General UA Tips

1. Basic Terminology

• Campaign: The advertising subject that requires promotion, which can be a mobile application or a web page.

• Offer: A specific advertising unit under the campaign that needs promotion. Multiple offers can be created for comparison testing based on factors like regions, bid prices, or creatives.

  • Customize the offer name in this format: compaign name_product name_targeted geo_OS_mtg.

• eCPM: By default, it refers to revenue per thousand ad impressions. On Mintegral, eCPM measures the competitiveness of an offer.

  • eCPM = eCPI*IVR*1000 = eCPI*CTR*CVR*1000

• eCPI: The effective cost per install, a metric that measures the cost of acquiring a new user.

  • eCPI = Cost/Conversion

• CTR: Click-through rate, the ratio of clicks to impressions in online advertising.

• • CTR = Click/Impression

• CVR: Conversion rate, the ratio of users who become valid activations, registrations, or even paying users after clicking on an advertisement to the number of clicks.

  • CVR = Conversion/Click

• IVR: Impression-to-conversion rate, a metric used to evaluate the performance of advertising creatives.

  • IVR = CTR * CVR

 

2. Pre-Launch Preparation

a. Creative Preparation Suggestions

I. Learning Phase

• • For detailed creative instructions, please refer to our Mintegral Creative Operation Guide first.

  • It is recommended to prepare full sizes of creatives according to Mintegral Creative Specs to cover more ad types and help you get the most out of your creatives and grow your user base faster.

  • It is recommended that you apply the creatives to the corresponding creative sets according to the theme/language/duration/gameplay method, etc., based on the actual delivery needs. If you divide the creative sets by language, you can group English creatives into one creative set (targeting English-speaking countries), Spanish into one set (targeting Spanish-speaking countries), and Portuguese into one set (targeting Portuguese-speaking countries).

  • If you do not have specific creative classification and creative testing requirements, it is recommended that you create 1 creative set at the beginning of delivery. It is recommended to upload 3-5 sets of video creatives (horizontal + vertical), 3-5 images (of which 1200x627 size is required), and 1-2 playables to the creative set.

  • In the early stage of delivery, please do not upload too many creatives at one time. Too many creatives will affect the learning speed of the algorithm in the initial stage, reduce the effectiveness of each creative data model, and thus lead to limited scale. Do not upload creatives multiple times within 3 days. If you need to test creatives, you can iterate on the creatives for testing after the overall delivery situation is stable.

  • It's important to apply playables alongside your videos, since they provide a different user experience, target and convert a wider variety of users, and bring added value to your UA activities, referring to the Playable Test Guidance for details.

II. Stabilized Phase

• • Iterate creative sets every 2-4 weeks and retain those well-performing ads. If there are too many non-performing ads in the offer, you could remove the ads with few impressions.
  • For ads that have been uploaded to the Mintegral system and have been approved for 5-7 days, if Mintegral cannot automatically generate significant volume, you could try deleting and re-uploading them.

 

   b. Campaign/Offer Configuration Suggestions

• Please refer to our Step-by-step Video Guide first for detailed configuration steps.

• Avoid narrow targeting that may affect volume. It is recommended to have an all-inclusive approach at the initial stage, without fine-grained settings based on user gender, age, device model, network, or industry tags.

• Create one campaign per product and merge country targeting into a global offer. Set bids and budgets for each country. Multiple types of creatives can be uploaded to the same offer (e.g., Product A targets countries B and C and uploads image D, video E, and playables F, all managed under Offer Y under Campaign X). Having more than three offers for the same product in the same region may result in significant competition that limits volume.

 

3. Initial Delivery Phase

a. Bidding Adjustment Suggestions

• To quickly accumulate learning period data, it is recommended to set the initial bids at 1.2-1.5 times the expected level, which will give more impressions and help achieve more impressions during the learning period.

• Avoid stopping or reducing bids frequently in the first 3 days of running the campaign, as this can affect the algorithm's learning period data model. Adjustments should be made with an interval of at least three days between each adjustment.

• Increasing bids stimulates the algorithm to deliver more volume. If you want to continue to expand your reach, it is recommended to actively raise bids.

• The recommended range for price reductions is within 10%. Reducing prices beyond this range may cause significant fluctuations in volume or even result in a serious drop in volume. It is recommended to observe the performance for three days after a price reduction and adjust accordingly.

 

b.  Campaign Performance Evaluation

💡 How to determine if the advertising campaign is performing well using data?

• Overall status: Check the offer's daily impression and conversion data for a positive trend and sustainable volume.✅

• Top regions: Campaign's top regions are sufficient Mintegral traffic, such as the United States, Europe, Asia, the Middle East, and South America.✅

• Ad types: Focus on video ads, including interstitial videos and rewarded videos, as they are the primary ad types.✅

• Ad output types: If dynamic Endcard/Playable creatives are uploaded, prioritize them as they drive volume effectively.✅

• Top sub-sources: Identify multiple sub-sources that generate significant volume and align with the product's target audience.✅

 

4. Stable Delivery Period

a. Daily Adjustment Suggestions

• If the volume is sufficient and the quality meets expectations, it's recommended to raise more budget to expand more quality sources and users.

• Bid adjustment range is controlled within 10%, and the adjustment cycle is recommended to be at least 3 days or more. Frequent or large-scale bid adjustments may cause algorithm disorder and lead to volume fluctuations.

• Regularly monitor the IVR (Impression-to-Conversion Rate) of creatives and analyze performance across different ad types. Optimize and expand creative content with the best IVR to continuously improve campaign performance.

• The creative iteration cycle is recommended not to be about 4 weeks. Retain high-performing creatives, and delete creatives that have no impressions or volumes promptly, except for necessary creatives.

• If the campaign has been running for more than 3 months and experiences a significant decline in volume, it is recommended to update the creative direction and implement appropriate bid increases to re-stimulate algorithm learning.

 

5. Troubleshooting & Deep Optimization

a. Troubleshooting for No Traffic

• Check if the user targeting range is too narrow, potentially limiting the traffic scope (e.g., user gender, age, device model, industry tags, etc.).

• Ensure whether all ad types are selected, and whether creative types and sizes are complete and approved.

• Check if the current bid is below the recommended range for the specific region.

• Check whether the delivery region is in an area where Mintegral's traffic is insufficient. Regions with low traffic volume may find it hard to achieve the desired volume.

• Check if there are too many offers created for a single product, as this may result in the dispersion of traffic and consequently impact the volume attainment.

• In cases where impressions are high but there is no conversion data, verify if the conversion link has any anomalies. Also, check if the impressions primarily occur in image ad types or on icons.

• Check if dynamic endcards/playable creatives have been uploaded and ensure they have generated enough impressions to drive volume.

 

b. Troubleshooting for Volume Drop

• Check if there were any changes to targeting settings, sub-source blacklists, or narrowing of targeting, and if the time of dropped volumes corresponds to these adjustments.

• Check if there were iterations of high-performing creatives, and if the volume drop aligns with the timing of these iterations.

• Verify if there were any adjustments to bids or budgets, and if the drop in volume aligns with these changes. If the adjustment range was significant, consider restoring the bid price.

• Determine if the drop in volume coincides with the end of working days or major holidays in the delivery region. Drops following holidays may be part of normal traffic trends.

• Identify if there is a noticeable decrease in volume from the primary volume-driving sources. If so, consider increasing bids for these specific sources or expanding to other high-quality sources.

• Evaluate if the IVR (Impression-to-Conversion Rate) of video creatives has significantly decreased due to creative fatigue. If so, iterate on the creatives.

• Check if the volume drop aligns with the timing of product update iteration or if there is a significant decline in CVR across multiple advertising platforms. In such cases, internal product investigation is recommended.

 

c. Deep Optimization

1. Sub-Source Level Adjustments:

   • 📌*Sub-source Optimization

2. Target ROAS Campaign:

   • 📌*Target ROAS Optimizer

3. Target CPE Campaign:

   • 📌*Target CPE Optimizer

*For more detailed information regarding regional benchmarks, bidding strategies, or feature introductions, please feel free to reach out to your account manager for consultation.