Optimization Suggestion
What are the ways to optimize data?#
Ways of optimization data includes ad source priority optimization, floor price optimization, traffic grouping optimization, ad scene optimization, and user analysis optimization.
How to do ad source priority optimization?#
According to the eCPM price of each ad source, it should be sorted from high price to low price in the mediation group.
How to do floor price optimization#
Why set a floor price?#
How to set floor price to maximize revenue?#
According to the above example, different floor prices can be set to ad sources. And the ad sources of different ad networks can be used interspersed according to the price, and placed in the "Sort by Price" module at TradPlus's Mediation function.
Waterfall optimization suggestions：
The floor price can be increased as much as possible in the high-price ad sources
When more traffic leaks to the lower-prices ad sources, it is necessary to add some middle-priced ad sources in between
For example: the second layer is 10 dollars and have 1000 impressions, the third layer is 5 dollars, and have 10000 impressions, then add a layer of 8 dollars can be added
You can set up several layers that accept all price from ad networkd to ensure that the app fill rate is higher than 95%
As shown in the picture above, several ad sources can set the same floor proce
How to use Bidding for optimization?#
What is the ad request logic of Bidding+ traditional waterfall?#
- The user opens the app, and the app initializes the TradPlus SDK;
- TradPlus SDK sends a bidding request to the ad network, and the ad network returns the eCPM information; TradPlus SDK merges the bidding PID into the waterfall according to the eCPM of the bidding PID, and generates a new waterfall in combination;
- The TradPlus SDK performs ad requests and impressions according to the new waterfall strategy.
How to use Bidding for optimization?#
It is recommended to use and optimize Bidding and waterfall flow to maximize the monetization revenue. Specifically, there are two points:
For ad network that support bidding, configure the corresponding Bidding PID
For ad network that do not support bidding, configure waterfall normally
Bidding platforms currently supported by TradPlus:
| Ad Network | How to configure the ad source | Link to the details of the ad source configuration |
|---|---|---|
| Meta | 1. In the Meta platfrom, create an ad source according to the conventional method, do not configure the floor price; 2. Fill in the PID in the TradPlus platform and turn on the bidding mode. | Click here to view |
| Vungle | 1. In the Vungle platform, create a bidding ad source (if you cannot create it, please contact the Vungle AM to open the permission); 2. Fill in the PID in the TradPlus platform and turn on the bidding mode. | Click here to view |
| Mintegral | 1. In the Mintegral platform, create a bidding ad source (if you cannot create it, please contact the Mintegral AM to open the permission); 2. Fill in the PID in the TradPlus platform and turn on the bidding mode. | Click here to view |
| Pangle | 1. In the Pangle platform, create a bidding ad source ; 2. Fill in the PID in the TradPlus platform and turn on the bidding mode. | Click here to view |
| MyTarget | 1. In the MyTarget platform, create a bidding ad source; 2. Fill in the PID in the TradPlus platform and turn on the bidding mode. | Click here to view |
| TencentAds | 1. Apply to the TencentAds AM to create bidding ad sources;2. Fill in the PID in the TradPlus platform and turn on the bidding mode. | Click here to view |
| Sigmob | 1. In the Sigmob platform, create a bidding ad source; 2. Fill in the PID in the TradPlus platform and turn on the bidding mode. | Click here to view |
How to use the bidding floor price function for optimization?#
- What is the bidding floor price function?
Bidding floor price is the minimum bid per thousand impressions of the bidding ad sources.
After setting the Bidding floor price, in this round of auctions:
If the price of the winning bidding ad source is higher than the set bidding floor price, it will be integrated into the waterfall to make an ad request.
If the winning bidding ad source price is lower than the set bidding floor price, it will be discarded. Until the ad is reloaded and an ad that is higher than the bidding floor price is obtained.
- How to use the bidding floor price function?
In Mediation - Group - edit, you can set the floor price of the bidding ad source for this mediation group.
All mediation groups support setting bidding floor prices, including default groups.
If no Bidding floor price is set, ads at all prices of the bidding ad source will be accepted by default.
When the currency of the ad network is inconsistent with the currency set by TradPlus, the set bidding floor price will be calculated according to the real-time exchange rate conversion.
- How to use bidding floor price for optimization?
When creating a new bidding ad source for the first time, it is not recommended to configure the bidding floor price.
After the price of bidding ad sources is stable, you can set the bidding floor price for optimization according to the conditions of bidding and waterfall ad sources.
How to use traffic grouping to do optimization?#
TradPlus supports the grouping function of 20+ dimensions, and developers can split groups according to users with different attributes to carry out refined operations
For example: the top 20 level users' user value is high, use the mediation group 1, the user value of the remaining users is low, use the mediation group 2
*If you want to know more details, you can click here
How to use the ad scene to do optimization?#
There are usually several ad scenes for the same ad type of a game. In order to better perform the refined operation of the scene dimension, TradPlus supports the dot data of different ad scene.
For details on how to use ad scences to do optimization, you can click here
The usage of ad scenes#
- Find out whether the number of users entering each scene is in line with expectations, and whether the number of users entering the scene every day has changed
- Know if the ad is ready when the user enters the scene, and whether the ad ready rate has changed
- Find out what percentage of users click on an ad after entering the scene, and whether the ad trigger rate has changed
The calling method of the ad scene#
- Rewarded video
① When a scene has only one video play button：
When the user reaches the following interface (that is, the interface where the user can trigger an ad to play),call the method TpReward.entryAdScenario("sceneId")，and when triggering the ad show, call TpReward.showAd(Activity context,"sceneId")，passing in the same scene id.
How to get Scene id： https://docs.tradplusad.com/en/docs/operation/tradplus_qa/app_management/#how-to-add-ad-scenes%EF%BC%9F
② When a scene has multiple video play buttons:
If you just want to know how many users entered the scene and the total number of buttons/the total number of users in the scene, you can refer to Usage 1.
If you want to know the trigger ratio of each button in order to optimize the reward content corresponding to each button, you can create a scene id for each button. When the user enters the ad scene, any one of the scene ids can be called; when the user triggers the ad shoe, the scene id of the corresponding button should be called.
- Interstitial：
An interstitial ad generally pops up when user jumps from page A to page B.
It is recommended that when page A is ready to jump to page B, call TradPlusInterstitial.entryAdScenario("sceneId")，and when triggering the ad show, call TpInterstitial.showAd(Activity context,"sceneId")，passing in the same scene id.
How to do user analysis optimization?#
Reports-Cohorts Report, is the user's retained ARPU on the Nth day, retained IMP/DAU and retained eCPM reports.
Example of use scenario 1: For example, through analysis, the IMP/DAU on the Nth day has dropped significantly, and the reward of the ad from the Nth day can be adjusted accordingly to slow down the decline in IMP/DAU.
Example of use scenario 2: For example, through analysis, the user's eCPM on the Nth day drops significantly, and different waterfalls can be set to slow down this phenomenon.