618. How to monitor the Double Eleven promotional activities?

Every year’s 618 is very special. At that time, are your friends busy working overtime and preparing for the activities? Especially for event monitoring, the leaders of each event would urge us to submit monitoring data again and again, which made people nervous.

So, how do you monitor activity? Let’s take an example of a small event that our company just did last month. This activity is a very simple welfare activity for all people.

From May 10th to May 31st, users can receive a coupon after logging into the APP. The coupon is valid throughout May and offers a discount of 80 yuan for purchases over 400 yuan.

The activity is so simple! So the report made by the girl in charge of operations was also very simple:

• May 11: 280,000 people attended
• May 12: 120,000 people attended
• May 13: 80,000 people attended
• …

I spoke word by word like a chick pecking at rice, and then was criticized by my leader on the spot: This is not how you report progress! Then I was asked to work overtime together.

Why was it criticized? Because to use the leader’s own words: “You can’t tell anything from this! At least add the goals!” Indeed, without goals, it’s unclear whether these numbers are good or bad. Therefore, the first step in monitoring operational progress is to compare with the goals and identify problems.

The total goal of the event is to have 1 million participants and 500,000 people use the coupons for consumption. So should we just compare the current data with this 1 million/500,000?

Of course not, because the event lasts for 22 days. If you only look at the overall goal, you will find out on the last day: Oops, we can’t meet the target! Then the situation is over. Therefore, the second step in monitoring operational progress is to break down goals and establish standards.

Theoretically, the simplest way to decompose it is 100/22=45,000. The target is met if 45,000 people participate every day. But judging from the actual data, it is obvious that more people participated on the first day, and the number began to decrease significantly on the second day. Therefore, it is very likely that the distribution of participants is not uniform. Therefore, you can refer to previous activity data to see the daily participation distribution.

Find similar events in April for reference. There were 800,000 participants at that time. From April 12 to April 30, the event participation data is shown in the following table. In order to see the distribution clearly, it can be divided into:

1. Step 1: Use the SUM function to calculate the total number of participants in the entire event
2. Step 2: Calculate the number of participants per day and the proportion of the total number of participants
3. Step 3: Draw a line graph (use a line graph if there are more than 10 points and the data is continuous)

This way you can see it very clearly.

Judging from this distribution, the first three days are the focus, and there is 3%-1% participation every day thereafter, with a small climax at the end. Although the activity time in May is different from that in April, the form of the activity is the same, so you can refer to the trend in April. Following the trend in April, the goals for May are broken down as follows:

Using the same method, you can break down the target number of people using coupons. Interestingly, the trend in the number of people using coupons is not consistent with the trend in the number of people receiving coupons, with a clear peak at the end of the month. Guess: There should be two types of users.

• Category 1: Using the coupons immediately after receiving them. Judging from the data, this ratio is relatively stable.
• Another type: The coupons are about to expire at the end of the month, so they are used only when you receive a reminder.

Therefore, to calculate the coupon usage rate on the last day, the formula should be: number of people using coupons / (number of people who received coupons - number of people who used coupons). After estimation, the coupon usage rate on April 30 was about 20%, as shown in the figure below.

This will simulate the number of people using coupons every day in May.

With these judgment criteria, we can determine the trend of this month's activities. When comparing goals, both daily completion status and cumulative completion status should be compared because the two have different meanings.

Daily completion status: What is the development trend? Are the current measures effective?

Cumulative completion status: Overall, how much is the surplus and how much is the shortfall.

The data for May was produced and the results are shown in the figure below.

This way of looking at it explains a lot more than just reporting numbers at the beginning:

First of all, the overall coupon collection situation is not good. After 5 days, the progress is short of 50,000.

Secondly, the development trend of coupon collection is not good. Except for the first day when there were a lot of people, the daily differences were negative.

Similarly, you can also make a comparison of the coupon table as follows:

Ah, the result of using coupons is that part-time jobs are even more miserable. If this continues, this month’s activities will definitely be ruined! We have to find a solution quickly. However, in a short period of time, the options that operators can think of are limited.

1. Put the coupon in a more prominent place
2. Push another round to all users
3. Send coupon reminders in advance to encourage users to use the coupons

Note that these three approaches have different effects:

1. Put it in a more prominent position: increase the number of daily user engagement
2. Push a round of information: immediately get a group of people to participate
3. Early coupon reminder: The peak at the end of the month decreases, and the coupon usage rate increases before the end of the month

Reflected in the data, the possible effects are as follows:

However, it seems that these are just last-minute measures. It is very likely that the activity itself was not well designed.

For example, a young man said: The effect of the activity in April was good because you could get 20 off for every 100 you spent. Although the discount of 80 yuan off for purchases over 400 yuan seems to be the same, both are 20% off, but it is definitely more difficult for users to pay 320 yuan in cash than 80 yuan.

Yes, that sounds reasonable, but the matter is done now. Changing the rules temporarily will not only increase the development burden, but will also be unfair to users who have already participated in the event, and will result in complaints. Right now we can only provide emergency relief.

As to whether it was caused by the problem of the coupons, we will have to wait and analyze it later.

The above is the general approach to monitoring and analyzing operational activities. It’s that simple and easy, you can do it with Excel.

There are generally three scenarios for data analysis:

1. Beforehand: planning analysis, forecasting analysis
2. During the event: monitoring analysis, cause analysis
3. Afterwards: Summary Analysis

Author: Coder Bear

Source: Coder Bear