2 tips to improve the efficiency of SEM account delivery! Collected~

In our promotion process, how to set up new SEM accounts has become a relatively basic and very important issue in the industry.

Many users often find in the subsequent promotion process that with the same basic account settings, the specific performance of traffic and effects in different accounts is quite different.

I believe that what is reflected here is not simply a problem of bidding, delivery location, and delivery time (so in this article I will not share with you the content that is already known to everyone).

Regarding the problem of different effects with the same settings during the promotion process, the author believes that there is a relationship between the historical value of the account in the promotion account.

Through exploration in the promotion process, the author discovered that there are some rules to follow in the relationship of historical value.

First of all, the launch of a new account means that a new enterprise has entered the overall competitive environment of the market. When we enter the specific competition link, we can conduct corresponding analysis on the specific business/products of the enterprise to find a more specific promotion rhythm and direction.

Secondly, the author found that in the early stage of account preheating, for companies with a certain brand appeal, it is better to set up the account in advance to only launch brand words. Whether it is the keywords at that time or the keywords added later, the quality performance is very good.

Next, I will share with you these two tips in detail.

1. Specific analysis of market environment

Before deciding to place an ad on a new account, we first need to conduct an index analysis based on popular keywords in the industry. The author uses a client he has served before as an example:

The client’s industry belongs to the auto insurance industry, so we can select keywords such as “auto insurance”, “ car insurance”, and “vehicle insurance” for analysis.

Since not all keywords are included in the Baidu Index, we should try to select the top keywords with larger traffic for analysis, so that the analyzed data is more representative. By analyzing a small amount of keywords, we can sort out the specific situation of the industry environment.

After selecting a keyword, we can use the index trend to view the specific search trend of the keyword and the different characteristics of the PC and WAP terminals. Take the keyword "auto insurance" as an example:

It is not difficult to see from the above figure that the average index of the mobile terminal has exceeded the average index of the PC terminal.

As mobile traffic has grown rapidly this year, the current search environment still emphasizes the importance of mobile, and information from big data such as indexes also shows that mobile traffic has become the mainstream market trend.

Through the above analysis, we can accurately draw the conclusion that today, the main search traffic in the auto insurance industry is concentrated on mobile devices. In the auto insurance industry, mobile devices have become the mainstream competition channel and the mainstream traffic channel.

When analyzing the behavior and needs of industry users, we need to judge the content of the user's demand map scatter plot.

This helps us understand users’ search behavior and their focus on the industry/product, so that we can understand which content in the industry should be promoted and which content should be blocked or denied before promotion, as shown in the figure:

The author divides the relevant demand distribution into four categories:

1. The red-framed part is classified by the author as “brand-related” needs.

This type of demand is for existing brands with a large voice in the industry. The key words mainly include brands such as Ping An, Huatai, PICC, CPIC, Taobao , Taiping, and China Life.

This shows that among the people involved in auto insurance, the market environment is divided by the above brands, and users have a high degree of recognition and demand for the above brands.

2. The green framed part is classified as “business-related” needs.

The user groups with this type of demand include those who have purchased auto insurance and those who have not.

Among those who have purchased insurance, we found that complaints, policy cancellations, etc. accounted for a large number of search demands, while among those who did not purchase auto insurance, the focus was mainly on price, brand ranking, and claims terms. From this level we can see.

When we are a new brand, most of the users we target should be those who are not purchasing auto insurance and are in a wait-and-see period. At this time, we need to deny and block key information such as complaints and policy cancellations.

3. The blue box part is classified by the author as “core needs”.

Through the keywords, we can see that this type of demand group is a relevant group that the auto insurance industry pays close attention to (car owners or car owners who are about to buy auto insurance). Their demand points mainly include insurance types (third-party, concentrated, ordinary, etc.) and forms (telephone auto insurance, online direct sales auto insurance, etc.).

By analyzing the demand scatter plot, we can roughly depict what the user's pre-sales and post-sales needs are in this industry.

After understanding these contents, we will have a clearer idea on the focus of promotion, the selection and negation of keywords.

4. The author classifies the purple-framed part as “policy category”.

People with this type of demand are basically those who have already purchased auto insurance. They are more concerned about the recently introduced auto insurance policies for the auto insurance they purchased. This type of users has relatively little intention to purchase auto insurance.

Therefore, when we launch new accounts, we don't need to pay special attention to this type of users.

After locking in the user's needs, when choosing keywords, we can find the strategy for choosing keywords through the "hot searches" in the Baidu Index, as shown in the figure:

From the Baidu Index demand map in the figure above, we can easily find that among the TOP10 search terms related to auto insurance, the top two words are "credit card" and "weight loss"; the 10th one is "suit customization", not keywords related to auto insurance.

In this regard, we can preliminarily conclude that among the netizens who pay attention to auto insurance, a large proportion of them are also concerned about credit cards (finance), weight loss (health and wellness), and custom suits (clothing), etc.

We further learned that the user groups who are interested in auto insurance also have a high concentration of interests in travel experts (travel), food experts (food), film and television fans (online videos), and family doctors (health and wellness).

So far, through the above analysis, we can summarize the following:

1. Traffic is highly concentrated on mobile terminals, so early placement should focus on mobile terminals;

2. In terms of keyword selection, we can give priority to "business", "brand" and "core demand";

3. While paying attention to keyword placement, we can expand the placement of related sites that the auto insurance population is concerned about.

2. Quality is king

When it comes to the issue of quality, many friends may feel very troubled, thinking that optimizing quality during the promotion process is quite difficult, unstandardized and unreliable.

I believe that there are definitely more than a few people who have this idea, because in my SEM career, I have also had such views on quality.

But by chance, the author discovered a very effective method to optimize quality (provided that the account is a new account and has never been advertised before).

The author encountered a bottleneck in promoting auto insurance to customers. After communicating the strategy with the customer, the author decided to open a new account to test the brand words and only use brand words in this account.

The keywords selected are basically brand words with good delivery effects in old accounts, and new ones are added, such as brand phone numbers, regional brand words, brand question words, brand alternative words, etc.

The bid, creative, promotion time, and promotion area are all consistent with the original account settings. After one month of launch, the author found that the keyword quality of this account is 30% higher than that of the old account on average, as shown in the figure:

The data in the figure shows that the quality advantage of new accounts is very obvious when they only use brand words in the early stage. From this, we can conclude that brand words have the following three characteristics:

1. The user's search intention for the brand is very strong, and the CTR is high;

2. Brand words are highly targeted and have better effects when placed in a small area;

3. Accumulate the overall quality of the account by placing brand words during the warm-up period.

With the combination of the above advantages, the new account of the client in this case not only performed well in terms of brand word quality, but also showed very eye-catching data performance in the subsequent placement of common words, core words, and long-tail words added by the author, as shown in the figure:

We can see that in the early stages of promotion, accounts that use pure brand words for preheating and data accumulation are inferior to accounts that originally used brand words, core words, and common words for preheating in terms of quality performance, customer acquisition performance, and cost performance, both in the early and later stages.

Summarize:

By setting up new accounts through the two methods of preliminary market environment analysis and prioritizing the use of brand words, the author believes that everyone can achieve the desired results in the new SEM account settings. I hope the above two tips can be helpful to everyone.