Consumption of 10,000, but conversion rate is less than 0.008! How to analyze and optimize accounts?

Today, I would like to share with you a case study of data analysis . As the saying goes, practice makes perfect. Dear readers, don't miss it if you pass by~

The purpose of data analysis is to improve account performance. So before the analysis, we need to be clear: What is the current disease?

Analyze with a purpose!

Taking weekly data as an example, the figure below shows the data collected from June 18 to June 22 using the marketing process chart. What do you think is the problem?

Impressions are decreasing, and clicks are also decreasing? In fact, such analysis is wrong.

Do not optimize based on data changes on a certain day!

Dear readers, please keep the above sentences in mind. Because one day's data is not representative. It is recommended to conduct comparative analysis with the previous period of the same cycle when analyzing.

We can compare the data from 6.18-6.22 with that from 6.11-6.15. The data from the above two weeks can be summarized and compared through the Pivot Table function in Excel.

As shown below.

By comparison, it will be easier for us to find the problems.

From the chart we can see that compared to last week, the number of impressions increased by 7,535, the number of leads increased by 3, the conversation cost decreased by 673.29 yuan, but the lead cost increased by 15.57 yuan.

Overall, there is a trend of declining conversion .

After clarifying the type of disease, we need to further analyze it. We can analyze and determine the direction of account optimization through four dimensions: plan, time period, region, and keywords !

From the figure, we can see that “ 2018 PC Night ”, “ Sauna room, brand word, competitor word -PC ” all have consumption but no conversion.

Optimization plan:

Its display and click rates are very low, and the data is too little to be of reference value.

The next step is to scale up these plans and check their conversion status.

From the above figure we can see:

1. In places like Fujian, Jiangsu, and Liaoning, high consumption leads to low conversion
2. Places like Zhejiang, Shanghai, Sichuan, Henan, Jiangxi, etc. have low costs and good conversion rates.
3. Places like Shaanxi, Guangxi, Hubei, etc. have consumption but no conversion. Especially in Hebei, Shanxi, etc., the conversion cost is extremely high.

Optimization plan:

1. In places like Fujian and Jiangsu, the bid price should be appropriately lowered.
2. Places like Zhejiang and Shanghai have low costs and good conversion rates, so they can operate in large quantities.
3. In high-cost, no-conversion regions like Hebei and Shanxi, try to reduce this part of consumption as much as possible, and try to optimize the page or quality to check the results.

From the figure we can see:

1. 10 o'clock, 18 o'clock, 19 o'clock, high cost, poor conversion
2. 9 o'clock, 12 o'clock, 15 o'clock, 17 o'clock, low cost and good conversion

Optimization plan:

1. Reduce the premiums during the 10:00, 18:00, and 19:00 hours periods to reduce consumption during these periods
2. Increase the time slot premium at 9:00, 12:00, 15:00 and 17:00 to get more conversions

As can be seen from the above figure:

1. The term "family steam room" has high cost and high conversion rate.
2. Sauna room, home sauna room, light wave room ranking and other words have low cost and good conversion
3. Home steam room, steam room quotation, steam room price are mostly high consumption without conversion

Optimization plan:

1. First check the traffic quality of the home steam room. If it is very poor, charge a match. If it is not bad, lower the price.
2. Increase the volume of keywords such as sauna room, family sauna room, and Guangbo room ranking to increase the volume of these keywords with good conversions
3. Reduce or suspend high-consumption, non-conversion keywords such as home steam rooms, steam room quotes, and steam room prices.

The above is the basic analysis of the above case.

1. First compare the marketing process table and find out the disease type
2. Determine the optimization direction from four dimensions: time period, region, plan, and keyword
3. Implement optimization operations

Author: Rabbit , authorized to publish by Qinggua Media .

Source: Houchang College (ID: semlianmeng )