How to build a funnel map to improve product conversion rate?

Traffic is becoming more and more expensive. Instead of continuously increasing users by purchasing traffic, it is better to increase product conversion rates and maximize the value of traffic. The following are methods I have summarized to improve overall conversion rates based on the paid video products I have made.

As traffic becomes more and more expensive, it is becoming less and less cost-effective to increase users and thereby increase revenue by purchasing traffic. Under the premise of limited traffic, improving product conversion rate and maximizing the value of traffic is the most cost-effective way. The following are methods I have summarized to improve overall conversion rates based on the paid video products I have made.

When talking about conversion rate, we must talk about the funnel model used in daily product and operation to analyze the conversion status of each step. However, the funnel model we usually make is relatively simple and only serves to analyze the final results.

In fact, when we build a detailed funnel map, it can help us clearly understand every detail of the product and enable us to be more targeted when optimizing the product.

1. Organize page funnel and operation funnel

First of all, we need to clarify our overall page hierarchy, organize it according to the process, and clarify the conversion status of the upper and lower level pages and the significance they represent.

Taking video-on-demand paid products as an example, they can be divided into five levels:

• First level: on-demand first level page. It includes various film recommendations, event recommendations, special recommendations, column entrances, and ranking entrances. The conversion from this level page to the lower level page shows whether the movies we recommend are in line with the user's taste, whether the operation activities and topics can attract users, whether the order of the columns we created is reasonable, and whether the column names and styles make users want to click;
• Second level: middle page. Including search page, column list page, activity list page, ranking page, etc. The conversion of this level of pages shows whether our search efficiency is high enough, whether the search results are what users want, whether the search recommendations can attract users, whether the column list sorting and recommendations are efficient and reasonable, etc.
• Level 3: Details page. Includes film introduction, selection, preview, ordering entrance, related films, actors, director recommendations, etc. The conversion rate of this level of page shows whether the information on the detail page is sufficient to support users in eliminating their psychological doubts and make them decide to place an order. When the data on this page is not high, you have to consider adding powerful auxiliary information to help users make decisions, such as adding Douban ratings, classic film reviews, and awards won by the film.
• Level 4: Order page. Including paid product packages, payment methods, scanning QR codes, etc. The conversion rate of this level of page directly indicates whether the ordering operation is smooth, whether the process is streamlined enough, and whether there are any payment problems;
• Level 5: Repeat purchase. It refers to the situation where the user purchases again.

However, data cannot be counted at the page level. We need to track various operations and buttons to count the various data indicators we need. Therefore, the conversion funnel we build is based on the data statistics of the operation steps. For example, the figure below shows the entire process from the user's entry into the on-demand column to the final viewing. By tracking these operations, the funnel data of this main process can be obtained.

After the operation process was established, we found that this funnel did not have much guiding significance. There were three important issues that needed to be addressed:

1. We usually check the conversion rate between pages, and the optimization is also for pages, but the statistics are the data of operation steps, which are not the same.
2. In actual products, processes often diffuse or converge, and different processes become entangled, which makes it difficult for us to perform precise product optimization and analysis.
3. When we analyze a process alone, we are often like a blind man touching an elephant, without seeing the whole picture. For example, we improve the conversion rate of a certain entrance on a page, but in fact it is because we have snatched traffic from other entrances on the same page. This does not help improve our overall conversion rate at all, and it is a waste of manpower.

Faced with the above problems, we need to use new ways to solve them.

2. Combine page funnel and operation funnel

In order to have a more instructive funnel model to help us with optimization, we need to take the page process as the main body and the operation as the connection, integrate both processes, and establish a detailed funnel map, so as to have a systematic and comprehensive understanding of our pages, operations, and the relationship between the two.

Through this picture, we can clearly see which pages are included in each level, the pages and operation processes, which operations are used to connect the jumps between pages, and the entrance and exit of each page traffic.

After counting the data from each embedding point, we can mark the dwell time, bounce rate, and click conversion data of each operation on each page. This not only allows us to understand the conversion rate of each level vertically, but also allows us to compare horizontally to intuitively see the traffic and user dwell time obtained by multiple pages of the same level, and then clearly understand user preferences and analyze the reasons.

After we have completed product optimization, we can use this chart to analyze and compare the optimization results from an overall perspective. For example, we found that the conversion rate of the "Selection Page" increased after optimization, not because it snatched traffic from the "Preview Page" and "Purchase Page", but because of the reduction in bounce rate.

However, when doing further analysis, there is still a problem that has not been solved: that is, for the same film, it may be in a list, a special topic, and a ranking list, so how to judge which form is more effective? In addition, for different topic styles, how do you judge which style is better?

3. Establish accurate traffic marking

To solve the above problems, we need to have a clear understanding of the conversion status of each form throughout the entire process, and be able to find a clear and corresponding thread in these constantly spreading and converging processes. At this time, we need to add a suffix to the traffic from each parent page to the subordinate page when embedding the points, in order to distinguish them.

Starting from the secondary page, taking the traffic of the "search page" and "ranking page" as examples, add "_Ser" to the "search page" traffic and add "_List" to the "ranking page" traffic. When the "search page" traffic passes through the "search results page", this part needs to be marked with "_Ser_Result". When the traffic converges to the details page and then disperses, you can clearly know the traffic composition of the next level and continue marking on this basis.

For this labeling method, it is important to note that the suffix needs to be added only when the process is diffused. When all traffic converges to one page, such as the "Details Page" here, there is no need to add the so-called details page label when it comes out.

In this way, we can make accurate analysis and judgment on the same level, different forms of effects, the same function, different styles of effects.

Similarly, on this basis, we can do further refined operations. After slicing the user base, we can add specific labels to each type of user. By using this funnel map, we can clearly see the focus of different types of users, differences in function usage, and other information, which will greatly help product and operation work.

The author of this article @Z. Fly compiled and published by (Qinggua Media). Please indicate the author information and source when reprinting!

Product promotion services: APP promotion services, advertising platform, Longyou Games