Vu's Blog

 1. Why you should plan for performance with Performance Planner - Understand the future spend potential of current Google Ads campaigns to help drive budget decisions - Take advantage of seasonality to capture incremental opportunities  - Set optimal bids and budgets across your campaigns to help ensure that ROI performance is maximized - Find new opportunities to grow your sales volumes with Google Ads 2. What's the Google Ads Performance Planner? --> new forecasting tool that uses machine learning to reveal the possibilities for your Google Ads campaigns.  3. How does the Performance Planner work? - Determines the optimal bids and average daily budget allocations across all of your campaigns to help increase the number of conversions.  4. How does Performance Planner forecast campaign performance? -  Forecasting :  made up of billions of searches per week — powers our forecast engine. - Simulator : engine simulates relevant ad auctions with query level variables, including

1. Display ad formats : Efficiency, control 2. Responsive Display Ads: built for performance, reach, and scale. Key benefits: Automation, Reach, Simplicity. 3. Uploaded ads :  - Image ads : unloadable as a .zip, only run in banner-eligible slots - AMPHTML ads : enable marketers, publishers, and technology providers to deliver faster, and more secure ad experiences across all platforms.

 1. What's automated bidding? some key factors: - Performance - Auction: dynamic nature of our auctions - User journey complexities: based on location, time, device, and so on --> bidding can be done through automation. 2. Benefits of Google Ads automated bidding - Machine learning - save time - Auction-time bidding - Depth of signals used and cross analysis 3. Choose the right bidding strategy - Awareness-based: make sure that your ad is visible for certain queries and even at certain locations on the page.  Bid strategies: Target impression share: make sure your ads are meeting a specific impression share threshold for a specific location  on the search results page: anywhere, top of page, or absolute top of page. - Consideration-focused : drive as many clicks as possible within a set level of spend. Bid strategies: Maximize clicks. - Conversion-focused : tracking actions post-click, valuing conversions equally, and looking to maximize the number of conversions.  Bid strate

 1.  Build awareness :  - Demographic targeting: - Affinity audiences: - Custom Affinity audiences: 2. Influence consideration - In-Market audiences: Reach potential customers. - Custom Intent audiences: create a tailored audience that isn't covered in one of our In-Market. - Similar Audiences:  finds users that are similar to an original remarketing list or other uploaded compatible list. 3. Drive action: available targeting options - Remarketing - Standard remarketing: Show ads to your past visitors as they browse network websites and use network apps (use when doesn't have a data feed). - Dynamic remarketing:  remarketing to the next level with ads.   

 1. Value of Google Display ads - Powerful intent signals: play ads again with most relevant content. - Massive scale - Measurable performance - Easy campaign setup. 2. Display campaign types - Smart Display campaign : accesses the best of Google’s machine and automate nearly all aspects of your Display campaigns. Note: eligible to set up need to  have gotten at least 50 conversions on Display — or at least 100 conversions on Search — in the last 30 days.  Component: Automated bidding, Automated targeting, Automated creatives. - Standard Display campaign : looking for full control over all aspects:  Targeting: can choose three marketing objectives: Build awareness, Influence consideration, Drive action.  Bidding strategies : Enhanced CPC, Maximize Conversions, Target CPA (cost-per-acquisition), Target ROAS (return on ad spend) Ad Formats: Responsive Display ads and Uploaded ads (Image & AMPHTML)

 1. Advance your business goals - drive sale  - Get leads - Increase website visit - influence consideration - Build awareness - promote your app 2. Benefits of Google Ads - Relevance: right people, at the right time, with the right message.  - Control:  control over your budget, change your strategy, adjust your ad. - Results: Pay only for results 3. Google Ads campaign types - Search, Display, Video, Shopping, App. - Specialized campaign types: Local, Hotel, Discovery  4. Optimize your campaign - Device target - location and language targeting - Bidding and budget settings. - Ad extensions

We develop one flutter app we need focus to 06 metric performance below: 1. Startup time to the first frame: Check the time when WidgetsBinding.instance.firstFrameRasterized is true. 2. Frame buildDuration , rasterDuration , and totalSpan 3. Statistics of frame buildDuration (*_frame_build_time_millis): monitoring four stats: average, 90th percentile, 99th percentile, and worst frame build time. 4. Statistics of frame rasterDuration (*_frame_build_time_millis):  monitoring four stats: average, 90th percentile, 99th percentile, and worst frame build time. 5. CPU/GPU usage 6. Release_size_bytes to approximately measure the size of a Flutter app

1. Controlling build() cost:  - Repetitive and costly work in build() --> can be invoked frequently when ancestor Widgets rebuild.  It can be invoked frequently when ancestor Widgets rebuild. - Split them into different Widgets based on encapsulation, Avoid overly large single Widgets. Avoid calling setState () high up in the tree  2. Apply effects only when needed 3. Render grids and lists lazily 4. Build and display frames in 16ms for rendering on a 60Hz display. 60fps leads to a smooth visual experience. 5. Pitfalls - Avoid using the Opacity widget - Avoid putting a subtree in the builder function When using an AnimatedBuilder . - Avoid clipping in an animation. - Avoid using constructors with a concrete List of children (such as Column () or ListView ()) 6. Flutter’s threads - Platform thread : Plugin code runs here - UI thread : executes Dart code in the Dart VM (code that you wrote, and code executed by Flutter’s framework).Creates a layer tree sends to the raster thread to

1. App size of a debug build is large due to the debugging overhead that allows for hot reload and source-level debugging, not representative of a production app end users download. 2. Your upload package to the Play Store and App Store not representative of your end-users’ download size.  3. The stores generally reprocess upload package to target , such as filtering for assets targeting the phone’s DPI, filtering native libraries targeting the phone’s CPU architecture. 4. To analytic size we can use DEV tool or option " analyze-size " when export. example: flutter build apk --analyze-size flutter build ios --analyze-size 5. Reducing app size: use flag  -- split-debug-info tag. Optimize resource, asset.