What challenges did Adspert face before AI implementation?

Automate product pricing optimization in ecommerce marketplaces to maximize visibility and profits. Handle large volumes of data from multiple platforms and competitors in near-real time. Balance between sales volume and profit margin for pricing recommendations.

What was the impact of the AI solution?

Achieved near-real-time repricing of products leading to increased visibility and maximized profits for sellers. Enabled cost-efficient model training on billions of daily transactions and seamless deployment via SageMaker. Improved seller competitiveness and profitability in dynamic ecommerce environments.

Source

Evidence: Low25/100

Adspert Ecommerce Pricing Optimization Using Amazon SageMaker

Use case typePricing optimizationUpdated Jun 13, 2026

Adspert, a Berlin-based ISV, developed a pricing tool to optimize product prices dynamically on ecommerce marketplaces to maximize product visibility and seller profits. Challenges included the need for near-real-time pricing decisions to increase product visibility amidst rapidly changing prices. The solution used Amazon SageMaker to train and deploy a Scikit-Learn Random Forest model predicting product visibility based on competitive pricing and other factors. AWS Glue performed ETL and data preparation, AWS Lambda optimized price recommendations, and Amazon RDS, S3, and SQS were used for data management and processing. The architecture enabled automated training and near-real-time inference to recommend optimal prices balancing sales volume and margins, increasing seller profitability.

Organization: Adspert
Industry: Retail
Location: Germany
Published: August 2022

Reported outcomes

Strategic outcomes

New product / capabilityAutomated dynamic pricing optimizationSpeed & agilityNear-real-time repricing of productsCompetitive differentiationImproved seller competitivenessCost efficiencyCost-efficient model training and deployment

Why do we believe this deployment?Customer identity, provider attribution, maturity, and source checks

Customer: Adspert
Provider: AWS
Maturity: Unknown
Linked source: AWS Machine Learning Blog

No explicit deployment-stage evidence found.

Customer identity supportedSource describes one deploymentMaturity evidence evaluated

Primary read

Use case focus

Showing 3 of 3

1Pricing Optimization
2Real-time Inference
3Machine Learning

Automate product pricing optimization in ecommerce marketplaces to maximize visibility and profits.
Handle large volumes of data from multiple platforms and competitors in near-real time.
Balance between sales volume and profit margin for pricing recommendations.

Leveraged Amazon SageMaker for training a Random Forest model predicting product visibility according to pricing and competitor data.
Implemented AWS Glue for complex, large-scale ETL tasks and real-time data transformation.
Used AWS Lambda to optimize price recommendations and update marketplace product prices through API.
Managed data ingestion and storage with Amazon RDS, S3, and SQS utilizing change data capture for real-time processing.

Technologies

Amazon SageMaker AWS Glue AWS Lambda Amazon S3 Amazon RDS Amazon SQS AWS Database Migration Service

Achieved near-real-time repricing of products leading to increased visibility and maximized profits for sellers.
Enabled cost-efficient model training on billions of daily transactions and seamless deployment via SageMaker.
Improved seller competitiveness and profitability in dynamic ecommerce environments.

Architecture

The system ingests product listing data into Amazon RDS and an Amazon SQS queue, uses AWS Database Migration Service for change data capture into a data lake on Amazon S3. AWS Glue performs ETL and transforms large-scale data for training and real-time inference with an Amazon SageMaker endpoint hosting a Scikit-Learn Random Forest model. AWS Lambda optimizes final pricing and updates marketplace APIs.

Sources & evidence1

Evidence: Low25/100Evidence strength