All-in-one retail media buying platform unifying campaign management across multiple advertising networks

Marketing agencies managed each retail platform separately, requiring duplicate campaign setup and preventing unified oversight of advertising spend and performance across networks.

Solution: Integrated Triffecta platform with Amazon Ads, Criteo, and CitrusAd, via custom APIs and connectors, enabling users to create and manage ad campaigns across multiple retailers.

Retail media networks differ in functionality. Without a unified logic layer, features could not be presented consistently to the end user.

Solution: Researched documentation of all retail media networks and conducted interviews with potential customers to identify the most valuable cross-platform features. We unified these capabilities into a single workflow and built an abstraction layer that maps each platform’s native features into a shared model.

Agencies could not compare performance across Amazon Ads, Criteo, and CitrusAd, or monitor spend in a single place.

Solution: Built a unified analytics engine that ingests data from all integrated media platforms through scheduled API jobs, normalizes metrics into a shared data model stored in a centralized analytics layer. Developed interactive, customizable dashboards enabling real-time performance tracking, retailer comparison, and spend analysis.

Retail media generates large volumes of high-frequency data (impressions, clicks, ATC events, conversions, spend updates), creating significant challenges for ingestion, processing, and storage.

Solution: Implemented large-scale data storage and processing capabilities using Clickhouse for analytical workloads and MongoDB for metadata. Background services ingest data continuously from external APIs, ensuring high throughput and fast analytical queries without compromising performance.

A monolithic or tightly coupled system would increase the risk of platform downtime, deployment failures, and performance bottlenecks as the product scales.

Solution: Adopted a microservices-based, cloud-native architecture that isolated critical platform components (campaign management, data ingestion, reporting, and user management) into independently deployable services. This separation allowed us to update or scale individual services without impacting the entire system.

Existing retail media software required extensive manual configuration with different interfaces, prolonging campaign launches and increasing operational overhead.

Solution: Introduced an AI-enabled optimization engine for quick ad campaign setup, requiring a customer to provide basic campaign information (product selection, campaign start and end dates, and budget), and the algorithm generates targeted keywords, tells where to place ads, and how long to run them.