When placing advertisements, operators need to keep an eye on the computer, adjust the number of advertisements, grasp the timing, observe the effectiveness of the advertisements , and also worry about the cost budget. Is there any way to reduce the operational workload? Based on his own work experience, the author of this article teaches you how to build an advertising strategy system. I hope it will be helpful to you.

1. Industry Pain Points

Advertising operators may be the group that needs the most care. Their daily routine is as follows:

• You have to keep an eye on the advertising results at all times, and even need to bring a computer with you when traveling, for fear of missing the best time to adjust the advertising.
• I am still looking at the computer late at night, trying to seize the opportunity and information of large volume.
• Manually adjusting multiple advertisements is mechanical and boring. I was in a hurry on my way to work, worried that the advertising costs would be out of control.

2. Delivery Strategy System

What is the delivery strategy system? Simply put, it is an automated advertising monitoring tool. This allows operators to let go with peace of mind, and the ads will be adjusted and optimized as preset.

However, before building this system, you must first be familiar with the interface documents of each media, and have a sufficient understanding of which ones can be implemented and which ones cannot be implemented for the time being. For some permissions, you also need to communicate with the official to apply for permissions.

Attached are the interface document addresses of Guangdiantong and Toutiao.

Guangdiantong: https://developers.e.qq.com/docs/start/knowledge/terms

Headline: https://ad.oceanengine.com/openapi/doc/index.html?id=37

So what does a strategy system consist of?

1. Data Items

Advertising will have front-end and back-end data, such as the number of views and clicks on the front end, and the number of APP downloads and registration costs on the back end. These data items constitute the first part of the strategy system. Usually operations need to use this data to monitor the performance of an advertisement.

In the early stages of advertising, operations are more focused on front-end browsing and clicks. If these front-end data cannot be increased, back-end conversions are meaningless. Therefore, only after your advertising traffic has reached the required level, should you start to verify whether your advertising creatives are attractive enough to users. That is, after users click on your ad, whether your advertising images and copywriting materials can convert users and make them pay and pay.

Therefore, the setting of data items needs to be supplemented by back-end indicator verification after the front-end data reaches a basic value.

2. Logical judgment conditions and values

After we have the data items, we need to determine whether these data items are greater than, less than, or between what ranges. This is easier to understand. The common logical judgment conditions are:

• Greater than
• Less than
• Greater than or equal to
• Less than or equal to
• Between
• Approaching (crossing the finish line)

If the "view volume" of an advertisement is greater than ××, and the "APP download cost" of the advertisement is less than ×× yuan, such an advertisement is a high-quality advertisement.

3. Conditional combination

Condition combination, that is, the properties set for the condition. The condition input can be satisfied either (or) or both (and), similar to the and & or of growing io.

Adding an and is equivalent to adding a scenario.

Adding an or is equivalent to adding a series condition to the scenario.

That is, whether the condition is met arbitrarily or simultaneously, the combination of these two conditions can create various actual delivery scenarios.

For example, when consumption and costs do not meet the budget, the budget needs to be reduced in a timely manner. You can use or to combine a strategy:

If "consumption" is greater than RMB ×× and "conversion cost" is greater than RMB ××, the budget reduction operation will be executed.

For example, when an ad passes the cold start phase, it is necessary to increase the budget in time to achieve the goal of rapid growth. The following strategy can be set:

If "Exposure" is greater than × ×, and "Conversion" is greater than or equal to ××, the budget will be increased to ×× yuan.

The above are two simple examples of scenarios, but the actual delivery is far more complicated than this.

For example, when a user first uses the system, a "conversion cost" will be set to control advertising, but no basic conversion number is set. A single composite indicator constitutes the strategy.

In the product design process, it is necessary to avoid losses caused by user misoperation. After all, any losses in the advertising process are real losses.

4. Execute the action

When the ad meets the set conditions, the corresponding action needs to be performed on the ad. Such as adjusting the budget, adjusting the bid, changing the targeting package, and pausing the ad. Any operation that does not trigger an audit can be performed.

It should be noted here that when adjusting the budget, there will be a maximum budget setting, and usually the budget setting will have a cap value. Adjust the amount by a certain amount each time until the cap is reached. The same applies to price adjustments.

In addition, execution actions can be combined with tasks. Users can create a new task and trigger the execution of the task when certain conditions are met. Compared with actions, tasks are more complex and cover more scenarios, such as the task of creating a new advertisement. Creating a new ad requires a lot of information and the process is lengthy. Users can fill in this information in advance and generate a task. When executing a strategy, you can select a task to trigger.

5. Supplement

After building the prototype of this strategy system, we must also consider the concurrency of the interface. Because each media has restrictions on minute and daily calls to the interface, this affects the frequency of pulling advertising data. If the data is updated once every hour, then the strategy system will be useless. Because if it is a large KA customer, one hour of consumption is already a lot.

As a qualified product, when connecting to these media, you have to know the functional implementation of the interface. It is best to learn how to use Postman by yourself. After all, if you often ask the developer to help you adjust the interface, he will rebel sooner or later.

Author: Little Blacksmith

Source: Little Blacksmith