What channels are used to purchase traffic for mobile games? This article tells you!

I recently spent two weeks traveling around the country, and had a lot of exchanges (mainly drinking) with friends who buy volume and distribute in Chengdu , Shanghai, Guangzhou, Shenzhen, Xiamen, and Beijing. In addition to in-depth exchanges with many veterans in buying volume, I also came into contact with many new teams that have just joined the army of buying volume and distributing. In this article, I still hope to continue to talk about the topic of buying volume. The content is suitable for all ages, and I hope it will be more or less helpful to veterans and newcomers.

• In-depth analysis of distribution channel data will reveal why you cannot afford users!
• The rich are coming, and the blockbuster of the year is about to be staged!
• Let’s talk about cheating again!

As we all know, both Android and iOS are facing the problem of increasingly expensive users. Let’s analyze it based on the data from the distribution channels.

First, let’s look at some data. I selected the top 386 channels and advertising platforms that everyone has launched in the past month (platforms with more than 500 cumulative APP activations in the past month), among which:

① The top five delivery channels account for 62.51% of the total delivery activation volume

②31 intermediate channels accounted for 24.32% of the total activation volume

③ 350 long-tail channels account for 13.17% of the total activation volume.

1. Analysis of top channels

From the data, we can see that the top channels are currently the most competitive in terms of buying volume and distribution. The top channels mentioned here include Toutiao , Guangdiantong , Zhihuitui (including Zhihuitui external links), UC Toutiao and WeChat MP . For experienced users, basically all of the above platforms can be used.

For those who are not experienced, the first lesson is to learn how to buy users from the above channels. Then problems arise, such as the poor conversion rate of Toutiao, the unaffordable purchase of Zhihuitui, the expensive WeChat MP, and how much money it costs to "maintain" an account, etc.

Here is another data to look at the competition among the top channels. Among all advertisers that have placed ads in the past month:

① More than 50% of customers have placed ads on Toutiao;

② 39% of customers have placed ads on Zhihuitui (including Zhihuitui external links);

③ 32% of customers have placed ads on UC Headlines;

④ 28% of customers have placed ads on Guangdiantong;

⑤ 10% of customers have placed ads on WeChat MP.

The reason why many advertisers give priority to advertising through top channels is that top channels have more users, better conversion rates, and considerable returns, and all platforms basically support audience targeting.

At present, we have found that advertisers' products and materials are highly homogenized (materials with good conversion rates can be easily copied), which leads to the platform users being washed back and forth and users are easily aesthetically fatigued.

When we look at the homogeneity of materials, it actually extends to the issue of increasing user expectations and managing user expectations in terms of return on investment. This involves the relationship between cost and advertising CTR , clicks to landing pages , downloads, installations, activations, and recycling.

To increase ad CTR, can I raise the bid? No, raising the bid can at most increase the exposure rate, but to increase CTR, you can only start with audience targeting and creative materials.

Assuming that the audience targeting is fixed and the exposure position is the same, whether the material can attract people depends on the creative designers (I personally think that the scenarios that programmatic creativity can adapt to are limited, and it has certain significance for specific industries, such as e-commerce ). The direct assessment indicator is CTR. Many advertisers will try their best to improve CTR, and many times there will be a problem of raising user expectations. For example, the material is a high-definition Maserati, but when downloaded, it is found to be a second-hand bicycle. The problem reflected in the data is that both retention and recovery are poor, so don't overdo it when raising user expectations appropriately .

Next, platforms like Zhihuitui will need to have landing pages. The design of the landing page will also affect the user's further conversion. Whether to further raise user expectations or deepen the impression of the product through the landing page, you can do more A/B Testing. The loss of users when jumping to the App Store or application center is also related to the size of the package. Making the package as small as possible can also increase the conversion rate . In short, the process from clicking on the ad to activating payment is a process of managing user expectations, and each step of conversion requires careful data analysis.

I have said a few more words above about the issue of user expectations for delivery management. Let’s get back to the topic: why are users increasingly unable to afford it?

From Q3 2016 to the end of Q1 2017, during this half year, we monitored the following data:

① The number of mobile performance-based advertisers increased by more than 40% compared to Q3 last year;

② The number of real deduplication users in the market increased by less than 20%;

③ More than 62% of the users acquired by the newly added advertisers and historical advertisers came from the top channels.

So now the question we all have to face is:

① The users are still the same users, but there are nearly twice as many advertisers competing for these users. The annoying thing is that these new advertisers are all very wealthy and their bids are getting higher and higher.

②Core users are washed away, products are highly homogenized, and user conversion rates gradually deteriorate.

③ In the future, competition in the top channels will become more intense, because more large companies and wealthy people will enter the market. Advertisers with limited funds or no obviously differentiated products will face greater challenges.

④How to find more high-quality media and channels is a great test for the media manager in the delivery team.

2. Analysis of middle and long-tail channels

So, the question arises again. After talking about the head channels for so long, there are still about 40% of the middle and long-tail channels. Why can't we afford them? Here we roughly distinguish the remaining channels, the four traditional portals, Hero App, and third-party advertising platforms:

1. Four major portals

Regarding the four major portals ( Sina , NetEase, SOHU, eh? Why are there three major portals?), the conclusion I came to is the same: KPI pressure (covering my face).

Because many of the three major portals sell on a CPM basis, and even some of the high-quality positions are reserved by others, the amount that can be obtained is not large. After comparison, many advertisers found that it is not cheaper than the top channels. Of course, the conversion and recovery are actually not bad, but it is very frustrating to balance CAC and ROI.

2. Hero App

For some Hero Apps, I think the main problem lies in user matching . Most Hero Apps do not fully understand their own user attributes. Especially when it comes to performance advertising monetization, fully understanding one's users is the basis for maximizing traffic value. There are also issues such as user habits. For example, I open xx weather every morning to see if there is haze. If there is haze, I will immediately become discouraged and close the App. If there is no haze and it is a sunny day, I will also close the App immediately.

3. Third-party advertising platforms for performance

As for performance-based third-party advertising platforms, most of them connect to BAT and various media's ADX, or package some media themselves, but most of them collect long-tail volumes that others cannot sell. This is a huge test for users' labeling capabilities and high-quality media acquisition capabilities. Third-party advertising platforms with good conversions and large volumes are rare .

After getting the top channels, how to gain a foothold in the middle channels is an advanced ability of the delivery team. If you can't do it yourself, I recommend that you find a reliable agent first. If you have this need, you can contact our BD colleagues. Of course, there are many ways to obtain traffic, such as video APP, WAP site, real CPA channels, etc. It is common for a graduated delivery team to have more than 30 channels for a single product at the same time (the most important thing is to directly bypass fraudulent traffic)

In summary, the price for users in the top channels will definitely get higher and higher, and the space to be explored in the middle and long-tail channels is full of imagination . Although it is already a red ocean, I believe that the domestic buying volume market is still in a very early stage, and everyone needs to practice basic skills and fight to the end.

After last year's "small-scale" trials and tribulations, reliable sources have revealed that this year a number of large companies and various hidden tycoons will enter the market with large sums of money, spending more than 50 million yuan on each product to snatch users.

This time in Guangzhou and Shenzhen, I met a tycoon who had 150 million in cash ready to invest in an SLG product. This reminded me of the web game era a few years ago, when manufacturers such as Kunlun and Efun released products in Taiwan. They used land, sea and air bombardment to grab all core users in the shortest time with the best-looking products.

In addition, some companies in the first tier that rely on purchasing volume to distribute content are attracting users at any cost (one manufacturer has offered CPC as low as 25 RMB, you can feel it), and the CPA cost has soared to more than 100 RMB (mortals, tremble).

In my previous article, I mentioned that the number of new users per week for a single first-tier product can be stable at over 500,000. This makes the survival space of B+ products that could have survived increasingly narrow. Apple is now very strict about cutting off Apple Pay and will remove any product it finds. This has left some unconventional buying teams, which could have had a profit margin of over 15%, helpless.

Ultimately, domestic volume-buying and distribution will be a blockbuster competition among the local tycoons, and volume buying in the top channels will be the first battlefield . Whoever can occupy a place will not only need an experienced volume-buying team, reliable products, and sufficient funds, but also courage and strategy, otherwise they will only become cannon fodder or pioneers.

From the perspective of the healthy development of the industry, we are looking forward to whether the second-tier platforms can bring out the value of performance-based traffic and whether advertisers can tap into the second-tier users. Reyun Data will release a product in the future, and its goal is to balance the overlapping users of the top channels and second-tier platforms and reduce the cost of buying volume. Stay tuned.

I have spent a lot of time before talking about the problem of some advertising platforms stealing attribution and organic volume by brushing clicks or reporting impressions as clicks, but some people still seem to understand it vaguely. Today I will take some time to talk about it.

If you find that an advertising platform asks you to settle on a CPA basis, but does not disclose where the ads will appear, and when you look at the data, the amount of "clicks" is very large, but the activation rate is only a few thousandths , you need to pay attention, because you may have encountered a fake CPA channel.

Regarding the Last Click attribution model, you can read my previous article. The platform mentioned above takes advantage of the loopholes in the Last Click model and cheats by brushing a large number of clicks or reporting exposures as clicks. Once such cheating occurs, your ads will either have no exposure and rely purely on machine brushing, or have no clicks but only exposures.

Cheating that relies purely on machine-generated content requires the cheating platform to accumulate enough real device IDs (usually tens of millions or more), and then use the machine to simulate the device IDs into clicks based on a certain pattern (some have basically no pattern) and report them to advertisers or third-party monitoring platforms.

Because advertisers or third-party monitoring platforms only receive an Http request, it is difficult to determine whether it is a real click. However, there is a certain probability that these users will download the APP from the App Store. At this time, the cheating platform will "grab" a user. If advertisers place ads on multiple advertising platforms at the same time, there is a certain probability that users of other advertising platforms will be snatched away by cheating platforms.

Cheating with only exposure but no clicks means that the advertising platform reports the exposure of the advertisement as clicks. The problems this brings are: first, it is absolutely unfair for everyone to use click attribution, and second, there is also the problem of snatching the organic volume from other platforms.

Therefore, it is recommended that you carefully distinguish between true and false CPA channels, and never think that you are getting an advantage. We found that a large number of delivery teams are actually still in this trap.

Mobile application product promotion service: APP promotion service Qinggua Media advertising

This article was compiled and published by the author @白冬立(Qinggua Media). Please indicate the author information and source when reprinting! Site Map