# Conversion Rate Optimization (CRO): The Complete Guide (2023)

Updated on November 2022 by Ivan Shumaylov

Here’s everything you need to know about conversion optimization (CRO).

In particular, after reading the article, you will learn:

• What is conversion and how to count it
• “Normal” conversion rates depending on the field (graph)
• How optimization can affect your results
• 6 key optimization strategies + how EXACTLY to apply them in your project

If you are tired of “sawing” more (traffic traffic traffic!) and feel that 80% of the result lies in the fact that you just need to “sharpen the saw” (increase conversion)...

If you want to get more leads, applications, and sales, without increasing traffic, without draining the advertising budget into the pipe…

## What is Conversion Rate Optimization?

In simple words.

Conversion Rate — the number of people who completed a target action on the site from the total number of visitors.

At the same time, the conversion rate may relate not only to the site.

This also could be:

• Number of buyers to the total store visitors
• Number of leads to the total website visitors
• etc.

These are so-called macro conversions (since these are the actions that directly affect revenue/sales).

Also, it could be:

• Number of likes on Instagram post to the total number of views (engagement rate)
• Opened the email to the total number of recipients (open rate)
• Number of clicks on the ad to the total amount of views (click-through rate)
• Number of clicks on the button on the website to the total amount of website visitors
• etc.

These are usually called micro conversions (since these are the actions that indirectly affect key indicators in business).

In simple words.

Conversion is the user's transition from state “A” to state “B” (was a visitor, and became a client, etc.)

Therefore, by optimizing the conversion rate we are increasing the overall efficiency of the business/website/advertising.

But if we break it down into separate components, then it becomes not that complicated anymore.

If we want to increase the overall business efficiency, we need to optimize macro conversions.

To optimize macro conversions, we need to improve its components — micro conversions.

But before we start, let's figure out the following.

## How to Calculate Conversion Rate?

As we’ve already mentioned, the conversion rate is not an absolute, but a relative indicator.

Meaning it’s always some %.

Basically, to calculate the conversion rate, we need to take the number of people who have done the target action and divide it by the total number of people.

Any other macro or micro conversion is calculated similarly.

1. Pick the target action
2. Count the number of people who have completed it
3. Divide this number by the total number of people

For example.

We want to measure the monthly conversion rate of the website (from visitors to leads).

First, we go to Google Analytics (or any other statistics) and look at the number of goals achieved (month period):

In this case, it's 142 people.

Next, we look at the total number of visitors for the same period:

In this case, it’s 5,501 people.

Next, we divide 142 people by 5,501 and get a 2.6% conversion rate.

Similarly, we can calculate this indicator for a day, week, year, etc.

Now that we have calculated the conversion rate of the site, a logical question may arise.

## Why is Conversion Rate Optimization (CRO) important?

Let's take a simple example.

Amazon.

A company that needs no introduction.

The average number of visitors to this site is 2.5 billion people.

Per month (!)

Their turnover is \$470 billion, i.e. ~ \$40 billion/month (according to 2021 data).

Let's make a rough calculation.

Imagine that the conversion rate from a visitor to a buyer is 10%.

10% of 2.5 billion visitors equals 250 million buyers.

It turns out that \$40 billion / 250 million buyers = \$160 average check.

(that seems to be quite similar to the truth btw)

The best part.

Imagine that we’ve improved the conversion rate of the site and achieved a slight increase: from 10% to 11%.

Plus 1 percentage point.

What would that mean?

Now with the same 2.5 billion visitors, we have 275 million buyers. Multiply by the average check of \$160 and we get \$44 billion in revenue.

One more time.

+ \$4 billion in revenue.

Per month.

With the same number of visitors on the site.
With the same products.

And that's only 1 percentage point.

What would happen if we increased this indicator by 20-50-100%?

And by the way.

This indicator is SO important that Amazon itself noted that every 0.1 second of site loading delay, on average, cuts off the conversion rate so that the company loses about \$40 million.

I guess you can ask a question now:

“OK, but how to increase the conversion rate of the site?".

That’s EXACTLY what we’ll cover in the next section.

We will show the key strategies step-by-step so that after reading you’ll be able to implement them yourself and significantly influence the effectiveness of your site.

## Conversion Optimization Strategies

Here we will analyze 6 key “battle-tested” strategies, the consistent application of which will inevitably have a positive effect on the conversion rate of your website.

At the same time, these are not some “hacks” or “best practices” that work occasionally (if you’re lucky enough).

Instead, these are the general principles.

Universal methods that will work as long as your client is a human being (not a robot).

Here they are:

• Speed
• CTA
• A/B testing
• Split URL testing
• Social Proof
• FOMO

Let's take each one apart. In the nuances.

So that as a result you can apply it to your project.

### Speed

Exactly the same as in the Amazon example above.

In order for a person to submit the request or purchase anything on your website…

...first, he should be basically able to get there.

Thank you, Captain Obvious!

But the devil is in the details.

The site shouldn’t just function and load — but do it fast. Very fast.

It doesn't matter how beautiful your website design is.
It doesn't matter how persuasive your copy is.

On average, if the site loads for more than 2-3 seconds, then you already lose more than 50% of the audience.

This is like a general online “etiquette”.
That should be the first priority.

OK, then how to make sure that your site speed is below the “radar”?

Step 1.

Just go to google page speed insights (it's a free service available to everyone).

Step 2.

Step 3.

To do this, Google gives specific recommendations on WHAT EXACTLY needs to be done in your case.

Just implement these changes (or pass them on to your programmer).

Important: please note that there is an evaluation for both the mobile and desktop versions (your first priority is the mobile version)

Actually, that's all.

Periodically monitor this indicator so that it is in the “green zone".

Let's move on.

### CTA

On the one hand, this is a simple and banal thing.

Probably everyone knows that on the website (as in any other material) it is necessary to complete the offer with a call to action.

But again the devil is hiding in the details.

Any offer should, in principle, end with a call to action that a person should take (the next step).

It is important to do this even when everything seems obvious. And explain it in such a way that even a schoolboy understands it.

Example.

OK, the point seems to be clear.

But if I want to learn more, what should I do then?

Call?
Send an email?
Should I find a profile on social media and text there?

Let's even say we’ve figured it out, but wait:

And where exactly can I find the email address?
Are there any in contacts or on the main page?
And if I leave this tab now, will it disappear afterward?

And let's say we finally found it, but:

What exactly to write?
Will they call me back?
Will they instantly do a hard sell pitch?

All this is called “friction".

Your task: to make sure that “the wheel does not slip, but slides like butter.”

The CTA should be:

• extremely simple
• as though anyone instantly understands it
• make it clear what will happen next (so that the user doesn't have to do guesswork)

For instance:

“Leave a request in the form below to receive a detailed calculation. We will contact you within 10 minutes to clarify the parameters.”

Next.

Compare these two sites:

• It seems to be clear on both sites what the companies are doing
• Both companies have the offer with a call to action
• Both CTA's have clear, understandable wording

Having everything else equal, where would you rather leave your contact details?

I bet it's on the second one.

The difference is that in the second case the lead magnet is used as a CTA.

In this case: “Get Project Brochure”.

That is a simple first step that a potential customer can take if he or she is generally interested in your product/service.

A way less friction than:

• Call
• Whatsapp
• “Make an inquiry” (What does it mean? Would I receive more info about a project or you would just call me back to sell it straightaway?)

If you want to buy a Mercedes…

...then you don't go into the salon, and then go straight to the cash register with a credit card.

Before buying, you do a test drive.
Before you do a test drive, you check the price.
Before you get a price, you basically look at the models available.

Every time we take a step back.

“what action do my clients usually make before the actual purchase?”

Its task: to reduce “friction" (to make it easier for a prospect to engage with you).

If you liked a girl, then you don’t propose to her right away and show “reviews from exes” to “deal with objections”...

An example of what you can use as a Lead Magnet:

• Catalog
• Price list
• Brochure
• Case study
• etc.

=>

This is how you’ll be able to increase your conversion rate

=>

Consequently, you’ll be able to dramatically expand the “incoming flow” of potential customers (who otherwise could just pass by).

Let's move on.

### A/B Testing

One of the key principles that directly affects conversion.

By the definition, it’s testing.

Formally speaking:

A/B testing is when part of the visitors (usually 50%) sees one version of the website, and the second part sees another with some modified element.

You can test anything:

• The location of the buttons on the site
• Location and content of visual elements
• The wording on the buttons on the site
• The wording in the site header
• Capture form
• CTA
• The cost of the goods
• Offer itself

(from the less to the most significant)

As a result of this approach, we can reliably, almost with scientific accuracy, draw a conclusion about what needs to be used and what needs to be removed.

This is a key principle that allows you to move not at random, but sequentially and pointwise.

We can assume that:

• The new design will perform better
• The new price will work better
• The new “trick” used by competitors will work for us
• “The red color of the buttons is too repulsive, it is better to use blue”

But the A/B test will show what will actually work.

An important point.

To make a correct conclusion, we need to test only ONE element, while all the rest of the content must be identical.

Example.

If we make a change to the site title, photo, and wording on the button, then as a result of testing we won’t be able to make a correct conclusion about what exactly influenced the overall result:

The title?
Photo?
The button?
Or are they all together at the same time?
Or just two out of three?

General rule: we change one element, and leave everything else the same.

OK, how exactly to conduct A/B testing?

Basically, you can do this in two ways:

Using an A/B testing tool, for example, VWO (just post this service to yourself on the website, and then freely make any changes).

Using Google Tag manager (a slightly more complex, but a free option).

P.S.

This method works great, but it suits better for large projects with big traffic (otherwise AB testing would take a while).

If you have a relatively small project OR you want to double or even triple your conversion rate fast

...then move on to the next point.

### Split URL Testing

The word itself might sound a little scary meanwhile the very principle is absolutely similar to A/B testing.

We also conduct testing and identify what exactly has a better effect on conversion, but here we test the ENTIRE site.

Or rather, the very method of converting traffic into a target action (converter).

(this option is primarily suitable for those types of businesses where lead generation is used, i.e. not “frontal sale")

Example.

Our task: to get the lowest cost per lead (real estate buyers) for the real estate agency.

To do this we:

1. Created a website
2. Launched traffic

To optimize this cost, we need to increase the conversion rate of the site.

To do this, we conduct AB testing, i.e. we try different:

• Photo
• Video
• Formulations
• Etc.

As a result of such actions, the conversion of our website increases, thereby reducing the cost of the lead to \$ 8.

However.

If we want to radically influence the CPL, then we have to test the very converter we use, i.e. the way we convert our traffic.

For example, we split our traffic and lead it to three different places:

1. Website
2. Quiz

Since in this case the way of traffic processing itself is changing, the conversion rate may not differ by a percentage…

...but AT TIMES.

Let's look at each of the options separately.

#### Website

Everything is obvious here — this is when we use the site itself as a converter (where we send traffic).

Pros: Usually, the quality of leads is the best.

In this case, we can strategically place all crucial information that the user needs to make a decision (for example, a catalog of our projects, portfolio, case studies, reviews, etc.). So the visitor who left a request there is more “hot” than the one who has not seen it yet.

Cons: Usually, the cost of leads is the highest.

Obviously in order to convey all the “meanings” we need to increase the size of the content itself, which is why not every visitor will just physically reach the CTA button.

As a rule of thumb: the more pages on the site or the more actions that a person needs to perform, the lower the conversion rate.

#### Quiz

In this case, we are sending traffic to the quiz (I’ll explain it using the Marquiz tool).

In this case, we ask visitors to answer a few questions before leaving a request, and in return, we offer some kind of bonus (lead magnet).

For example:

“Answer 4 Questions & Learn A Quick PRO Tip: How To Get More "High Intent" Leads (Bonus: Quiz Marketing Guide + Try Marquiz.io for free)”.

“Wait, but here, in addition to the capture form, you also need to answer questions — will the conversion rate be even lower?”

Absolutely correct logic.

Meanwhile, the answer is counterintuitive — everything happens exactly the opposite way and therefore the conversion of this method is usually significantly higher than that of the website.

Why?

It is based on one of the basic principles of persuasion by Robert Cialdini: “Commitment and Consistency”.

See how it works.

The offer itself is extremely simple and causes a minimum of “friction” — you just need to click on the button.

At the same time, it causes Curiosity: “Hm, OK, let’s look what’s there, go ahead, surprise me.”

Then the user gets to the first question, the answer to which (strategically) is extremely simple to give.

Something like: “Is the grass green?”

And then comes the best part.

1. Each subsequent question is more and more point-based (which segments the prospects) while being easy to answer

2. At the same time, the number of questions itself is small (usually no more than 4), i.e. obviously won't take long to fill out

And finally, when the user finds himself in front of the lead form, he has ALREADY gone such a way that he just wants to finish what he started.

“Well, okay, since I've already answered so many questions, let's bring it to the end.”

Pros:

• As a result, the overall conversion rate of this approach becomes significantly higher than on the site
• You receive pre-qualified leads (you know what a user is interested in and his approximate preference profile even before contacting him)

Cons: yes, the quality of applications is slightly lower than on the website (because the user hasn’t seen detailed information about our offer, only a basic description)

In this case, we send traffic to the Facebook lead form (Meta).

You must’ve already seen such forms:

A key nuance: in this case, a user leaves a request in the interface and infrastructure of Facebook/Instagram (Meta).

Since a user:

• Does not make an extra click on the site
• The user’s contact info is automatically pulled up in advance (using Facebook’s profile), and the one thing that remains to do is only to hit “Submit”
• The user remains within the Facebook platform (and its algorithms especially encourage such behavior in order to maximize the users’ time spent)

Then usually the conversion rate becomes unprecedentedly high.

Pros: the conversion rate of such forms is usually the highest

Cons:

• This option is only available for advertising in Meta (not available for different traffic channels like Google, YouTube, TikTok, etc.)
• Usually, the quality of leads is minimal here (often people leave a request by accident because their data is already pre-filled)

In reality, in fact, this does not help a lot.

PRO TIP:

It’s much better if you use the lead form as a quiz in the same way as we showed in the previous paragraph:

That is, to ask the same questions as in the quiz, thereby prequalifying the audience + minimizing “accidental” leads

Let's underline the main point once again.

AB testing — a basic principle that allows you to increase the conversion of the site.

However, as a rule, you only achieve +10-30% increase here (for instance, 3% before and 4% after).

But if you want to double or even triple your conversion rate (for instance, 3% before and 6% after), then use Split URL testing or, simply put, fundamentally different ways of converting your traffic.

Let's move on.

### Social Proof

One more persuasion principle. This is what directly affects the conversion.

Question.

“Choosing from the two restaurants below, which one would you rather go to?”

I bet you’d rather go the #2.

A queue creates a queue.
All other things being equal, a person would rather do what others do.

All these statements come from one thing — social proof.

If you want to do 1 action that is almost guaranteed to increase the conversion rate of your site…

...then just use the reviews.

It doesn't matter what niche you have.
It doesn't matter how “specific” your products or services are.

It works smoothly. Like clockwork. Every time.

In response to this, I’m often asked: “OK, how many reviews should I use?”

The principle here is simple: there are never too many reviews.

Moreover.

It works so well that you often even don't even need to use any specific offers or unique selling propositions.

And how many of them there are.

Example.

Which of these hotels would you rather book?

Moreover.

You don't even have to read them — their very number is already enough.

Based on the reviews, a potential client will sell to himself for you. Because “so many people can't simply be wrong”.

You may ask, “OK, what kind of reviews should I use?”

There are 2 key points here.

First.

Your customers’ speech, wording, and slang may sometimes seem strange and inappropriate — however, this is what resonates with LASER's precision with their own kind, i.e. other potential customers.

BAD: a staged, formal, “correct” review
GOOD: live, real, “unfiltered" reaction

Compare it yourself:

Second.

The more visual elements in the review, the better.

And let's go back to the Amazon example again.

On the left is what reviews look like in 90% of Ecom stores.
On the right is Amazon.

Every action they do is for a reason.
It's like an instruction for airplane pilots — every point there is written by the “blood" of previous flights.

GOOD: there is a review
BETTER: there is a review with a photo
EXCELLENT: there is a review with a video

The key task that reviews solve — is to build trust through Social Proof.

The more visual elements in the reviews, the higher the trust.

(roughly speaking, it is possible to fake text reviews, but obviously, it’s way more challenging to do in a photo or video format).

So, let's summarize once again:

• If you want to raise the conversion rate, add reviews
• There are never too many reviews
• The more visual elements in the reviews, the better

Let's move on to the final strategy.

### FOMO

In general, its essence lies in the name itself: Fear of Missing Out.

In fact, this is the same as Social Proof but executed somewhat differently (from another angle).

Here’s the logic.

If there is a queue for the product => it’s popular => it’s good
If there is a low quantity of products left / it was sold out => it’s popular => it’s good

And it also works in the luxury segment:

As a trigger, you could use:

• Discount
• Limited Collection
• Limited Quantity
• etc.

One of the most common, but no less effective methods.

To optimize the conversion rate your task is to say what needs to be done (CTA)…

...and also to give a clear and logical reason why this should be done now.

Not tomorrow.
Not in a week.
Not “a little later".

GOOD: Halloween sale -20 discount%
BETTER: Halloween sale -20% discount up to 11.1
EXCELLENT: Halloween sale -20% discount up to 11.1 (less than 50 items remained)

1. Find all the potential reasons for the buyer to purchase later
2. To “destroy” them

This is our complete guide to conversion rate optimization.

Now I want to hear from you.

Let me know below:

Are you measuring your conversion rates?
Are you optimizing it?
What strategies are you using?
Maybe there’s something that you want to add?

Leave now a quick comment below.

Ivan Shumaylov
Marketer
Share