How to review an event you have organized? From which aspects?

This question comes from a reader's question, which is a very good one. His original question is as follows:

How to review an event you have organized? From what perspectives should information and data be collected to optimize future operational strategies and methods?

So, today we will use a short story to talk about the details of review.

The Nature of Community

Xiao Wang organized an activity, hoping to increase the transaction volume of users within a period of time through the activity. Xiao Wang made the following design:

During the event, all users who participate in the transaction can get one chance to draw a lottery, and then they will be ranked according to the actual transaction amount (payment-refund). The top 100 users can draw grand prizes including iPhoneX (which is actually impossible, after all, the unit price of a single prize cannot exceed 5,000 yuan), and the number of chances to draw a lottery will be directly increased from 1 to 10. Users ranked after 100 can draw other prizes including Xiaomi note3. All users have a 100% chance of winning. The expected transaction amount during the event is 3 times that of the non-active period. The total cost is controlled within 50,000 yuan. Everyone will win a prize. The prize pool includes physical prizes and various exchanged coupons. At the same time, registration is required for the event, and the warm-up period is one week before the event.

Now that the activity is over, what dimensions of data should Xiao Wang need for review?

Preparation for replay

1. Effectiveness data of the activity warm-up period

The data that needs to be prepared are:

• UV and PV of the activity preheating page
• Number of clicks on the sign up button
• Number of registered users recorded in the background
• Source Channel
• …

2. Consumption data of registered users and non-registered users in the period before the event, such as the previous two weeks, or one month, or the same period last year

The data that needs to be prepared are:

• UserID
• Trading order volume
• Consumption amount
• Consumption time
• Consumption frequency
• Refund order volume
• Refund amount
• Refund Time
• …

3. Consumption data of registered users and non-registered users during the event

The data that needs to be prepared are:

• UserID
• Trading order volume
• Consumption amount
• Consumption time
• Consumption frequency
• Refund order volume
• Refund amount
• Refund Time
• …

4. Consumption data of registered users and non-registered users for a period of time after the event, such as the next two weeks, or a month, or the same period last year

The data that needs to be prepared are:

• UserID
• Trading order volume
• Consumption amount
• Consumption time
• Consumption frequency
• Refund order volume
• Refund amount
• Refund Time
• …

What is the problem?

What do these data show?

1. Preheating data is used to illustrate the traffic source pull of the promotion channel
2. Compare the data of registered users during the event period and the data of non-activity period to prove whether the event has a driving effect on registered users.
3. Compare the data of non-registered users during the activity period with the data of non-activity period to prove whether the activity has an impact on non-registered users
4. It is extremely important to compare the final successful transaction order data to feedback whether there is a phenomenon of brushing during the activity, as well as the relationship between cost and income.
5. The data after the event is used to provide feedback on whether the event itself will have a long-tail impact on users.

There are a number of things to note here.

1. Some data may not exist in the original system, such as the promotional channel tracking points during the warm-up period, page monitoring tracking points, button click tracking points, etc. These data need to be confirmed and communicated with the developer before development.
2. There are two types of data comparison: month-on-month and year-on-year. Month-on-month data can be compared with the previous week or month to see the time of the activity, and it is continuous. Year-on-year data is compared with the same period of the previous year, and it is not continuous. Which comparison method to use needs to be discussed in the actual scenario, and usually a year-on-year comparison is sufficient.
3. In addition to data, behavior should also be reviewed. Data is used to verify the effect, while behavior is to find the reasons that cause data changes. There are many things to record here.
4. The purpose of reviewing is not to investigate whether the event went as expected, but to accumulate experience so that you can understand which pitfalls to avoid or which aspects to strengthen in the next event, including channels, the event rules themselves, copywriting , etc.

summary

To sum up, if you want to investigate data, you must first have data. The premise of having data is that you know what kind of comparison you want to make, so that you can see whether to add new statistical sources.

In terms of statistical tools , UV and PV traffic detection tools can use third-party statistics such as Baidu Statistics as input sources. For other statistics, if they are daily data, they should have their own point monitoring. If not, then it may be a part that needs to be strengthened.

That’s all for now.

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

Product promotion services: APP promotion services, information flow advertising, advertising platform