This is the 10th week I have learned. I am very excited to write about my study progress here From About CXL CRO Course.
First of all, it must be visually simple. People are not willing to ask for complicated things. Use more familiar layouts-that is, to think about the problem from the perspective of consumer behavior.
Then have a sense of visual hierarchy. For example, if you need to zoom in over there, you need a color difference over there. Put the most important information on top.
The third is to improve concentration. Don’t put stock pictures, you can use some people’s life photos.
The fourth is that there should not be too many paragraphs.
The fifth is one operation per page to maximize the product image.
You have just started working on the website to increase the conversion rate. Where to start?
The typical mistake of a rookie is to use the spaghetti method-start throwing things on the wall to see if it sticks. People start with random A/B testing and then wonder why no gains have been realized. This kills emotions and motivation.
Instead, follow the step-by-step process. Conversion optimization is a process and should always be described as such. How is this going:
1. Determine business goals: determine the indicators we want to increase.
2. Business questions: ask as many questions as possible from the user’s perspective
3. Data collection: Set up related data collection and tracking. Waiting for data input.
4. Insight stage: Analyze all the different data sources, determine the model, and draw conclusions.
5. Identify the problem: Refer to the data to find problem areas that need attention.
6. Hypothesis: Make the correct hypothesis for the test. Sort them according to the likelihood of hypothesis and ease of implementation. This determines the ones we need to test first.
7. Design: Based on assumptions, you can develop wireframes for treatment. Wireframes are mainly used as communication and planning tools. Once completed, the wireframe will be passed to the designer, who will convert it into a design screen (alternative layout/content of the page on the website).
8. Technology integration: This is an essential evil — a new design screen needs to be integrated into the website. If the change is small, it can also be done through the testing tool itself.
9. Test: We will conduct appropriate controlled a/b tests in accordance with scientific methods, and then run each test until statistical significance is reached.
10. Post-test analysis: Evaluate the data to discover how different changes to the target are implemented.
11. Learn and improve customer theory: We analyze the test results (after correct completion, each test is a source of learning), gain new insights, and propose new hypotheses. Through each test, our customer theory should be improved.
12. Document: Archive your test results and learning materials.
Go back to step 2.
This is what I call the conversion cycle.
How long does a cycle take?
There is no answer. This may take 2 weeks, a month or 6 months. it depends
1. The complexity of the website (simple landing page and huge e-commerce website),
2. The data we have (how much data still needs to be collected and how long does it take to process the data),
3. And website traffic (this determines how long the test must run to reach statistical significance).
Without affecting the quality of work, the sooner you are, the better. The more experiments you can conduct in a 12-month period, the more content you will learn and the more benefits you may get.
When will conversion optimization end?
The correct answer is “never”. You can always do better than last month.
The world is always changing, your business is constantly evolving, and the competitive landscape is evolving. New technology, whatever you want. The change is the name of the game, you need to keep up with the trend. Conversion optimization is part of it.
Are you always stealing ideas from other people’s websites and putting them on your own does not work? -That’s because other people’s ideas are wrong on your side. We need to better tap the potential needs of users!
What is the conversion rate optimization test: It is about ideas, not tests! Many people do a lot of tests around a lot of ideas. But what has the most influence on our value unit is our thinking!
Don’t test content a lot, because some content doesn’t need to be tested!
We have to learn: 1. How do we view data. 2. How to interpret the data. 3. How to know that these data are correct! Which data is useless.
CRO personnel only need to learn to do three things: 1. Collect questions. 2. Organize ideas. 3. Find a solution (high-quality ideas)
How to manage ideas: build a database of ideas. ②. For the idea, if I did it, what would I expect to be, and how would this be measured. ③. Grade ideas: confidence, simplicity, and influence
How do you know whether to do this idea-confidence? Data is the source of insight. :①. Google this idea and how many people support it. ②. Market research. Consumers mainly care about things. ③. Consumers care about the price, you always have to display the price, the price is suitable for the explosion range. Consumers care about trust, then you can find JD power and some credible review agencies to do it. ④. Analyze the most important data ⑤. User test-focus group website: www.userlytics.com It can be a click-through rate test, a comparison chart test, or a questionnaire ⑥. Artificial intelligence analysis with User intelligence tools: www.tableau.com
A/B test requirements:
1. Sufficient data
2. Data measurement requires a certain amount of time to change, so it takes two weeks, four weeks.
3. The data is the present, not the past
4. This is a quantitative analysis
5. Is your client or potential client
6. They want to be unknowing.
What is conversion?
The conversion converts X to Y. Website visitors are converted into buyers or potential customers. There may be many other things.
When someone buys something from us, we call it a conversion. We have converted visitors into customers (conversions of all e-commerce sites). Software-as-a-service companies may try to convert visitors into free trial users. That is also conversion. For some companies, such as advertising agencies or construction companies, the generation of potential customers is the most important conversion a website can provide. Therefore, they want to convert visitors into potential customers.
Therefore, when someone says “My conversion rate is 5%”, we don’t know what he is measuring. Visit to lead? Visit to buy? Converting free trial users to paying customers? All of these are of compliant behavior.
Every company needs to figure out what the main conversion they want to measure is. In order to understand the effectiveness of the website, we measured its conversion rate. Conversion rate is the number of conversions divided by the number of visitors (in a given time period) * 100.
So, if you have 5000 visitors, and 20 of them have taken action (such as buying something), your conversion rate will be:
20/5000 * 100 = 0.4%
Conversion rate ≠ revenue
The last is income. Your conversion rate may drop, while your revenue will increase at the same time. It is not uncommon to conduct A/B testing on e-commerce sites, where variant A has a lower conversion rate than variant B, but it brings in more revenue. This is what we want.
What does the conversion rate depend on?
The conversion rate depends on many factors.
* The relevance of the quotation/traffic source. When you offer shampoo to a bald person, you won’t sell a lot. If you provide allergy relief to people with allergies, your odds will be higher.
* Relationship with visitors. If the visitor knows you and trusts you, it is much easier to sell him. Therefore, all your content marketing and relationship building efforts are very important for conversion.
* Business vertical. We need certain products more often than others.
* Product Cost. Compared to selling a Ferrari 458 for $229,825, the conversion rate of a “Wash&go” shampoo for $2 will be higher. The conversion rate for more costly products is usually lower. If the product you sell is free (for example, you provide a free pdf report in exchange for their email address), you can potentially get a higher conversion rate.
* Cost/value ratio. The perceived value of your offer compared to the cost (even a free offer-sending you an email is a transaction).
* Copywriting. Words are important. What you say and how you say it can change everything.
* Design, U SER experience and usability of the website.
Therefore, when we compare two e-commerce stores and their conversion rates, it is not a real apple-to-apple comparison.
Measuring conversion rates: the right and wrong ways
Wrong method: At the end of the month, you divide the total traffic by the number of sales. Any manual measurement is wrong. When you get an “average” conversion rate, it doesn’t tell you much.
The right way: You need to use analytics tools to measure conversions. In this way, you can measure the effectiveness of each segment (for example, new and returning visitors), traffic source, device, country, etc. You can make observations and draw wise conclusions to optimize your website.
What is a good conversion rate? How much should I fight for?
The only correct answer is “better than last month.” not kidding. A reasonable goal is to figure out how to increase the conversion rate by 5–10% per month (relative increase, not absolute increase).
Of course, there are “averages”, but they are not really useful. It is too contextual. The best benchmark is yourself.