RichRelevance leverages a robust, fully-owned global infrastructure to accelerate and innovate personalization for leading brands and retailers. Our enterprise-class infrastructure, combined with the first open AI stack for experience personalization results in a high-performance, scalable and extremely reliable platform. Plus, our extensive use of customized Open Source software—including Hadoop, hBase, Parquet, Avro, Spark, various proxy services, Postgres and Kafka — provides an agility, modularity and speed to market that has helped to make us the #1 name in personalization.
$35 B
incremental sales delivered
1 BILLION
decisions driven daily
5 PB
Multi-tenant Hadoop installation
Infrastructure Built for the Enterprise
Blazing fast page loads
RichRelevance is the only personalization provider with enterprise scale infrastructure. Our 14 global data centers ensure industry leading speeds and uptime. Read how we delivered over 630 million page views for 2016 Black Friday.
Proven and efficient
We’ve chosen to use proven, efficient and more accessible ways to develop our software in order to deliver a high quality and cost efficient personalization platform to you.
We use Hadoop, Spark, Kafka, Parquet, Pig, Hbase, and R as our underlying open source technologies for reliable, scalable, distributed computing.
Giving back
We actively contribute knowledge and expertise to improving these collaborative tools so that the larger open source community/ecosystem, which includes academics, researchers, and non-profits, can innovate as well.
22
patents issued
14
global data centers
630 M
page views on Black Friday
EARLY ADOPTERS
Of leading edge Big Data technology
18,600
requests/sec. during shopping peak
#1
robust infrastructure with industry's fastest response times
7,500
avg. requests/sec. (Twitter averages 9,100 tweets/sec.)
Let's dig a little deeper
Data Science
RichRelevance uses customer behavior to discover interests, purchases, and interactions with your business across all channels.
User profiles are updated in real time to allow our advanced predictive models to make effective decisions.
A comprehensive ensemble of supervised and unsupervised predictive models future proofs your investment and ensures the right context-relevant algorithm is applied to every decision. To combat cold starts, we use hierarchical and K-means clustering and other classification systems to leverage the Wisdom of the Crowd (WoC). Not all crowds are created equal, so our engine automatically identifies sub-populations with different types of behavior in order to take advantage of those differences. We use factorization machines and matrices to predict affinities, as well as deep learning neural networks to mine unstructured text and images to better understand and leverage shopper intent and sentiment.
Our algorithms dynamically tune their behavior (using real-time metrics and Gaussian process and other regression techniques) to optimize performance for each client, and also to adapt to seasonality and variable customer behavior.
RichRelevance comes out-of the box with 125+ pre-built predictive models encompassing the full spectrum of available AI/ML algorithms. We use a goal-seeking master algorithm, or Experience Optimizer, to pick the best choice for each interaction. The Data Science Workbench extends the capability to your data and data science. Once the Experience Optimizer determines the most effective algorithm for each customer and context, we use Gaussian Process Regression (aka Kriging) to fine-tune the weights and parameters for the selected algorithm to ensure the greatest impact on the customer interaction and potential revenue. With the Experience Optimizer, personalization is no longer constrained by manual parameter selection; it is dynamically automated and optimized in real time, ensuring the personal experiences delivered to each and every customer will generate the most effective response.
Consumer Privacy
Your customers’ privacy is paramount to us and we ensure that we remain in lockstep with the evolving regulatory and legislative landscape.
Our privacy policy, which can be found here, and practices have been awarded Trust Arc™ Privacy Seal.
Fully compliant with the U.S.-EU Safe Harbor Framework and the U.S.-Swiss Safe Harbor Framework
Certified for adherence to the Safe Harbor Privacy Principles of notice, choice, onward transfer, security, data integrity, access, and enforcement.
GDPR encourages businesses to use anonymization technology when possible to avoid the processing of personal data. In providing its services, RichRelevance uses an anonymous identifier that does not store shoppers’ personally identifiable information (PII). This identifier isn’t connected to an individual’s personal data, and so would not typically be considered personal data under the definition provided in the GDPR.
In the normal course of operation, RichRelevance has no ability to connect such identifiers back to any of the personal data held by our customer about that individual shopper. Thus, when our customers transmit a website visitor’s anonymous identifier to the RichRelevance personalization service, it is doing so in a manner that prevents RichRelevance from having any ability to know who the user is.
Learn more about our commitment to GDPR compliance here.
Data Science
Your customers’ data are safe with us and we have taken following steps to secure it.
We do not capture or sell any PII/PCI or financial data.
All of RichRelevance’s co-location and hosted state-of-the-art facilities are SSAE16/ISO27001/ISO9000/SOC1-2-3 compliant.
All of our customers’ data are logically siloed to prevent unintended access by irrelevant parties.
Remote access is setup with two-step authentication, and thoroughly monitored and logged to prevent intrusion.
Our data centers detect DDOS and brute-force attacks and, block the offending IPs.