Advertising Agencies on AWS
With 175+ services and 13 years of proven expertise, AWS has everything agencies need to innovate faster, optimize costs, and drive client success with cutting-edge advertising technology, intelligent creative and petabyte-scale consumer analytics.
Why AWS?
Innovate faster with more machine learning and analytics capabilities
Accelerate innovation for identity resolution, advertising analytics, consumer analysis, and creative development by choosing from more compute, analytics and machine learning capabilities than any other cloud provider.
Drive efficiency with Amazon Advertising
Agencies and marketers can easily set up secure data integrations, analyze campaign performance, and onboard partners to Amazon Advertising using solutions such as AWS Lake Formation and Real-Time Advertising Analytics.
Achieve unmatched cost-efficiency
Save up to 70% on petabyte-scale event pipelines and analytics workloads; and uncover more opportunities to reduce costs through a wider selection of compute, storage, analytics, and machine learning capabilities than any other cloud provider.
Performance at scales that would otherwise break the internet
Unmatched elasticity, reliability, availability and security from the most widely-adopted cloud.
Use cases
-
Consumer & Advertising Analytics
-
Identity Graph
-
Machine Learning
-
Real-Time Bidding
-
Consumer & Advertising Analytics
-
Consumer & Advertising Analytics
Analyze billions of advertising and customer events per day at up to 70% lower costs than co-lo and on-premises.
Track customer behavior, advertising performance and drive efficiency across the marketing funnel with efficient data collection, processing and analysis. AWS enables companies to centralize both anonymized data for data management platforms (DMPs) and PII data for customer data platforms (CDPs) with the most secure and privacy-compliant cloud services available.
Why AWS?
Cost-Efficiency with EC2 Spot Instances
Customers processing billions of events per day for advertising analytics and data pipelines can save up to 70% on costs over on-premises data centers by using tools such as AWS Lake Formation, Amazon EMR, Amazon Kinesis and AWS Glue catalog. And no other cloud provider has a comparable compute alternative to Amazon EC2 Spot Instances, which enables companies to save up to 90% on On-Demand instances.
Breadth & depth of analytics tools
No cloud has more purpose-built services for database, analytics, storage or machine learning than AWS—especially services for storage, processing, real-time analytics, forecasting and visualization.
Business agility
Set up data processing, data lake and warehouse in days with AWS solutions for real-time advertising analytics.
Amazon Advertising Ecosystem
Increase agility and simplify data integrations using the same cloud as Amazon Advertising and the most popular cloud provider for agencies and advertising technology partners.
Case studies
Nielsen
Nielsen built a data lake capable of storing 30 petabytes of data in Amazon Simple Storage Service (S3) and leveraging Amazon Redshift, AWS Lambda, and Amazon EMR. According to Brown, this allowed Nielsen to grow from measuring 40,000 households daily to more than 30 million households each day. "Working with AWS and the services they provide allows us to do all of that at a much faster pace, with much greater velocity than we could have ever achieved before," says Scott Brown, general manager of TV & Audio at Nielsen.
GroupM
GroupM – the world's largest media investment group with more than $113.8 billion in billings - discusses their unique challenges and how they created a fast and reliable data analytics platform using Amazon Redshift, Matillion and Tableau.
Annalect
“We built a solution on AWS to manage, enhance, and analyze event log data from our ad technology partners. This scalable, on-demand architecture has proven to be a high-performance solution for our analytics users and is very cost-effective.” – Eric Whitney, CTO, Annalect
Learn more »
Quantcast
Used Amazon EC2 Spot Instances to save up to 60 percent for its QCLearn machine learning system, and spending about 25 percent less on Amazon EC2 Spot Instances than it would have on Amazon EC2 reserved instances for its real-time bidding solution.
Related products
-
Identity Graph
-
Identity Graph
Identity resolution & graph enrichment at any scale
Map and unify billions of relationships and query customer data with millisecond-latency using AWS Neptune— the only fully managed graph database in the cloud —along with managed machine learning and serverless computing tools from AWS. Extend audience reach and improve cross-device performance and audience activations. Reduce resources with fully managed services for identity graph and machine learning.
Why AWS?
Purpose-Built Databases
AWS is the only cloud with a purpose-built, fully managed graph database service available worldwide—Amazon Neptune—a high-performance graph database engine optimized for storing billions of relationships and querying the graph with millisecond latency. Customers can also store and query data from Amazon DynamoDB, a fast and flexible NoSQL database that delivers single-digit millisecond performance at any scale.
Unmatched Cost-Efficiency at Scale
Reduce total cost of ownership (TCO) by storing and quickly querying hundreds of billions of nodes and edges with lower latency and lower costs for storage than other databases
Reliability at Scale
AWS has a ready-to-deploy solution to help customers set up an identity graph quickly to map behavioral data and enrich profiles for audience segmentation, customer 360s and retargeting. AWS has more experience with big data workloads than any other cloud—more than 10,000 data lakes run on AWS.
Case studies
Salesforce DMP
Salesforce DMP has “acquired and scaled a global customer base without any concern about the limitations on the amount of data it can process. Our platform captures data from interactions from three billion devices and about 40 billion page views each month.”
Zeta Global
Zeta Global uses Amazon Neptune to map 450M requests per day into 1B user profiles that they use for people-based targeting and audience enrichment. “By leveraging [Amazon] Neptune and other AWS services, we were able to achieve a cost-efficient data platform at scale in a very short period of time."
- Sasikala Singamaneni, Engineering Manager, Zeta Global
Related products
-
Machine Learning
-
Machine Learning
Accelerate ML innovation in advertising technology by using unmatched tools for computer vision, model-training and deployment at massive scale
Create a new generation of ML-powered applications for contextual analysis, next-offer decisioning, ad fraud detection, dynamic creative optimization, and traffic filtering using Amazon SageMaker, a fully managed service that provides every developer and data scientist with the ability to build, train, and deploy machine learning (ML) models quickly, and Amazon Rekognition, a managed computer vision service that enables you to identify the objects and scenes in images that are specific to your business needs.
Solutions
Traffic & Cost-Optimization
Reduce data transfer on bidding workloads costs by up to 75% using Amazon SageMaker and Amazon EMR.
Contextual Analysis & Brand Safety
Use Amazon SageMaker and Amazon Rekognition to analyze advertising creative and media content at massive scale
Dynamic Creative Optimization
Use Amazon SageMaker, Amazon Rekognition and Amazon EMR to programmatically create creative assets that drive personalization in ad serving.
Fraud Analysis
Analyze billions of ad events per day at single-digit millisecond latency to detect fraud and bots using Amazon EMR, Amazon Kinesis and Amazon SageMaker.
Why AWS?
Accelerate time-to-market with the most ML capabilities of any cloud provider
Customers can choose from 12 pre-trained AI services for computer vision, language, recommendations, and forecasting; use Amazon SageMaker to quickly build, train and deploy machine learning models at scale; or build custom models with support for all the popular deep-learning frameworks including TensorFlow, Apache MXNet and PyTorch.
Most cost-effective for ML and analytics
Amazon SageMaker GroundTruth enables customers to build highly accurate training datasets using machine learning and reduce data labeling costs by up to 70%. Amazon EMR has native support for Amazon EC2 Spot and Reserved Instances, which enables can also save 50-80% on the cost of the underlying instances for ML workloads.
Most experience running ML workloads
Tens of thousands of companies use AWS for more machine learning, more than twice as many customers run ML on AWS than any other cloud provider, and 81% of all deep learning projects and 85% of TensorFlow workloads in the cloud happen on AWS.
Case studies
Smaato
Smaato sought to reduce its networking costs on outbound bid requests sent to advertising partners, so it developed a traffic-filtering service that uses machine learning with Amazon SageMaker and Amazon EMR to predict advertising partners’ likelihood to bid in auctions.
“We used to pay very high amount of money in outbound traffic for our bid requests,” says François Guerraz, systems architect at Smaato. “Thanks to machine learning, we’ve been able to predict effectively which demand partners are going to be interested in specific ad requests and only send the request to a specific subset of the demand partners. We have achieved a 75% cost reduction in outbound network costs.”
Mobvista
Mobvista used Amazon EC2 Spot Instances, to save nearly CNY 10 million (USD $1.4 million) a year in IT costs and reduce the cost per request by 60%. “Ultimate simplicity and extreme scalability are what I conceive of AWS. The use of AWS has enabled Mobvista to expand and innovate quickly." Cai Chao, Vice President of Technology and Chief Architect”
White Ops
White Ops runs MediaGuard and other applications on Amazon EC2 and uses Amazon EMR for data processing. “We built our AWS environment to achieve low latencies for our products,” says Walker. “Specifically, we optimized on the most powerful Amazon EC2 instance types for the best performance.” White Ops also stores multiple petabytes of data on Amazon Simple Storage Service (Amazon S3), and it relies on AWS services such as Amazon CloudTrail to enhance its security posture.
“Running our bot-detection solutions on AWS, we can help customers respond to threats in minutes or hours instead of days.” Sarah Walker, Vice President of Engineering, White Ops
GumGum
GumGum uses AWS to process more than 1 billion events and reach 400 million consumers per day as they view images and content across the web. “With AWS, we can smoothly process billions of events daily with 24/7 reliability, and can scale quickly to meet spikes in demand.”
- Ken Weiner, Chief Technology Officer
Curalate
Curalate runs Apache MXNet on AWS EC2 Instances to analyze “millions and millions” of images stored in Amazon S3 to automate the collection and curation of user-generated social content in the pipeline, and then connects it to the right product in a catalog.
Related products
-
Real-Time Bidding
-
Real-Time Bidding
Achieve unmatched business agility and cost-efficiency at ultra-low latency
With more compute capabilities, networking services and cost-optimization tools than anyone, AWS continues to be the cloud standard for advertising platforms that buy, sell and serve hundreds of billions of ads per day at single-digit millisecond latency. Improve business agility, unlock development velocity and achieve up to 25% cost-efficiency compared to on-premises data centers by running real-time bidding in the cloud with AWS.
Why AWS?
Unmatched Compute Capabilities
AWS is the only cloud provider with 100 Gbps network connectivity for standard instances, and provides more types and sizes of bare metal, memory-intensive, and storage-intensive compute instances than any other cloud provider—enabling workloads like real-time bidding to run cost-effectively with ultra-high throughput and ultra-low latency.
Cost-Efficiency with EC2 Spot Instances
No other cloud provider has a comparable compute alternative to Amazon EC2 Spot Instances, which enables companies to save up to 90% on On-Demand instances and helps advertising technology customers achieve cost-savings of up to 25% compared for compute-intensive programmatic advertising and bidding workloads.
Largest Global Availability
AWS has 69 availability zones in 22 regions — more than any other cloud provider —with 199 edge locations across 37 countries and 78 cities; and plans for 16 more Availability Zones and five more Regions.
Breadth of networking & load balancing tools
AWS offers more capabilities for load balancing and networking high-volume, single-digit millisecond latency advertising workloads than any other cloud provider, including tools such as Elastic Load Balancing and AWS Global Accelerator.
Capabilities for Containers
AWS has more capabilities for containers than any other cloud provider, which is why 81% of all containers in the cloud are running on AWS. We recently cut prices in half for our managed Kubernetes service, Amazon EKS—another reason why 84% of Kubernetes workloads than run in the cloud today run on AWS.
Experience Supporting RTB Migrations
AWS has over a decade of experience helping companies handle virtually any scaled cloud workload for advertising and marketing technology—including RTB migrations and low-latency programmatic advertising workloads. Our leadership position and sheer volume of usage means AWS has learned lessons about reliability, security, and speed at usage tiers that other cloud providers simply haven’t reached yet.
Case studies
The Trade Desk
The Trade Desk used more than 70 AWS services to build real-time bidding sites in Tokyo, Hong Kong, Singapore, and Frankfurt, while processing 800 billion queries per day and 10 million per second. “It used to take somewhere between 3-6 months to build out a new site when we were entering a new market. With bidding in the cloud, that’s cut down to 2-3 weeks. There’s much more agility.”
- Zak Stengel, SVP Engineering, The Trade Desk
Mobvista
Mobvista used Amazon EC2 Spot Instances, to save nearly CNY 10 million (USD $1.4 million) a year in IT costs and reduce the cost per request by 60%. “Ultimate simplicity and extreme scalability are what I conceive of AWS. The use of AWS has enabled Mobvista to expand and innovate quickly." Cai Chao, Vice President of Technology and Chief Architect”
AdRoll
AdRoll uses a combination of On-Demand, Reserved, and Spot Instances to operate the RTB platform. By taking advantage of the Spot market for intensive, non-time-critical computation processes, AdRoll is able to reduce costs by nearly 80 percent compared to On-Demand Instances. “Leveraging the AWS platform, we’ve been able to seamlessly scale our infrastructure, better serve our customers across the globe, and reduce our fixed costs by 75% and operational costs by 83%. Because we’re running on AWS, we’re able to focus 95 percent of our staff on new product development.”
- Valentino Volonghi, CTO, AdRoll
Quantcast
Quantcast: By using Amazon EC2 Spot Instances for its critical business systems, Quantcast reduced its operational costs in several ways, including saving up to 60 percent for its QCLearn machine learning system, and spending about 25 percent less on Amazon EC2 Spot Instances than it would have on Amazon EC2 reserved instances for its real-time bidding solution. "As we roll out more infrastructure to AWS, Amazon EC2 Spot Instances are helping us control costs and scale our systems to meet demand.”
- Leah Blank, Senior Systems Engineer, Quantcast
Smaato
Smaato reduces networking costs for auction platform by 75% using Amazon SageMaker and EMR.
Beeswax
Beeswax uses Amazon EC2 Auto Scaling and Amazon EC2 Spot instances to reduce costs on compute instances for its bidding platform. “If there’s one thing from this talk you should take away, it’s to go check out [Amazon] EC2 Spot,” says Ram Kumar Rengaswamy, co-founder and CTO of Beeswax. “If you care about operating costs, Spot is going to save you a ton of money.”
Related products
Resources
Whitepaper: Building a Real Time Bidding Platform on AWS
This whitepaper helps you understand real-time bidding (RTB) and the services available in AWS that can be used for RTB. This paper will showcase the RTB platform reference architecture used by customers today, as well as additional resources to get started with building an RTB platform on AWS
AWS Reference Architecture
AWS provides services and infrastructure to build reliable, fault-tolerant, and highly available ad serving platforms in the cloud. In this document, we describe the two main parts of such a system: ad serving infrastructure and click-through collection featuring a data analysis cluster.
Building an identity graph with Amazon Neptune
This post provides an overview of how to build a customer identity graph on AWS. It reviews key business drivers, challenges, use cases, customer success stories and the benefits of the solution. You also walk through the solution, sample data model, AWS CloudFormation templates, and other technical components that you can use to kick-start your development.
Lambda@Edge for Cookie-Syncing
Cookie syncing involves mapping cookies from one AdTech platform to another. While the concept itself is straightforward, because of high request rates, cookie syncing infrastructure design requires some careful consideration. With serverless technology such as Lambda@Edge, you now have another option for implementing cookie syncing, that includes using a Lambda function that executes in CloudFront edge locations.