Activity operation data analysis method!

Data analysis is one of the core capabilities of operations, especially in event operations .

The application of data analysis can help event operations go from subjective to objective, from chaotic to controllable, and from defective to perfect. It is a compass that runs through the entire event and guides event operations.

Today, let’s have a systematic discussion about how to make good use of data in event operations, involving specific methods of data analysis, data analysis of different types of events, event cost-benefit evaluation, and AB comparison experiments. I hope it will be helpful to you.

The full text is long and the structure is as follows:

1. Basic skills of operational data analysis

When it comes to event operations, it’s easy to think of wild ideas and crazy ways of spreading the game. However, creative ideas and planning operations also require strong support from data analysis.

It cannot be said that many event operators do not have the ability to analyze data, because they do not have the awareness of data analysis.

Data analysis awareness means paying attention to data, valuing data, and making decisions and actions based on data analysis in all aspects of operations.

Focusing on operational goals, reducing reliance on experience, increasing data support, and establishing data analysis awareness are the prerequisites for doing good data analysis.

Once you have the awareness of data analysis, then we can talk about data analysis capabilities. Data analysis capabilities focus more on analysis methods and ideas: How to break down and locate problems? How to find the analysis dimension? Commonly used analysis method models? This is what you should focus on when learning applied data analysis. In addition, when talking about data analysis capabilities, many people think of data analysis tools. For operations, the importance of tools is much lower. Excel is basically sufficient. Other tool capabilities are a plus, but not necessary.

The focus is on introducing the basic skills of data analysis: analysis process and analysis methods.

1. Data Analysis Process

The end point of data analysis is to draw causal conclusions and form recommendations/decisions to guide subsequent directions and actions.

Reaching the end point of data analysis requires a three-step process:

The first step is to identify the problem or goal.

This is the starting point of data analysis, which determines the center and direction of data analysis and is the prerequisite for effective data analysis. When clarifying a problem, it is necessary to avoid preconceived problem definitions, start more from phenomena and data, and define the scope and objectives of the problem, which can then guide the subsequent scope and ideas of analysis.

The second step is to break down and analyze the causes.

Based on the identified problems or goals, further breakdown and analysis are carried out to identify the key factors that cause the problems/affect the goals. In most cases, there will be multiple influencing factors, and it is necessary to judge and analyze the impact and verify the role of the factors. Efficient disassembly and analysis requires a combination of data analysis methods and operational practice experience. The methods are shared below and experience is accumulated slowly.

The third step is to draw recommended conclusions.

The value of the last step of data analysis is to find a solution to the problem or a path to achieve the goal, which requires outputting conclusions and suggestions. The conclusion refers to the key factors of the problem, the way/size of the factors' influence, and other related factors. The suggestions refer to how to influence the key factors, what actions need to be taken, and the ideas and strategies for the actions. After analyzing the data, take action.

2. Data Analysis Methods

There are many data analysis method models, which need to be flexibly selected based on analysis needs. For complex operational analysis problems, multiple analysis method models are generally required. Here I share with you 3 commonly used and essential analysis methods.

Method 1: Comparative distribution analysis

Comparative distribution analysis is the most commonly used analysis method, which helps to identify difference definition problems in the early stage of analysis.

The core of comparison is to solve the problem of how to compare effectively and find out the clear characteristics of differences. External comparison focuses more on differences, while internal comparison focuses on changes.

When making specific comparisons, you can compare scale: focus on total amount/average/median, you can also compare fluctuations: focus on variance/standard deviation/range, and you can also compare trends: focus on month-on-month/year-on-year/change.

Solution 2: Path Funnel Analysis

The path funnel analysis method is particularly suitable for event operations, and can effectively monitor process effects and locate key factors.

To use the path funnel analysis method, you first need to split the activity stages into the publicity and reach stage, the participation and sharing stage, the fission and conversion stage...then clarify the user behavior at each stage, especially the high-value user behavior, and then sort out the corresponding data indicators to form a complete path funnel. In actual monitoring, combined with comparative analysis, attention is paid to change differences and abnormal data, and analytical applications are carried out.

Method 3: Dimensional analysis

Dimensional decomposition analysis plays a huge role in locating the causes of specific problems and breaking down target components.

In many cases, we are faced with the manifestation or result of a problem, and data analysis needs to find the specific cause. At this time, decomposing the overall data and the problem into dimensions is an effective means to discover the cause.

Common decomposition dimensions include time dimension, channel dimension, user stratification, etc. For example, if the daily sales of a promotion activity declines, it can be decomposed from the channel dimension to see whether the number of orders has decreased or the number of orders has decreased. It can also be decomposed from the order dimension to see whether the number of orders has decreased or the order amount has decreased. By continuously decomposing, the ultimate cause can be located to guide subsequent operational actions.

Data analysis is an ongoing job that requires awareness and attention. At the same time, data analysis is also a complex job that requires learning methods, accumulating experience, and laying a solid foundation. Only then will data analysis be easier.

2. Data analysis throughout the activities

The application of data analysis in event operations runs throughout the entire process, and the value of data can be brought into play in every link.

1. How to use data before the event

What needs to be done before the event? The first step is to establish an activity. Under specific goals and background, decide to do an activity; then clarify specific goals and resources, guide activity planning and resource investment; then conduct activity planning communication, form an activity plan and promote its implementation.

What role can data play in this?

The first is the early stage of project establishment, which requires data support to illustrate the necessity.

Why do we do this activity? What problems can it solve? Through research analysis, data breakdown and operational insights, the necessity of explanation activities can be supported.

Secondly, in the stage of clarifying goals, data support is needed to clarify the direction.

How to set activity goals? Is it reasonable? Through competitor activities and historical activity data, compare the differences and define the overall approximate goals. Then, break down the activity process, determine resource investment, and gradually estimate the effects of each link to improve the accuracy of the overall goals.

Finally, during the event planning stage, data support is needed to improve feasibility.

Why are the activity process rules designed this way? How to ensure the effectiveness of the activity? From the perspectives of time rhythm, scenario channels, activity rewards, user characteristics, process characteristics, and cost-benefit, combined with data comparison/disassembly, we support the formulation of activity gameplay processes.

2. How to use data in activities

After the preliminary preparations for the event are completed, the next step is to go online for operation. At this time, event data monitoring and iterative optimization become the focus.

Activity data monitoring ensures control over the functional status and operational effects of activities, and ensures the stable progress of activities.

Activity monitoring breaks down the activity process, focusing on the process indicators and final goals of each link. It can also further break down the overall data from the time dimension, channel dimension, user dimension, etc., so as to monitor the activity results more accurately.

Based on data monitoring, it is necessary to combine data analysis to make adjustments and optimizations during the activity.

There are many angles to find optimization ideas, such as looking at data fluctuations from the time dimension, paying attention to abnormal data increases or decreases, and analyzing the specific reasons; looking at feature differences from the user/channel dimension, paying attention to differences in activity data for different channels or different types of users, and focusing on advantageous channels and users; you can also look at the room for improvement from the historical data dimension, clarify the effects and room for improvement of current activities based on historical activities and industry competitor data, and determine the value and investment of optimization.

3. How to use the data after the event

After the event, data is an effective way to reflect the effectiveness of the event and conduct a review and summary.

The activity results should highlight the core indicator results, auxiliary indicator results and process indicator results, compare the differences between the target and the actual, and simultaneously perform dimensional decomposition to identify the specific reasons why the activity is good/poor in effect. Is it because a certain channel is not working? Or is the participation of a certain group of users exceeding expectations? Only in this way can we gain valuable event experience and guide the planning and operation of more subsequent events.

Data analysis plays an important role before, during and after an event. It can be said that it is related to the success or failure of the event. Pay attention to data and make good use of it.

3. Cost-benefit of return target

In event operation, "money" is indispensable, which includes the cost of event investment, the various benefits generated by the event, and the relative relationship between cost and benefit, that is, the return on investment (ROI).

1. Let’s talk about the cost of the activity first

Most new customer acquisition/conversion/promotion activities involve cost investment, which requires accurate calculation and evaluation:

(1) New customer acquisition activities have new customer acquisition costs

That is, how much money it costs to bring in a new user, also called customer acquisition cost (CAC), which is usually calculated by dividing the total cost of the new user acquisition activity (including promotion costs, prize costs, etc.) by the number of new users brought in by the activity.

CAC reflects the cost of acquiring users and affects the sustainability of new customer acquisition activities. What really determines the effectiveness of new customer acquisition activities is the relative relationship between user acquisition cost and user lifetime value. It is also easy to understand that a new customer acquisition activity can spend 100 yuan to acquire a new user, and the user ultimately contributes 1,000 yuan in consumption within the product, then such a new customer acquisition activity must be vigorously carried out.

(2) Conversion activities have first-order conversion costs

That is, how much money is spent to get a new user to complete the first order. The first order conversion cost reflects the effectiveness of the conversion activity and is also a way to judge user quality. The higher the subsidy cost for user conversion, the poorer the user quality and the expected subsequent retention effect is not good.

(3) Promotional activities have promotional costs

That is, how much money is spent to bring about a certain amount of sales. For example, a subsidy of 1 million brings 10 million in sales, and 1 million is the cost of the promotion. You can also pay attention to the cost of each promotional order/ordering user from the order dimension and user dimension, and monitor and analyze the cost of promotional activities more accurately.

2. Let’s look at revenue and ROI

In the long-term operation of the product and the life cycle of the user, there is a very important concept: the net value of the user's lifetime value:

Net value of user lifetime value = user contribution value – product cost

The value contributed by users refers to the amount of money they spend on the product and the profits it brings. It can also be broadly defined as the friends they invite and the word of mouth they spread. The cost of the product includes the new customer acquisition costs, conversion costs, and promotion costs mentioned earlier.

When a user first uses a product, the net value of the user's lifetime value is generally negative. As the user's active consumption contributes more value, the net value of the user's lifetime value will become positive. The larger the user base with a positive net value, the higher the value of the product.

In specific activities, more attention is paid to the ROI (return on investment) of the activities, that is, the benefits divided by the costs .

(1) ROI of new customer acquisition activities

In new customer acquisition activities, customer acquisition costs need to be invested, but the quality of users acquired and their lifetime value cannot be directly measured. It is necessary to refer to the performance of historical activity users within the product and the estimated lifetime value benefits to calculate the ROI of new customer acquisition activities.

For example, the average contribution value of new users brought by the previous new user acquisition campaign was 100 yuan, the customer acquisition cost of this campaign was 110 yuan, and the ROI of the activity was less than 1, which means it was a loss and this activity cannot be sustained.

(2) Conversion activity ROI

Conversion activities are aimed at new users to promote their first conversion. Although they can bring sales, new users are often given large subsidies, and it is difficult for the ROI of conversion activities to be directly greater than 1.

Conversion activities need to focus on both short-term ROI and long-term ROI. Short-term ROI is the sales volume of users in conversion activities minus the cost of conversion activities, while long-term ROI focuses on the contribution value of users within a certain period of time, minus the cost of conversion activities. Generally speaking, long-term ROI is more important.

(3) ROI of promotional activities

Promotional activities are the main form of activities to increase sales and user value, and they focus more on short-term sales revenue and ROI. In addition, there will be new user conversions and silent user recalls in promotional activities. For these users, we can pay proper attention to subsequent retention and long-term benefits.

Organizing an event means spending money. Paying attention to costs means "spending money wisely", while paying attention to revenue and ROI means "spending money with results". Cost-benefit thinking is something that must be possessed and should be valued in operational work.

IV. Iterative and upgraded activity experiments

1. Understanding A/B testing

When it comes to data application-driven operations, A/B testing is a very important tool. A/B testing speaks with facts and is a method of comparative analysis, where plan A, plan B, and so on are compared to select the best one.

Strictly speaking, A/B testing means that when the effectiveness of multiple plans cannot be determined, these plans are randomly tested on target populations with similar composition and characteristics within the same time period. Finally, the plan with the best effect is evaluated through comparative analysis of the plan result data, and then it is officially used in full. In summary, the premise is uncertainty, the core is a single variable, and the conclusion is survival of the fittest.

A/B testing can effectively support the design, operation and iterative optimization of operational activities, especially long-term or periodic activities, which can effectively improve the effectiveness of the activities with the help of A/B testing.

From the design of activity processes and rules to activity rewards and gameplay selection, to the style of activity pages and copywriting, A/B testing can be used to verify and optimize.

2. Apply A/B testing

The application of A/B testing can be divided into a 4-step process:

Step 1: Define your goals and form your hypothesis

When doing an A/B test experiment, you must first be clear about what you want to verify, and you must have prior analysis and judgment, rather than blindly coming up with multiple solutions for testing. After all, A/B testing also requires a certain amount of cost investment and time period. Try to choose the best from the best, rather than picking one out of a million.

Step 2: Determine indicators and select users

Determine the key elements to be tested and verified, and clarify the key indicators that affect them, so that after the subsequent tests are completed, effective comparisons can be made to draw test conclusions. At the same time, A/B testing is a verification of uncertain solutions. Try to avoid using too large a number of users. Randomly select some users for A/B testing.

Step 3: Design a plan and conduct an online experiment

The most important thing about the A/B test experiment plan is to ensure the uniqueness of the test variables. Except for the variable elements to be verified, all other aspects of each plan are the same to avoid interference with the analysis and judgment of the test results.

Step 4: Analyze the results and determine the solution

After the A/B test experiment has the result data, it is time to analyze the pros and cons of each plan and decide on the best plan. The main comparison is the key indicators initially determined, and also pay attention to process indicators and experience-related indicators.

There are some key points in data analysis for A/B experiments. I have previously shared them in detail in an article. You can also read: 5 Things You Should Know About AB Testing

V. Conclusion

That’s all I have to say about data analysis for event operations. The basic skills of data analysis must be consolidated, the data analysis in the process must be thorough, more attention must be paid to cost-benefit, and data experiments must be used to improve results.

There is no need to say more about the importance of data analysis to operational work. In your daily work, if you apply data consciously and methodically, you will get better feedback.