How to choose promotion channels for efficient delivery?

Traffic is the essence of all business, whether online or offline. Traffic can be obtained for free or for a fee. For a business, paid traffic often achieves its goals more quickly and easily. Therefore, for paid traffic, how to screen channels and deliver them efficiently becomes a basic rule!

For traffic channels, in addition to monetary costs, we must also consider labor costs and time costs, as well as the total amount of traffic in the channel and the quality of the traffic, so as to ensure a reasonable ROI. For example, SEM and information flow channels require professionals to operate because there are too many links and factors involved.

Some third-party channels, such as traditional portals or local website forums in the past, and some new media paid channel advertisements today, have much less manpower and time costs. They are used by some startups or teams that have just been established, or by mature companies that are looking to expand new traffic (there is always a limit to the traffic of a channel, and it costs more to get more traffic).

Third-party channels do not require professional operation. For some advertisers , the easiest way may be to follow the strategy: if competitors are investing, I will invest too. Choose a location, insert a banana picture or other simple advertising form, without too much manpower cost, just wait for traffic in a fool-proof manner. Little do people know that any third-party channel placement must go through the two steps mentioned above: channel investigation and efficient placement and rapid trial and error.

Today, I will use a practical case review to explain in detail how to choose a third-party advertising channel for efficient delivery? (To avoid unnecessary disputes, the channel is a real name, and readers are requested to figure out the characteristics of the channel themselves.)

I have previously managed an in vitro fertilization project. Due to policy reasons, this project had very few advertising channels available in the country. In order to quickly obtain traffic, basically all kinds of traffic channels, big and small, are swept through. Among them, Hudong Baike , as a channel with a large number of users, has its advertising space basically occupied by competitors.

Although there are a large number of users and competitors are flocking to it, it seems that good advertising space is in short supply. However, as a qualified advertiser, you must do a good job of diagnosing and analyzing the channels.

After some explanation from the salesperson and understanding of the advertising PPT and other materials, I learned that one of the larger advertising systems of Hudong Encyclopedia is to advertise on the encyclopedia entry page of keywords . The entire entry page has a word segmentation king (head bananer), a middle listening section, a bottom pop-up window, etc., and there are also many segmented position ads. The price of advertising space is based on position rather than the similarities and differences of keywords.

Therefore, apart from the advertising material due to its own reasons, the biggest influencing factors are the traffic size of the keyword and the advertising position. First, we need to look at the source of traffic from the Hudong Baike channel itself, whether it is direct input, external traffic, search engines , social channels, etc. What is the quality and retention time of users? Secondly, look at the keywords you have chosen, the commercial value of the keywords, the location and other factors. Let’s analyze them one by one according to the process below!

To obtain the traffic of your own website, you can just add statistical codes, such as Baidu Statistics, 51la, CNZZ , etc. However, estimating external websites and understanding the real traffic of competitor websites similar to Alexa, as well as domestic SEO analysis tools (Aizhan, Webmaster Tools, 5118) are very inaccurate. You must know that a website's real traffic comes from many sources, and SEO analysis tools only estimate search traffic and do not include other traffic channels. Generally speaking, similarweb is more accurate in analyzing external websites.

In addition to browser plug-ins, its data sources are also verified and complemented by web crawlers, network packet analysis, web page information, semantic analysis, and statistical algorithms. The final data can be said to be more accurate. Of course there are paid and free versions, and I will review the free version.

Log in to the official website of similarweb (translate it into Chinese using Google ), enter the domain name baike.com, and you can see from this screenshot that the total number of visits is not low, reaching more than 30 million. However, looking at the next three indicators, the average visit time is 47 seconds, which is a bit too short, the number of pages visited is 1.41, and the bounce rate is 84.8%.

This is related to the user usage scenarios of Hudong Encyclopedia itself. After all, using the encyclopedia is to query the details of the entries on the current page, which is similar to the bidding page, so the last three indicators should be questioned. This is worth everyone's attention.

Continuing the analysis, from the above figure, we can see that the vast majority of Hudong Encyclopedia's traffic comes from search, reaching 88.89%, while direct visits, especially external recommendations, are rare. So we can focus on search engine traffic, which is consistent with what sales said about maintaining keyword rankings.

External links account for 3.85%, which has little to do with our advertising analysis. We do this just to make the review more accurate and to let everyone know how to use this tool to identify channels. As can be seen from the above figure, most of the external links of the encyclopedia still come from search engines, but this external link traffic is not search keyword traffic, and the top five exported are Hudong Encyclopedia itself and a data knowledge service platform.

In addition to the fact that most of its traffic comes from search engines, Hudong Baike does not run any paid keyword ads or display ads. The traffic imported from social networking sites is almost negligible, and the proportion of searches for brand words ranks in the top two. This shows that Hudong Baike's own traffic is too single, similar to the search optimization diversion in the traditional PC era for simple advertising monetization.

After clarifying the traffic sources and characteristics of Hudong Encyclopedia, and combining it with its keyword entry advertising system, we can conclude that in order to maximize the acquisition of high-quality traffic, it is nothing more than adopting SEO thinking, for example, keywords with strong commercial nature, high popularity, high ranking, and good position.

Our market at that time was mainly focused on overseas test tubes, with in vitro fertilization as the core word. We started to select words through Aizhan keyword mining .

Some long-tail words do not have entries in the encyclopedia. In addition, the advertising system of Hudong Encyclopedia is based on position, rather than differentiating prices based on the keyword index.

Therefore, we examine the keyword parts of speech one by one for the expanded keywords, that is, the commercial conversion value of the word. The first choice is of course words with country and region roots, such as Thailand and the United States, followed by roots such as hospital and price. The conversion rates of words like forum, website, technology, and process are lower. Then there are the large indexes. It is worth noting that the indexes are divided into PC and mobile, and some advertising positions in the encyclopedia are also distinguished between PC and mobile.

After examining the part of speech and terminal index, we also need to look at the ranking of the selected words, that is, the different rankings of the keyword encyclopedia on PC and mobile terminals. (There is a time difference between the ranking in the figure below and the actual delivery, it is just an example), and the keyword ranking of the encyclopedia is also changing at any time. After all, they also use SEO strategies to optimize entries. We can only refer to the initial ranking when placing advertisements.

The next step is to refer to the click-through rate of the keywords in different terminals and different ranking positions. For example, some words are bidding words, so they are naturally ranked first and tenth. The click-through rates of PC and mobile are also significantly different.

So basically, there are not many clicks on the second and third pages of keyword rankings . The possible clicks can be estimated based on the index and ranking. (The estimated traffic algorithm of the webmaster tool also comes from this)

When the traffic from external search engines is factored in, including index, conversion value, ranking, PC and mobile click-through rates, the remaining factors come from the encyclopedia itself, namely the strength of the location and the material.

As mentioned above, Baidu Encyclopedia provides different system advertisements for PC and mobile, but it all comes down to the position. The price of some ad positions may be several times lower than that of strong positions in a year, but with the same budget, you must spend a lot of money on strong positions with high click-through rates, rather than dispersing funds to buy some small ad positions cheaply, because in the end there may not even be a chance of exposure, and the click IPs brought are very few.

After screening the characteristics of channel traffic and how to choose words, there is an even more important step, which is to estimate whether the cost is worthwhile. For this, we can find benchmarking channels, such as SEM channels and controllable precision channels, to measure whether the encyclopedia is worthwhile.

If you have not worked on SEM channels yourself, you can use the account opening agent and the keyword prices in the promotion background, and the estimated conversion rate of each link to derive the approximate consulting cost of the SEM channel (there is a previous article that explains the data reverse calculation method in detail). For a channel that has no contact, is less controllable and may encounter false traffic, its consulting cost estimation must be lower than that of the SEM precise channel, so that the subsequent ROI is reasonable. Without considering the quality of traffic, the lower the consulting cost, the better.

Therefore, through analyzing the characteristics of channel traffic, selecting keywords, and calculating benchmark channel costs, we can eventually come up with a rough delivery plan: should we deliver or not, and how should we deliver?

The placement of third-party advertising channels is not a simple follow-up strategy. As a qualified advertising placement personnel, it is necessary to analyze and judge in advance. Even if the final result may not be as expected, because there are many factors involved and they are dynamically changing, but after some actual practice, rapid trial and error and rapid improvement are always what Internet advertisers should follow.

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

Product promotion services: APP promotion services, advertising platform, Longyou Games