Russian Super Retail Chain Adds Azure Confidential Computing to Customer Data Platform to Improve Targeting Accuracy
Moscow, July 10, 2020 — Magnit and Aggregion launched Azure Confidential Computing data processing capabilities as part of the Magnit decentralized customer data platform (CDP). The platform, developed by Aggregion in cooperation with Microsoft, provides the FMCG brands with a range of products aimed at improving the audience targeting and analytics in digital marketing. The platform provides professional access to the transactional data of Magnit, one of Russia’s largest supermarket chains with 22,000 supermarkets, 35 million customers enrolled in the loyalty program, and over 6 billion identified purchases.
Marketing specialists from top advertising agencies use the self-service portal to construct desired consumer segments based on hundreds of attributes: social demographics, children and pets, interests, events, geotargeting and locations, travels, budgets, preferences, visits to retail, behavioral, retail purchases, and other categories. The attributes are derived from first party data provided by Magnit and other companies. The goal is to build an audience that most closely reflects the portrait of the consumer of a particular brand’s product, thus increasing the success rate for advertising campaigns.
Results of campaigns conducted through the platform with leading FMCG brands show that conversion — the percentage of customers that click on the ad — increases by up to 100% percent and the audience targeting accuracy is getting a five to tenfold boost depending on the product advertised.
The platform is designed to empower marketing professionals and offers a range of products and services based on multi-source data: advanced audience targeting and segmentation, advertising campaign management, online-to-offline conversion reporting, omnichannel marketing, and ecosystem data enrichment and matching scenarios.
MATCHING AND PORTRAIT 360
The core feature of the platform is its open decentralized architecture that allows the companies to resolve the issue of trust and work jointly with their combined data. Not only is there more data available to create a digital profile of each consumer, but it is also different types of data provided by different types of data suppliers: retail, e-commerce, banks, telecom, and others. The result is an accurate Customer Portrait 360, a multi-dimensional image of a customer’s interests and preferences.
The decentralized protocol enables the instances of AFS — Aggregion Full Stack, core technology deployed within the companies’ own IT landscapes — to interact without the need for any central hub or storage. This ensures that the data never leaves the premises of the owner company. All operations are written on blockchain and the blockchain nodes are stored within each AFS instance.
The critical component of this technology is the mechanism for matching hashed customer ID’s between datasets. The AFS algorithm stored within secure enclaves in each company’ own instance allows it to securely perform matching calculations and produce correlations between datasets. The combined dataset — the audience segment — is dynamic, generated per request through the instance and is not stored on any single central server.
The absence of a central server requires a reliable solution to protect the environment from possible leakages of source data and ensure that even the enclave owner does not have access to the calculations performed there.
This solution is Azure Confidential Computing technology, which allows the protection of data-in-use through isolating computations to a hardware-based trusted execution environment (TEE). While data is traditionally encrypted at rest and in transit, confidential computing protects your data while it’s being processed. A TEE provides a protected container by securing a portion of the hardware’s processor and memory.
Azure confidential computing infrastructure is currently comprised of a specialty SKU of virtual machines (VMs). These VMs run on Intel processors with Software Guard Extension (Intel SGX). Intel SGX is the component that allows the increased protection that we light up with confidential computing.