Asset generation runs allow you to batch-generate multiple app store assets using AI. This powerful feature enables you to quickly create numerous variations of descriptions, feature lists, and other content, giving you plenty of options to test and optimize.
A generation run is a batch process where PressPlay's AI creates multiple app store assets based on your configuration. Each run produces several variations of content that you can review, test, and deploy.
Think of a generation run as a focused creative session where the AI explores different ways to present your app based on:
Your selected AI model (Gemini or Kontext AI)
Your chosen research method
Your competitor analysis configuration
Specific parameters and context you provide
To start generating assets with AI:
Navigate to Asset Generation: Go to your app's asset generation section in PressPlay
Click "New Generation Run": Start the configuration process
Select Asset Types: Choose what you want to generate (descriptions, feature lists, promotional text, etc.)
Configure AI Settings:
Choose your AI model
Select research method
Set generation quantity (how many variations to create)
Provide Context: Add information to guide the AI:
Target keywords
Key features to emphasize
Tone and voice preferences
Target audience details
Any specific requirements or constraints
Review Competitor Settings: Ensure your competitor list is current
Launch Generation: Start the AI generation process
Each generation run can be customized with various parameters:
Asset Types: Which elements to generate (short description, full description, subtitle, etc.)
Platform: iOS, Android, or both
Language: Primary language for generation
Character Limits: Platform-specific constraints for each asset type
Model Selection: Gemini for creativity or Kontext AI for ASO optimization
Research Method: Strategic approach for content generation
Generation Quantity: How many variations to create (typically 5-20)
Temperature/Creativity: How much variation to introduce between generations
App Context: Core value proposition, unique features, target users
Keyword Targets: Specific terms to incorporate naturally
Brand Voice: Tone preferences (professional, playful, inspirational, etc.)
Dos and Don'ts: Specific language to include or avoid
Competitor Focus: Whether to emphasize differentiation
Once you launch a generation run, here's what happens:
Analysis Phase: AI analyzes your app, competitors, and market context
Strategy Development: AI develops multiple messaging angles based on your research method
Content Generation: AI creates multiple variations of each requested asset type
Optimization: Generated content is refined for keyword integration and character limits
Quality Check: Content is validated for compliance and readability
Delivery: All generated assets are made available for review
Most generation runs complete in 2-5 minutes, depending on the number of assets requested.
PressPlay keeps a complete history of all your generation runs:
Run History: View all past generation runs with timestamps
Run Details: See the configuration used for each run
Generated Assets: Browse all content created in each run
Status Tracking: Monitor which assets you've reviewed, approved, or rejected
Performance Data: See how assets from each run perform in experiments
Running multiple generation batches helps you explore different approaches:
Create multiple runs with different configurations:
Run 1 - Gemini + User-Centric: Creative, benefit-focused content
Run 2 - Kontext AI + Keyword-Optimized: ASO-focused content
Run 3 - Gemini + Competitive Research: Differentiation-focused content
Compare results to see which approach resonates best with your audience.
Build on successful runs:
Initial Run: Generate broad variations to explore options
Review and Select: Identify the most promising approaches
Refined Run: Generate new variations based on what worked
A/B Test: Test refined options in experiments
Final Run: Create polished versions of winning concepts
For efficiency, PressPlay provides quick access to your most recent generation run:
Last Run Summary: Quick overview of your latest generation
Recent Assets: Immediate access to newly generated content
Continue Working: Pick up where you left off in the review process
Regenerate: Quickly create more variations with the same configuration
Start with Multiple Runs: Generate assets using different methods and models initially
Provide Rich Context: The more context you give the AI, the better the output
Generate in Batches: Create 10-15 variations per run for good diversity
Review Systematically: Use a consistent framework to evaluate generated assets
Track Configuration: Note which settings produce the best results
Iterate Based on Data: Let experiment results guide your next generation runs
Don't Over-Generate: Quality evaluation is more valuable than endless options
Generate comprehensive assets for your initial app store presence:
Create 3-4 runs with different research methods
Generate descriptions, feature lists, and promotional text
Review and select the strongest positioning
Test top options with small user groups before full launch
Refresh your app store presence for holidays or events:
Generate seasonal variations of your core messaging
Create time-limited promotional content
Test holiday-themed vs. standard messaging
Update assets when you add major functionality:
Generate descriptions that highlight new features
Create variations emphasizing different new capabilities
Test feature-focused vs. benefit-focused approaches
Create localized content for new markets:
Generate culturally-adapted messaging
Test different positioning for different markets
Create market-specific variations
React to market changes or new competitors:
Add new competitors to your analysis
Generate differentiation-focused content
Test new positioning angles
Not all generated assets will be perfect. Expect a range of quality:
Ready to Use: 20-30% may be immediately usable with minimal edits
Good Foundation: 40-50% will be solid starting points requiring refinement
Inspiration Only: 20-30% may contain useful ideas but need significant rework
This distribution is normal and expected. The goal is to generate more good options faster than manual creation would allow.
Provide more specific context about your unique value proposition
Add more detailed competitor analysis
Try a different research method
Include more specific dos and don'ts
Explicitly list features to emphasize in the context field
Ensure your app profile is complete and current
Try feature-driven research method
Reduce generation quantity for more focused output
Provide clearer brand voice guidelines
Try adjusting creativity/temperature settings
Switch to keyword-optimized research method
Try Kontext AI if using Gemini
Provide more specific keyword targets
Learn how to Review Generated Assets effectively
Understand AI Models to optimize your generation settings
Configure Research Methods for better results
Set up Experiments to test generated content
