Conversion rate optimization is inarguably a hot topic for most businesses due to its relationship with sales and general business success.
In fact, a 2015 report by Comscore showed that CRO technologies have established their presence.
In addition to that, CRO has been proven to be successful, especially after Venture Beat survey report showed that only 2 out of 36 companies failed to generate substantive Return On Investment (ROI) with conversion optimization. Additionally, only 5% of the 2938 survey respondents were unable to generate ROI.
But keenly watch this:
Out of the 2938 respondents, only a meagre 173 achieved an ROI of over 1000%.
Clearly, conversion optimization has the ability to skyrocket your profits. However, the big question is why is it that only few people are seeing these amazing results?
Glad you asked:
Truth is marketers are continuously making certain conversion mistakes which sadly, they don’t know about. This is why most websites don’t have a massive traffic problem but are suffering from conversion problems.
The gravity of the situation is seen when reports reveal that conversion rates range from mere 1% to 3%.
In real sense:
Only 22% of businesses are satisfied with their website conversion rate. No wonder in 2014 alone, 85% of marketers planned to focus more on conversion rate optimization.
Where do you start from?
Well, there are so many points you could start from, one being, understanding the conversion optimization mistakes you are making.
In this post, we will look at 5 crucial parts of your business and how your marketing team could unknowingly be making conversion losses through certain mistakes.
To make your reading easier, I have created this table of contents below. You can use it to jump to a specific title.
Download this Checklist of 30 Conversion mistakes you need to avoid Right Now!
Let’s get started:
6 User Experience Mistakes and How to Solve Them
Did you know that 79% of your traffic will search for another site to complete their task if your website’s user experience is poor?
This backs up the fact that 83% of people say that a seamless experience across all devices is somewhat or very important.
It is fair for us to conclude that a bad UX will eventually lead to loses in your business. Most probably, this realisation is the reason 84% of companies expect to increase their focus on customer experience measurements and metrics.
Well, where are the tangible numbers? You ask.
I will give you specific case studies but before then, read this:
Can you believe that in 10 years, a $10,000 investment in design centric companies will have yielded returns which are 228% greater that when the same amount of money is invested in S&P?
In addition to that, ESPN.com revenues jumped 35% after truly listening to their community and incorporating suggestions into their homepage redesign. (Please note that this could be a mistake you make-you don’t listen to your customers. You can learn through ESPN which improved by simply listening)
UX has immense potential to either drive up or drive down your sales. Here are 6 UX mistakes you could be making.
1. Failing To Understand Human Psychology
Humans have been proven to be lazy beings. As Steve Krug pointed out in Don’t Make Me Think, users want to think as little as possible. This means that you need to use human psychology and influence your users’ actions.
What do I mean?
As a user, I don’t have to start scrutinizing your web page to find what is interesting and even locate CTA buttons. If what I see does not help me decide in seconds, then I leave your page.
To curb this:
There are certain psychological tricks that you can use to influence your prospects or buyers. Just briefly, we will look at three important psychological tricks:
The psychology of colors : What colors make people want to buy
People make a subconscious judgment about your company or product within 90 seconds of initial viewing. In actual sense, between 62% and 90% of that assessment is based on colour alone.
Additionally, adverts in colour are read up to 42% more often than the same ads in black and white.
Definitely, you have to wisely incorporate colour in your website.
As you will notice, colours sometimes perform differently. This is likely because elements such as differences in upbringing, person preference and cultural differences among other factors determine our perceptions to different colours.
Generally though, you should learn how to create right emotions and perspectives using colour. For example, orange promotes friendliness, energy and uniqueness while green promotes growth, financial themes, stability and environment.
That’s why I love to think that the psychology of colors is one of the most interesting and controversial marketing topics. To know what colors make people want to buy it is important that you do a/b testing on your website, and use the one that performs best.
To understand better the psychology of colors in marketing for improving conversions, please view this infographic by Hammad Akbar:
As a teaser, the screenshot below shows you what colours your business should use, depending on the type of shoppers:
Now, you ask:
Where has colour actually improved UX?
In 2015, Moz.com wanted to see how it would improve UX in their website, and opted to do split testing to find out colors that make people buy.
The original CTA button was green, which performed well. However, as earlier said, the business knew that the Call to action button could be made better.
Here is the original CTA button that was green in colour and had a conversion rate of between 10% to 15%.
After testing with different colours, it was realized that a shift from green to yellow increased conversions by 175%.
Here is the improved CTA button:
After testing with different colours, it was realized that a shift from green to yellow increased conversions by 175%.
Here is the improved CTA button:
The use of personality in design
What does it mean to design with personality?
In real life, there are people with whom we are close and warm up to and there are also people who are repulsive. This is the atmosphere they exude every time we are around them. Usually, it is these traits that we use to classify people as bad, good or great.
This principle also applies to web design.
The aura your website gives the moment a user peruses through is what will determine if your UX is good, bad or great.
Let’s see how popular sports brand Adidas did this:
During the 2014 FIFA World Cup, Adhemas Batista pitched a very interesting version of the Adidas logo. The games were played in Brazil, a culture characterized by colourful bahia bands. On the other hand, Adidas is a brand that is known for being fun-loving, hip and classy.
Where these aspects were combined, it resulted to a logo deep in personality and culture.
See the image below to see the image of the logo:
The idea floated in this pitch can serve as your lesson on how to creatively incorporate color in your design.
While you focus on using your brand personality to improve user experience, remember to also consider your user personas, to further improve your CRO strategy.
2. Poorly Designing Customer Feedback Forms
Customer feedback forms could be for different purposes such as lead generation or collecting suggestions. Generally, these forms are drafted as questions.
How do you do this?
Do you include all the questions you ever wanted to ask your users in one form?
Do you give too many options for one question? Do you ask your respondents to create an account before submitting their form?
Do you love very long feedback forms?
If you answered ‘yes’ to any of the above questions, then you are making a BIG MISTAKE!
Since there are different types of forms, we will focus on one of the most popular ones-The Contact Form, which sadly, converts at a meagre 3%.
Why does it have a low conversion rate?
In my view, collecting contact information should be a very simple and short process. After all, you only need to generate leads then from that point, work on the rest of the conversion process.
In fact, if you mess up at the lead generation stage, how do you expect to increase your prospect-client conversion rate?
Sadly, some people think otherwise. Some prefer long forms to collect as much information as possible.
But, did you know that?
It has been showed that asking for a telephone number (or implying that a user will be called) causes a 5% drop in contact form conversion rate, asking for street address causes a 4% drop while asking for age causes a 3% drop.
What does this mean?
There are certain fields that you need to remove from your contact form in order increase chances of a higher conversion rate.
Let’s look at what Expedia did for example.
The illustration below, which is a mock-up of the experiment, shows an old form and the new form.
In a second case study, we see that Blivakker.no , Norway´s leading online beauty shop with approximately 20,000 visits per day, increased customer registrations by 11% after removing 3 fields in the contact form.
The red marks in the image below shows the fields that were removed to improve their contact form design:
How do you create best contact forms? Here are a few tips to improve your contact form design:
- Kill the captcha
- Use the right form layout
- Don’t ask for phone numbers
- Focus on just three fields
- Optimize for mobile users
3. Not Using Visual Content Marketing
While it is okay to use text to talk to your site visitors in an effort to entice them to buy, you need to ask yourself,
How effective is a text-only strategy when the average human attention span is 8 seconds, less than that of your aquarium’s tiny goldfish?
This means that you need to use more than text to capture and retain your audience’s attention. Visual elements such as images, infographics, illustrations, memes, presentations and videos will definitely improve user experience.
- Coloured visuals increase someone’s willingness to read a piece of content
- Colour increases your audience’s attention spans and recall by 82%
- Information can be located 70% faster if it is in colour
- Colour improves brand recognition and survey participation by 80%
No wonder content with relevant images get 94% more views that those without relevant images.
When people hear information, they are likely to remember only 10% of that information after three days. On the other hand, when the same information is paired with an image, people will remember at least 65% of it.
To be effective enough, ensure that you optimize videos for mobile phones. In 2015 for example, mobile video plays accounted for 44% of total video plays.
The trend in the image below shows us over 844% increase in mobile video plays since 2012, an indication that optimizing videos for mobile phones will help improve user experience.
4. Keeping Your Users In The Dark
Have you ever visited a website that kept loading and you didn’t know what was going on?
What was your reaction?
You most probably got annoyed and left. This is what you don’t want happening to your own site.
Well, unless your website is super-fast and the page load time is undetectable, you need to find a way to keep your audience engaged while they wait for a webpage to open.
One tip is informing the user what’s going on as illustrated in the image below:
From the example above, you can tell that the user wanted help finding a hotel. To improve this person’s experience on the website, the site owners ensured that the user was kept informed while the page was loading to find a hotel.
What are the best ways to keep your users interested?
It is simple:
- Use loading sequences
Instead of loading the page as a whole, you could load in it parts in three easy ways.
- Use multi-part transitions whereby different parts use different transition effects to load
- Use skeleton screens where the skeleton screen is appears fully then content is slowly loaded into it
- Load a webpage in sequences. In sites like Facebook for example, your chats sections, homepage contents and adverts load differently.
- Confirm progress immediately
To explain this point, we will use the example below:
From the image above, the user is warned not to click twice. But, how we s/he knows that the first click was successful?
Instead, show progress like Stripe does:
As you can see, the client pays $25.00 for the order. Then, as the image below shows, the business confirms that the client has indeed paid, by showing a tick.
5. Having a Non-Responsive Design
Simply put, a responsive website design is one which adjusts from one device type to the next, ensuring that users on all these devices get the best user experience tailored for them.
The image below shows you an example of how your typical corporate website should look across devices of different sizes:
So, what are the disadvantages of a non-responsive design?
40% of people will choose another result if the first one they land on is not mobile friendly. Remember that 60% of internet access is occurring on mobile devices and 4 in 5 smartphone users access retail content on their device monthly.
How do you ensure that you have a responsive website design?
A responsive design should at least have these 3 features:
- Adaptation to all devices
- Open support for all browsers
- Use responsive front-end frameworks such as Foundation and Bootstrap.
6. Using Carousels
One of the biggest questions marketers have asked themselves is if they should include a carousel in their website.
Also known as slider banners, image sliders or content slider, carousels are bad for UX and kill conversions. Why?
If you have encountered them, then you sure know how hard it is to read content that is moving fast or even get the slide to stop.
To prove this, eye tracking heat maps were used to see where people clicked on a webpage. From the image below, you can see that the carousel part of the webpage receives very little attention if any.
Well, how does this affect conversion? You ask?
A test conducted by Notre Dame University showed that only 1% of the total visitors clicked through from the carousel and majority of these visitors, 84%, interacted with only the first slide of the carousel.
Now, question is:
Suppose the business had an offer located in the second part of the slide, how many people would click though?
To answer this question,
The image below compares 4 different sites and the carousel click-through rate. You will notice that the click-through rate reduces with subsequent slides.
Instead of using an image slider, consider the following:
- Convert each slide into a separate, targeted homepage for specific site visitors
- Focus on your best offer and use that on your homepage
If you must include one, at least have your important information in the first slide. After all, people get less interested with subsequent slides. In addition to that, use a manual slider as opposed to an automatic slider. That way, you user is in control of the slideshow.
6 Call to Action Mistakes and How to Avoid / Solve them
1. Over-selling and Under-delivering
Some businesses are fond of promising the sun, moon, stars and entire galaxy in their call to action and then fail to deliver.
Well, you might argue that humans thrive on promises. But, did you know, as Temkin Group reported, customers are more likely to share bad experiences more than good experiences?
This means that if you state in your call to action copy that someone will get a free report and software trial, you MUST deliver just that, instead of offering either of the two.
2. Displaying Unclear CTAs
Out of 200 small business website evaluated, Online Marketing Coachfound out that 70% of them failed to display clear CTAs for anything on their homepage, including newsletters, specials, interactive tools, how-to-guides and demos.
So, how do you create a clear CTA you ask?
Generally, this is in the design of your CTA button and text.
An unclear CTA has the following characteristics:
The Call to action phrase is vague
One of the principles of creating an effective call to action is to ensure that your site visitors are not confused by your CTA text.
Imagine asking a potential lead to ‘click here’ or ‘submit’ without telling them why they need to do so.
Different texts can be used in CTA buttons. Below are some best call to action phrases:
- Start my free trial
- Get my free ebook
- Register now
- Sign up
- Install free trial now.
Please remember that these call to action phrases sometimes perform differently so it is important to do a/b testing to know what text works best for you.
For example, some studies have showed that landing pages that use ‘submit’ as their CTA text have lower conversion rates than other types of text.
Do colors of call to action buttons affect the conversion rates?
In the case study below, which was presented by copyblogger, we will see a comparison of different CTA buttons and their performance.
This was the original copy -A black CTA button against a grey background. (Notice that the fonts are small and the CTA button is small)
A second variation with bigger font and green button was also used.
The third variation, which had the same font as the green button and was orange in colour was also tested.
Compared to the original version, the green button had a 81% increase in conversion while the orange button had 95% increase in conversion.
3. Placing Too Many CTA Buttons in One Page
Sometimes we tend to think that the more choices we give people the higher the chances that they will choose something.
Have you listened to the paradox of choice lecture?
If you have, then you understand the kind of paralysis people experience when they are forced to choose out of a wide variety of choices.
This principle applies to your website too. Placing too many CTA buttons in one page will only lower your conversions.
4. Not testing Your CTA
When it comes to conversion, be sure that not all tips will work in your website as expected. This is the reason why we always encourage you to test the performance of different aspects in your business. The same of course applies to CTA.
Failure to test CTA could be causing you a conversion loss.
An anonymous Danish company knew the importance of a/b testing and decided to test its CTA placement.
The business had a long landing page from paid search ads that sought to increase sign ups to one of the subscription services.
Initially, the CTA button was placed on top of the very complex landing page and the designer though that this position would make the business seem too aggressive, consequently forcing the customer to make a decision before understanding what the page was about.
After running a/b tests with a sample size of approximately 100 conversions and 98% statistical difference, a new page with new CTA position was found, with a conversion increase of 304%.
5. Not Conveying Value
Sometimes, your users need to know more than one element of your offer. For example, one would need to know the value you are offering other than a free software trial.
To prove this, an experiment on Friendbuy was carried out.
The first CTA copy simply offered a free trial of the program as demonstrated below:
The second copy included a testimonial:
The third variation which reduced any anxiety caused by credit cards and also included another key benefit had 34% increase in conversions compared to the original CTA copy.
6. Not Leveraging Numbers
One of the best ways to lure people into taking the action you need them to take is by showing social proof. One option is leveraging numbers in your call to action copy.
These numbers will serve as some form of testimonial.
Let’s see how TruConversion did that:
6 Landing Page Mistakes and Their Solutions
1. Not Including Customer Testimonials
Before making a purchase decision, 61% of consumers will read an online review. If they find an online review on the product, 63% of these visitors will be more likely to purchase.
Site visitors who interact with both reviews and customer questions and answers are 105% more likely to make a purchase and even spend 11% more than visitors who don’t interact with the testimonials.
The image below shows us the original page and the new page that had a testimonial attached.
The result was 50% increase in conversion.
To learn effective ways of using testimonial, read this post.
2. You Don’t Have A Trust Symbol
48% of shoppers use trust badges as one of the factors they consider before making a purchase. In their view, trust seals give them a great sense of trust, increasing their chances of buying.
The illustration below shows us how trust seals compete against other conversion factors.
To prove this:
BagServant, an Ecommerce site, won a notable sector award and decided to include the award symbol in their landing page to see how it would influence conversions.
From the image below, you can see the trust symbol circled in red:
The business was able to achieve 72.05% increase in conversions.
Do you know why?
The trust symbol was a sign of credibility as well as quality. Though simple, such an award would reduce the chances of a customer hesitating on basis of fear of getting bad services or goods.
Additionally, the customer would feel safe transacting online because the business has been proven to be legit.
What this means for you?
3. Not Clearly Explaining Who You Are And What You Do
Your landing page should be your business pitch to potential leads and customers. This means that one should find all the necessary information about your business within your landing page copy.
Failure to fully present yourself will only lead to a low conversion rate like Groove experienced.
The business suffered from a low conversion rate of a mere 2.3% to the point that it thought it would go out of business.
An analysis of the landing page, was able to quickly tell that the landing page, which is illustrated below, was not clear on what the business was about and what problem it was solving.
From the image above, you can see that the headline was only talking about the product. At no point do we see the problem this business sought to address.
The headline is followed by a testimonial which, instead of stating the problem, only compares the product to competitors.
This copy was later changed to a more effective one that doubled the conversion rate.
Just like Groove did to improve themselves, here are a few questions that you need to ask before you create your landing page copy.
- Why do we exist?
- What problem are we trying to solve?
- What difference are we making in our customers’ lives?
- Who are our most ideal customers?
What makes us better than our competitors?
4. Focussing Too Much On Videos
I know that you are usually advised to include images and videos to your landing page in order to increase conversions.
Yes, it is true that videos increase conversions by 80%.
Do you know that only 10% of your users are watching your landing page video?
Well, that’s why we say continuous testing and improvement is what will improve your conversion rate. Simply don’t include a killer video and sit hopefully waiting for skyrocketing conversions.
Sometimes images outperform videos as BrookDale Living found out.
Their initial page was full of text and no visual attractions. Two variations were then created, one with images and one with a video as seen below:
After running the test on over 30,000 visitors for two months, the business realized that images outperformed videos in conversion.
The image version increased conversion by 3.92% while the video version increased conversion by only 0.85%.
In terms of revenue, the simple 3.92% conversion increase with the image version helped the business realize $106,000 additional monthly revenue.
Take home lessons:
It is shocking that images can outperform videos given the numerous reports showing that videos are better.
After an audience demographic analysis, which is shown in the image below, it was evident why videos did not perform as expected.
The audience had no college education and browsed from home, causing them to be mindful of their internet charges.
For your business, the key on what to use depends on:
- The type of online persona your target market is. (In this case, the business had spontaneous type of audience which is usually ready to buy without much marketing)
- How established your business is. Most of the time, established businesses don’t need videos to further convince clients.
5. Excluding Product Filters
Well, in your main site, you probably have product filters to boost conversion.
But, do you include product filters in your landing page as well?
The original landing page copy had text and sample product images. The variation included the same text but allowed a potential customer to search for desired product using filters as demonstrated below.
As a result, the business achieved 76.1% increase in revenue and 26% increase in conversions.
6. Giving In-House Testimonials Only
It is good that you have included testimonials. But, who has given these testimonials?
By common sense, we know that you cannot shoot yourself in the leg in business. Giving yourself a bad review will reduce conversions faster than third-party bad reviews will, in my opinion.
So, why would a prospect believe your staff members when they already know that their testimonies are biased by default?
If you must include in-house testimonials, balance them out with third-party testimonials from real customers.
You can also use other means of getting customer testimonials such as Facebook reviews, YouTube comments and LinkedIn recommendations.
6 E-commerce Design Mistakes and Solutions
1. Not Giving Detailed Product Information
The easiest way to post a product description on your ecommerce website is to paste the exact manufacturer description. Though easy, this idea could be killing your conversions
Taking time to give a detailed product description actually increases conversions.
So, how how to write a good product description that converts?
- Use emotional language which has been proven to have greater influence on customer than the advert content.
- Use descriptive words when writing product descriptions
- Talk to the customer, writing in a conversational tone
- Describe the results of the product not the function
See how this product description example uses all these elements:
The product description tells me exactly what the bed can do by directly talking to me. I can only imagine the feeling of a good night’s sleep and this product description trigger emotions by promising comfort and personalization.
2. Not Indicating Steps In The Checkout Process
Sometimes during the checkout process, some websites could ask you to ‘submit’ after every form, without indicating how far into the process or where exactly you are.
Initially, users were asked to ‘continue’ with the process. The business went ahead and tested different ways of wording this CTA and found out that ‘review’ had the best conversion as demonstrated below:
As you can see, the business increased checkout funnel conversion from 30.5% to 39.4%.
3. Missing Your Contact Information
In a test to find out the importance of including contact information in your website, it was discovered that 53.96% of sign-ups originated from the home page variation with the phone number while 46.04% of sign-ups originated from the original home page without a phone number, showing a conversion rate increase by 0.5% where contact information is included.
The image below shows the pie chart comparing the conversion rates with and without the contact information:
Where money is involved, in my view, I need to know who to contact in case something goes wrong.
Have you ever read an online review stating that a particular website is a scam and a reviewer even goes ahead to say that that particular website did not have contact information?
Such a testimonial of course further drags your company through the mud.
You don’t want this happening to you.
To avoid this,
Ensure that you include contact information to build trust with your prospects. Additionally, these contacts should be real and active, providing prompt customer support.
See how Zappos does it:
It even promises 24-hour customer service and shows support for an additional language, Spanish.
To learn how to increase your contact us form conversion rate.
4. Lacking A Proper Value Proposition
Why should I buy your product and not your competitor’s?
This is the question your prospects will ask and they expect you to answer it by drafting a great value proposition.
In this caption, we see how Peep Laja describes the importance of having a great value proposition.
To create a great value proposition:
- Use an attention-grabbing headline
- Have a subheadline to briefly describe the business or product
- 3 bullets points giving specific reasons why you are the best option
- Include visual stimuli such as images to show the product
- Avoid hype and business jargon
5. Not Showing the Product Price
While it is reported that part of the reasons for shopping cart abandonment is price related, it has been showed that including product price can actually increase conversions.
Variation A, as depicted below did not include the price or the cost of the product
The second variation included the product price in the landing page copy as seen below:You will also notice that the price has been made conspicuous.
The result was 100% increase in lead generation.
6. Poorly Choosing a Form Layout
Here is how the horizontal layout looked like:
The second variation included the product price in the landing page copy as seen below:You will also notice that the price has been made conspicuous.
The business realized that vertical layout, due to its slim and compact design, helped visitors see important information quickly. No wonder the vertical layout had 52% increase in conversion.
The choice of layout is dependent on your e-commerce website design. Remember to test different layouts and choose the best.
6 A/B Testing Mistakes and Solutions
1. Not Split Testing At All
It is sad that only 44%, instead of 100%, of companies use spilt testing.
How will you know what to improve and what changes have the best results if you haven’t tested?
In case you don’t do a/b testing, then this is a very grievous mistake. Actually, one accurate measurement is better than a thousand expert opinions, as Admiral Grace Hooper says in the caption below.
So, why split test?
- Best practices always don’t always work
- You challenge conventional marketing wisdom
- Small changes could lead to big wins. For example, a small change in a bad landing page did not improve its quality much but increased conversion by 76%.
2. Split Testing Without a Plan
Now that you know the importance of testing, how do you go about it? Just like any other problem-solving process, you need to have a strategy.
People who do split testing without a strategy miss out on conversion big time. Instead of guessing what to test and why, you should have a strategy in mind and know what your aim is.
To start planning for your split test, you could ask yourself a set of questions:
- Where are users confused in the funnel?
- What’s our hypothesis for the test?
- Is this test the best test needed to impact results?
- It is necessary to test this element? What is the proof?
- What should we learn from this test?
3. Running Just One Test
Sometimes split testing can be time consuming but the results are definitely worth it.
Due to laziness and time limits, we tend to run only one test on our sites.
But, have you ever asked yourself how testing more than just one element would improve conversions?
To help you see the importance of this, we will look at one case study.
ABNRegistrations.com increased conversions by 25%
Conversion is a process, no wonder we talk about the conversion funnel. Generally, this means that all the elements in your website have an effect on your conversion rate.
What this means:
Testing just one element is not enough. You need to optimize every element in your website just like ABNregistrations.com did as illustrated below.
Test 1: Bullet points
The business initially had their three-step ordering process described in text as shown below:
A small tweak was implemented by adding green ticks next to the steps as seen below:
This small change increased conversions by 2%.
Test 2: Headline
Headline 1 which is showed above promised ABN provision within 24 hours. On the other hand, headline 2, as shown below, promised ABN provision in 4 hours.
The result was a 10% increase in conversions.
Test 3: Order process
The order form was also redesigned to look better.
This whole process resulted in a conversion increase from 30.88% to 38.88%,which was an increase of 25%.
You need to be careful especially when you decide to test too many elements due to the risk of encountering false positives. This is because even at 95% statistical difference, the likelihood of a false positive is 88% when you do many tests.
4. Testing Too Many Elements or Variables at A Time
We have agreed that you need to test more than one element. But, how do you do it?
Do you test subject lines and different CTA buttons at the same time? If you do, be sure that it is impossible to determine which of the two was the success factor in Website conversion optimization
To avoid this,
Corinne explains that limit your testing to one variable at a time. This way, you will better understand what content strategy or website changes are more effective and why.
5. Using an Inappropriate Sample Size
One of the biggest questions in a/b testing is how to measure the appropriate sample size.
Now, we will not get into the mathematics of it and bore you. I will just give you a general idea of how to do this.
First, let’s observe the image below:
Out of 44,654 visitors, do you think that 100 visitors per variation are enough to call an a/b test? Is 100 representative of 44,654 visitors?
By common sense, of course not.
So, what do you do?
Simple, get a/b test sample size calculator online and fill the fields as stated. Alternatively, you can read this post on exact steps to follow when choosing a sample size for your a/b test.
6. Lack of Understanding of False Positives
It is common knowledge that a statistical difference of 98% usually is enough to conclude on a test. On the other hand, a statistical difference of 50% tells you that either of the two variables could work.
Therefore, we usually conclude on tests with 98% statistical difference.
Now, as we said earlier on, when running several a/b tests, the chance of getting a false positive even at 95% statistical difference could be 88%.
To prevent this from happening, there are three concepts you can apply to get true positives.
- Design of Experiments (DOE), which is used in one or a series of controlled tests to understand the influence of one or more elements in a complicated environment.
- Statistical power, which is all in your sample size, the bigger the sample size, the bigger the power and the lesser chance of a false positive.
- Regression to the mean, whereby a test could show a strong initial result due to novelty effect rather than a better user experience. The more you do the regression analysis, the lower your chances of getting a false positive.
You now know some of the conversion mistakes you are probably making. Instead of simply having a light bulb moment and do nothing, I advise that you seriously run an in-depth analysis of your website and business.
How is your user experience, how is your landing page as well as CTA and other elements?
To properly make an analysis, please use the all-in-one analytics tool – TruConversion and note down key areas that need improvement. Develop a conversion strategy and begin working on the mistakes we have mentioned in this post.
As always, we would love to hear from you.
What trivial or grievous conversion mistake have you been making and how did you correct it? Let’s engage each other in the comments section below.