Tik Tok Operation Data Analysis Skills

Before we create a Douyin account, we must find a track that suits us. The prerequisite is that we must determine the IP positioning of the account and master certain market data to better judge which track is more suitable.

Secondly, we need to use data to analyze how our peers are currently playing this track, and whether the same approach is also applicable to us. Third, we need to learn the logic and rules of the explosion of short videos, which also requires data support. Finally, data analysis is the key to live streaming monetization. Live streaming data indicators can help us continuously optimize and iterate products and live streaming techniques.

In short, we need to adjust our operation direction based on the results of data analysis.

The two most important data for adjusting the operation direction are live streaming data analysis and short video data analysis. It is recommended that 60% of the time be spent on live streaming sales analysis for these two core data. As a merchant, the focus should still be on monetization. The other 40% of the time should be spent on short video content analysis. Douyin itself is a content ecosystem platform, so we should also pay attention to content.

Short video operation data analysis direction:

• Traffic analysis by account domain. For example, if our account is for maternal and child care, we need to understand the size of traffic in this field and what the traffic ceiling is so that we can better judge the next layout.
• Account audience analysis. Only by clearly knowing the gender ratio, distribution area, age group, etc. of current fans can we match more accurate content.
• Analysis of account content operation direction. We need to make real-time adjustments based on fans' feedback on the current content, whether they like it or not.
• Analysis of fans’ activity over time. It is important to know the time periods when fans are active. The activity time of different accounts is also different. Only by understanding the life patterns of fans can we better monetize live broadcasts and attract traffic through short videos.
• Analysis of account content and work performance. What is the data feedback for a single work or all works, and whether the content needs to be adjusted. For example, if the content of your account has been continuously posted according to the same template and script, but you find that its playback volume is basically around 300 to 400, and you want to break through the playback volume bottleneck, then at this time you need to do a performance analysis on the content, understand whether the performance of the work is good or bad, whether users like it, and then adjust the video type and content script.

Data analysis direction for live broadcast operations:

• Single live broadcast data analysis. Review the data promptly after the live broadcast, and do not wait until the next day. Only by reviewing the data in real time can you get feedback on the live broadcast results.
• Live real-time data analysis. The anchor needs to make real-time adjustments to the speech through real-time data analysis, and the operation also needs to make strategic push or product marketing plan changes at different rhythms based on the real-time data.
• Analysis of live streaming traffic sources. Understand whether the audience comes from short video recommendations, follow-up pages, or live broadcast square recommendations. Only by knowing the source of traffic for this live broadcast can we grasp the direction of traffic. If the traffic for this live broadcast is all from short videos, then the next live broadcast should focus on the ability of short videos to attract traffic.
• Analysis of live broadcast promotion. For example, how much money did we invest in DOU+ for this live broadcast? Was the final data feedback good? If it is not good, can it be suspended indirectly?

Short video refined data analysis skills

1. Core methodology of data analysis for short videos

Let us first understand the methods of short video data analysis. The dimensions of core data analysis are mainly divided into four categories.

• The average playback time of the video. Pay attention to the average playback time of the video to understand the quality of the video content.

Viewing path: Open Tik Tok and select one of your own works. Click “Data Analysis” at the bottom to jump to the data performance of a single video.

• Crowd tags. Observe whether the crowd tags introduced by the video are consistent with the direction of the later conversion crowd and whether they are consistent with the final monetization traffic. If you are selling children's clothing, most of the people who buy children's clothing are mothers. If the video content attracts only male users, it means that the video content is not suitable.

Viewing path: On the Douyin Enterprise Account webpage, you can see information such as fans’ active distribution time period, gender, region, device, interests, etc.

The duration of the video. Test the length of each short video content. Different types of short video content require different playback times. It is recommended to prepare videos of the same type but different lengths for testing to see which length gives better feedback for the same content. In addition, the shorter the playback time, the better the completion rate. For example, when shooting food, it is difficult to show the food production process or the tasting and evaluation process in a short time, so the content must be clear and complete.

View path: Douyin "Enterprise Service Center" or Douyin Enterprise Account computer backend.

Number of views. The basic number of video views can reflect the weight of an account. The better the account content performance, the higher the number of views. The basis of video playback volume determines the early weight of the account, so the playback volume of the first ten newly released videos of the account should be around 500 as much as possible. The account weight with playback volume below 300 is poor, and the content needs to be optimized in the later stage to increase the basic playback volume.

Viewing path: On the computer backend of Douyin Enterprise Account, go to “Short Video Management” in “Operation Center”.

1. Deep optimization method for short video data

The three core points of short video data: average playback time, crowd labels, and video playback volume. If the short video data is not good, how should we optimize it in depth?

How to optimize the average video playback time can be divided into three steps. Step 1: First choose the type of video you want to shoot, prepare 3 to 6 videos (the duration can be divided into three stages: short, medium and long), and publish them at the same time, observe the overall average duration of the videos within 24 hours, and choose the content type that is suitable for your account through testing (average playback time of more than 70% is excellent data); Step 2: If the average playback time is unstable and has not reached the excellent indicator standard, you can consider changing the video content type or imitating the popular video content to follow the shooting, which will help improve the account data and also help us find direction and inspiration. Step 3: If the completion rate has not increased, you can shorten the playback time of the video and quickly express the core content without dragging it out. This is the last resort and should not be used unless absolutely necessary. The premise is that we must first ensure the quality of the content, which is the most important thing.

Crowd label data is very important for the early stage of account operation and is related to the subsequent monetization effect. In the early stage, you must check what kind of people the video content attracts. If the labels of the people attracted by the video content are wrong, change the video content immediately. Here is a method to forcefully pull crowd tags, which is to use Dou＋. In the early stage, you can first select fans similar to those of influencers for delivery. When the crowd tags are stable, you can directly deliver them intelligently in the later stage.

1. Live broadcast refined data analysis skills

1. Core methodology of live broadcast data analysis

We will definitely encounter some problems during the live broadcast process. Which data indicators should we use to optimize the data? Let’s take a look at the core data analysis dimension model of live broadcast. From bottom to top, they are live broadcast duration, viewing duration, live broadcast room conversion rate, GMV and per capita GMV.

(1) Live broadcast duration. In the initial stage, the live broadcast duration should be extended as much as possible. The longer the duration, the better the live broadcast data performance. The recommended duration of the live broadcast is about 3-6 hours.

• In the early stages of live streaming, it is recommended to increase the duration and frequency of live streaming. Although the popularity of live streaming is low in the early stages, increasing the duration of live streaming is undoubtedly the best way to increase live streaming traffic.
• If the live broadcast time is too short, the system will not be able to recognize the data tags of the live broadcast room, resulting in the live broadcast room being unable to obtain free recommended traffic. Only by extending the playback time can the free traffic flow in continuously be felt.
• While increasing the length of live broadcasts, it is also important to interact with fans.

(2) Viewing duration. Be aware of the length of time fans watch the live broadcast. During the process, the host should try to guide fans to stay, whether through the host's words or through the use of products. Watching time can be used to retain fans through products, anchors, and marketing activities.

Tips for retaining people:

• What you see (the information displayed in the live broadcast room) guides fans to take actions through the visual presentation of the live broadcast room.
• What you hear (the host’s guidance) is that it is important to design a good script for the entire live broadcast.
• What I felt (the overall atmosphere of the live broadcast room)

(3) Conversion rate of the live broadcast room. The conversion rate of the live broadcast room should be around 3% as much as possible, and discounts or promotions can be used to increase conversion.

The key to improving conversion rates in live broadcast rooms lies in the rhythm design and product selection and pricing of the entire live broadcast. If the products in the live broadcast room fail to attract users, the product prices cannot retain users, and the host's guidance and rhythm are not well controlled, users will feel that it is too slow and will be unwilling to spend time in the live broadcast room. Adjustments must be made at this time. The design of the overall rhythm can be based on the three gameplays of the product. You can set up lottery models, flash sale models, and regular models, and adjust the real-time conversion effect of the live broadcast room through different product pricing and marketing strategies.

(4) GMV and per capita GMV. These two data allow us to clearly understand the sales of the entire live broadcast and the average spending power of each person, making it easier for us to adjust and optimize the product later.

GMV refers to the sales data of the entire live broadcast, and per capita GMV refers to the average consumption value of fans. If you want your live broadcast room to receive a steady stream of free traffic support, you need to increase the per capita GMV of the entire live broadcast. Generally, live broadcast rooms with high sales have relatively excellent data performance in terms of traffic and fan retention. In the early stage, these two data indicators can be achieved by guiding old customers.

2. In-depth optimization method for short video data

There are too many data indicators in the live broadcast room. When we analyze the live broadcast data, the three core points are the average viewing time per person, the number of shopping cart clicks, and the transaction amount of the store. The average viewing time per person should be as high as possible, and the number of clicks on the shopping cart and the transaction amount of the store should be as good as possible, so that the system will continue to push free traffic. Therefore, if you want to achieve better live broadcast results, you can perform in-depth optimization and adjustments based on these three data dimensions.
Finally, let’s take a look at methods for deep optimization of live broadcast data. Quickly improve the monetization capabilities of live broadcast rooms in three directions.

1. Live content quality analysis. In the live broadcast content quality analysis, if the live broadcast content quality is not good, the reason may be that the interaction rate is not high or the conversion rate is too low. At this time, adjustments can be made through related factors. The related factors include: traffic accuracy, product selection attractiveness, product display, marketing activity, and anchor guidance.

1. Live sales efficiency analysis. Data indicators: conversion rate, order conversion rate, average order value, average order quantity, and UV value. Related factors include: whether the traffic is accurate, whether the product is powerful, related sales capabilities, live broadcast display, anchor guidance, etc.

1. Live streaming traffic optimization analysis. By analyzing the traffic sources, we can clearly know which traffic source can bring more people and what our core advantages are. As long as we continue to strengthen and optimize our advantages, we can ensure that the traffic for each live broadcast is relatively high-quality.

We can directly refer to the above three tables, write down the data of each live broadcast according to these data indicators, find its related factors, carry out refined operations and adjustments, and make our live broadcast data and our live broadcast room better.

Summary: Mastering the basic data analysis capabilities of Douyin Enterprise Accounts, and learning the data analysis methods and optimization logic for short videos and live broadcasts can help us quickly implement and master the gameplay of Douyin Enterprise Accounts, create popular Douyin Enterprise Accounts, and learn to optimize live broadcast data to maximize live broadcast effects.