# ADVPERF04-BP01 Choose a data management strategy that matches your availability, latency, and access requirements
<a name="advperf04-bp01"></a>

 Customers need to have a clear data management strategy for their advertising workload datastores. The factors to consider are latency needs, availability needs which will help them chose the right AWS data service 

## Implementation guidance
<a name="implementation-guidance-47"></a>

 The following are the most common data stores available in adtech: 
+  **User data:** Demographic data (age, gender, and location), behavioral data (browsing history, interests, and purchase history), and device data (device type, operating system, and browser). 
+  **Audience data:** Segmentation data (personas and target audiences) and geo-location data (IP addresses and GPS coordinates). 
+  **Campaign data:** Ad creative data (like images, videos, and text), ad placement data (websites, apps, and platforms), and campaign performance data (impressions, clicks, and conversions) 
+  **Inventory data:** Publisher data (website or app details and traffic data) and ad space data (ad sizes, formats, or placements) 
+  **Pricing and bidding data:** Bid data (bid prices and bid strategies) and auction data (bid landscape and winning bids). 
+  **Third-party data:** Data from Data Management Platforms (DMPs) and data from data exchanges or marketplaces. 
+  **Analytics and reporting data:** Conversion data (sales, leads, and actions), attribution data (tracking user journeys), and engagement data (view-through rates and dwell times)  

 For latency, consider the following: 
+  **Low-latency data (real-time or near real-time):** This data needs to be processed and acted upon within milliseconds to ensure optimal ad delivery, real-time bidding, and accurate tracking of user interactions. 
  +  Bid (bid requests, bid responses, and auction data) 
  +  User (device data, location data, and contextual data) 
  +  Ad impression (ad requests and ad responses) 
  +  Real-time campaign performance (clicks, impressions, and conversions) 
+  **Medium-latency data (near real-time or batch processing):** This data can be processed in near real-time (within minutes or hours) or in batches, as it is used for audience targeting, campaign optimization, and attribution analysis. 
  +  User behavior (browsing history and interests) 
  +  Audience segmentation 
  +  Campaign optimization (performance metrics and engagement data) 
  +  Attribution (user journeys and conversion paths) 
+  **High-latency data (batch processing or offline):** This data can be processed in batches or offline, as it is typically used for analysis, reporting, and long-term decision-making rather than real-time ad delivery or optimization. 
  +  Historical campaign 
  +  Detailed analytics and reporting 
  +  Third-party (from DMPs or data exchanges) 
  +  Ad creative (images and videos) 

## Resources
<a name="resources-42"></a>
+  [Architecture III: Picking the Right Data Store for Your Workload](https://aws.amazon.com/blogs/startups/how-to-pick-the-right-data-store-for-your-workload-1/) 
+ [ Amazon DynamoDB: Ad tech use cases and design patterns ](https://aws.amazon.com/blogs/database/amazon-dynamodb-ad-tech-use-cases-and-design-patterns/)