced a serious few thoughts that you might find useful.
What do you do when you’ve reached a conversion threshold? I have a five-point checklist that I recommend.
(Note: Keep in mind that this slippery thing we call a “conversion” doesn’t equate with “sale.” A conversion can be as something as innocuous as a newsletter signup.)
1. Deny that there is a threshold.
Allow yourself a bit of hubris.
Deny limits. Deny thresholds, and tell yourself that you can go higher.
Industry averages are meaningless.
There are a lot of studies that display the average conversion rates by industry or search network.
- Average conversion rate for the search network in Q3 2012: 5.63%
- Average conversion rate from the display network in Q3 2012: 4.68%
Larry Kim of WordStream has provided an often-cited study on the median conversion rate across industries, which is 2.35%.
Obviously, every industry is going to have different conversion rates, which is why Kim also created a helpful chart plotting the CVR (conversion ratio) by industry and performance levels.
But what do these metrics really mean? They mean nothing for you. Why? Because they are simply an aggregate of random companies. These companies don’t directly affect you. They don’t tell you what your CVR can or can’t be.
You have a right to engineer your own conversion rates.
The only real threshold is 100%.
There’s only one true conversion threshold, and that’s 100%. It’s a mathematical impossibility to have more conversions than you have visitors. It has to stop at 100%.
I often read statements like this: “You can’t convert 100% of your visitors, because some of them, no matter what you do, may never be prepared to take action.”
I get that. I understand what he’s saying. But that’s placing artificial limits on the world. In the first place, I don’t like the word “can’t.”
Second, this statement is built off of a hypothetical — the “may never be prepared” bit.
Third, the author of this discouraging declarative doesn’t know what I’m going to do. (He says “no matter what you do.” Little does he know what I have up my sleeve ...)
When I look at the CVR universe it all its dismal reality, I tend to agree with the statement that you can’t convert 100% of your visitors.
But I choose to aim higher than averages. And while I’m at it, I might as well aim as high as possible.
That’s the first step. Deny thresholds, then get ready to work really hard.
2. Create an A/B test list of new variables
The second step in overcoming your “threshold” is to start A/B testing with renewed passion. I see a common mistake among many in-house marketers. They either lose their energy, interest, or passion in A/B testing.
The only way to increase your conversion rate is by A/B testing. You can’t raise conversion rates without knowing exactly what factors will improve your specific conversion rate.
Remember the following rules about A/B Testing:
- Only one split test at a time.
- Only one variable at a time.
- Test every change — even the small ones.
- Test features at the top of the funnel and at the bottom.
- Hypothesize before you test.
What kinds of tests should you run? Anything. Here are some ideas to get you started:
- Your headline
- Your subheadline
- Your introductory text
- Your font size
- Your font choice
- Your font color
- The location of your CTA button
- The size of your CTA button
- The copy on your CTA button
- The color of your CTA button
- A video vs. an image
- The size of your image
- The subject of your image
- The autoplay vs. non-autoplay on your video
- The structure of your content
- The length of your content
- Trust signals
- Your site navigation
- Your pricing
- Your number of offers
- Your value proposition
- The number of benefits
- The amount of whitespace
- Your return policy
- Your guarantee
- The location of your guarantee on a landing page
- The number of products offered on a single landing page
- A popup vs. no popup
- Live chat vs. no chat.
Any one of these suggestions could translate into huge conversion upticks. Just see CopyBlogger’s article on seven simple A/B tests that boosted conversions.
I’m not saying that any one of these will boost conversions. But you won’t find out unless you test.
You can break any conversion threshold by simply testing, testing and testing.
3. Tear everything down and start a new landing page.
This is my most Grizzly Bear recommendation of all time. I do not recommend it lightly, and I do not recommend it for everyone. I recommend it only if you are in dire straits, if your conversion rates are near negative levels, and if you are considering a career change to professional base-jumping.
Tear everything down and start from scratch.
It could be the best thing that ever happened to your life. (Exaggeration level of previous sentence: Medium to high.)
Here’s why a complete teardown might be in order:
- Your conversion rates might suck for reasons that you won’t discover until after months of A/B testing.
- Your landing page or site might be so bad that no amount of testing or tweaking can change as many variables as you need to change.
- You built your website in 1976.
There are several options you have for a complete tear-down:
- Redesign your entire website. Do you need a new website? Conversions have to do with more than just landing pages, so you may need to do a total redo on the site as a whole. If people keep asking you, “So, when you are going to redesign your site?” this is actually a tactful way of saying “Your website is awful.” If your budget allows, get a new site.
- Redesign a landing page only. The more modest approach is to simply work on your landing pages. There are resources for creating landing pages (Unbounce, HubSpot, etc.) that look good and are easy to test.
Do some basic research on what a landing page is, and how to make one that’s conversion ready. You can take the simple approach, like this example:
And you can also copy templates for more sophisticated landing pages like this:
Be sure to look at lots of examples of good landing pages before you start making your own awesome landing page.
I recommend researching a few of these innovative options to make your landing page as awesome as possible.
- Long-form page. In some industries, more is better when it comes to landing page optimization. The famous Crazy Egg test from Conversion Rate Experts produced a landing page that was 20% longer than the control, and produced a conversion increase of 363%.
- Video. More CROs are discovering the potential of video on landing pages. It may work for you; it may not. It’s worth some research.
4. Change your conversion action.
One of the ways to supercharge a lagging conversion rate is to change the conversion action. Usually, amateur CROs are more concerned about jiggering button size and optimizing the popup.
I take a more aggressive approach. I say that you need to try a new conversion action. The easiest way to raise your conversion rates is to lower the conversion action.
At first, this may sound like cheating. It’s like lowering the high jump bar to make yourself feel like you’re awesome, right?
Lowering the standard of your conversion action is, instead, introducing more users into the conversion funnel.
Every funnel starts wide and ends narrow. If you widen the wide end of the funnel, then you’ll be able to achieve a higher percentage of users who follow through to the narrow end of the funnel.
The way to widen the funnel is to make the conversion action easier.
(Image from SeeWhy. Right arrow annotations mine.)
Here are some conversion actions that you could use:
- Sign up for a webinar
- Enter email address to get a newsletter or email updates
- Download a PDF or whitepaper
- Get a quote
- Download an app
- Start a free trial
- Read the next article
- Chat with a customer service representative
- Order a sample
- Explore more options
- Call a phone number
- Watch a video
- Fill out a contact form
- Like on Facebook
- Follow on Twitter
- Share a link
- Comment on article.
The specific conversion action you choose depends on the nature of your product or service, the intent of your audience and your overall goals.
Experiment with conversion actions until you find one that actually moves people down the funnel, not just makes your CVRs look better.
It’s important to understand that higher conversions don’t automatically mean higher revenue. You need to make clear in your mind the distinction between these two numbers, because they have two different levels of importance. It’s better to have higher revenue levels than it is to have high conversion rates.
Often, raising conversion rates will improve revenue. But not always. Just be aware.
5. Segment your audience.
Audience segmentation means dividing up your visitors into groups based on different features.
Marketing isn’t just about getting a ton of people to view your product. It’s about getting the right people to view your product. Interacting with these people depends on their intent, their goal and their demographic features.
Different people act in different ways in different situations. Your basic tools, like Google Analytics or Facebook ads, allow you to understand what kind of “different people” you’re dealing with, and create experience pages that are designed for them.
I’m a strong proponent of A/B testing according to landing page segmentation, and that’s one approach I recommend. Another approach is to create a variety of landing pages, each designed for specific visitors.
There are several ways to approach segmentation — using keywords, allowing users to make their own choices, and then segmenting your customers once they convert.
Keyword-based segmentation is the best and most logical place to start. Creating keyword-focused queries is a simple matter of targeting the right PPC bids and/or optimizing for the right query intent.
Make as many landing pages as is necessary to target the right visitors.
User-directed segmentation allows the visitor themselves to choose what segment they belong in. A landing page might look like this:
The conversion path then would flow according to the user’s situation and interests. For example, this landing page for the New England Journal of Medicine helps users select the best offer for receiving the journal.
Intuit uses a different form of segmentation on the landing page below. Rather than segment the audience demographically, which is most common, they segment them by where they are in the buy cycle.
Users who are at the top of the funnel may only be interested in seeing it, whereas a more engaged user wants to “Get it.”
Segmentation can create a huge difference in your conversion rates. Rather than simply throw up a few landing pages and hope that visitors will convert, you instead shape a variety of experiences that are tailored for specific audiences.
Your conversion rates can be higher than they are right now. That much is at least obvious. The question is how much higher?
In an article from Conversion Rate Experts, they pose this issue:
Many people ask us if there’s a point at which conversion rate optimization gives diminishing returns. If that point does exist, Facebook, Amazon and Google don’t appear to have hit it yet. It’s astonishing what you can achieve with a sophisticated conversion strategy and a skilled team.
I tend to agree. Conversion rate optimization is a source of incredible power. With the right strategy, you can improve your conversion rates.
What techniques will you use to improve your conversion rates?