How to use data operation methods to improve article conversion rate

Nowadays, many products need to be sold on official accounts. Official accounts are now one of the important channels for promoting and advertising many products, and the most important thing is the conversion of official account articles (bringing goods). So how can we increase the sales of articles in public accounts? This article uses a case to tell you how to quickly increase the sales of goods sold on public accounts through data operation methods and ideas.

Take the sale of a course I managed with a customer unit price of 169 yuan as an example. This course is about vocal music, that is, singing. In the end, the article was read 3,000 times, 200 courses were sold, and the order conversion rate was as high as 15%, achieving a revenue of over 30,000 yuan for a single article.

We didn’t spend a penny on promotion, and it was accomplished by pushing messages through a public account with less than 50,000 followers and 6 WeChat Moments. So how can we achieve this through data operations?

Step 1: Calibration

Calibration is the process of determining a core data indicator for an operational activity or project. Usually, in order to find the core data indicators, we must first clarify our operational objectives.

For example, for sales-oriented public account articles, the operation purpose is generally to sell goods (sell courses). At this time, the core data indicator is the final sales volume of the article.

The significance of clarifying the final sales volume is that all our subsequent operational strategies and behaviors must be firmly centered around sales volume and must not ultimately deviate from our core indicators.

Step 2: Modeling

Modeling , establishing the relationship model between core data indicators and secondary and tertiary indicators. Break down the core data indicators into secondary and tertiary data indicators that can clearly guide operational strategies, ultimately achieving the goal of data-guided operations.

So how do we break down the core data indicators?

The key method is to sort out the user's entire path. Determine the data indicator for each path step, and this indicator is the second and third level data indicator we want.

For example, the user path of this public account selling course articles is:

It is obvious that the secondary indicators are the number of article readings and the number of visitors to the products, while the tertiary indicators are the total exposure of the official account articles, the number of article clicks, the number of clicks on the products in the article, and the number of clicks on the purchase button. At this point we can establish a correlation model between core indicators and tertiary indicators.

Assuming that the core indicator, that is, the number of orders is X, the exposure is A, the article opening rate is B, the product click conversion rate is C, and the order conversion rate is D, then we have come up with the formula model for the number of orders X=A*B*C*D.

Step 3: Attribution

Attribution , sort out all factors that affect the third-level indicators. After knowing what data the number of orders is related to, our focus is to improve the data in each link. The factors that influence these data are obvious, and this step is also the most important step in the overall data operation process.

First of all, A is the exposure of the article. This data is basically fixed and it depends on the overall resources of your WeChat ecosystem. For example, how many followers does your official account have? Do you have some private WeChat accounts and WeChat groups?

This time, our exposure number is only 6 WeChat accounts, each of which has about 2,000 friends, and the public account has less than 50,000 fans. Then the three data that can be changed are B, C, and D.

B is the opening rate of the article, and the opening rate depends on your title. In addition, the opening rate in the circle of friends also depends on the wording of your recommendation.

C is the click-through conversion rate of the product, which actually depends on the position of the product in the article and whether the display of the product is obvious and attractive enough.

The last thing is the product order conversion rate D. This idea depends on whether your product details page is more attractive to users. At this point, we clearly know which paths or details we need to optimize in order to increase the number of orders.

Step 4: Test and launch

After testing and launching , we have identified all the factors that affect the third-level indicators. The next step is to test and determine our operating strategy and all the details. The first step is to improve B, which is the opening rate of the article. Our focus is to test the title with the highest opening rate. First, we came up with 5 titles, and then asked users to vote and select the top two titles with the most votes.

Then we made it into a permanent link and posted it to two WeChat Moments with more than 1,900 friends each at the same time. We found that the second title had a significantly higher opening rate.

At this point we have determined the title of the article in the official account. That is to say, the maximization of B is achieved. The second is to improve C, which is the click-through conversion rate of the product.

We still tested and compared two methods, one of which was to add a red button for clicking on the product in the text, and the other was not. As a result, C increased from 10.38% to 19.38%, nearly doubling.

As shown below:

The last thing is to improve the conversion rate D from product visitors to final orders. This data mainly depends on your product details page. Whether the details page explains the highlights of the course and solves the user's trust issues and is attractive enough to users to place orders. Another thing is whether the entire path to the final purchase is smooth and simple. So far we have completed the complete data operation method from calibration to modeling to attribution and testing. In the end, I completed 200 orders and generated 30,000 yuan in sales revenue.

The data comparison is as follows:

Step 5: Review

Review: No matter how well prepared you are before the event, no matter how perfect you think the details of the event are. After the entire activity is launched, there are always some details and strategies that can be optimized, and this is the charm of operations, that is, you can always do better than before.

Each activity should be optimized more detailed and better than the previous one. Review is an essential step. After the project is completed, a comprehensive and in-depth review of the entire project is conducted through data analysis reports and project logs to find some details and strategies that can be further optimized.

For example, this time we can actually promote user forwarding and activity through a lottery in the comment section of the article, strengthen the stickiness between public account users and public accounts, and will not weaken the connection between users and the public just because it is a sales article.

Finally, let’s summarize the methods of improving the conversion rate of public account sales through data operations:

• The first step is to determine the core indicators
• The second step is modeling to find the correlation model between the core indicators and the third-level indicators.
• The third step is attribution, sorting out all the factors that affect the third-level indicators
• The fourth step is to test the launch and optimize all details and strategies according to the influencing factors to maximize each third-level indicator.
• Finally, after the project is completed, the entire process is reviewed to find details and key points that can be optimized again, ultimately achieving the goal of data-guided operations.

Author: Operation Weekly

Source: Operations Weekly