This article is the second part of the 'Enrich the Digital Profile of your Customer' series. We look at it from two perspectives – firstly, from the standpoint of anonymous matching, and secondly, scale.

In this blog, we share learnings from our work with several large enterprises in enabling them to augment their customer information with outward origins (social media, external organisations, sales and marketing, loyalty). We hope that some of the insights shared through this blog will provide you with 'food for thought as you pause and reflect on your strategy and transformation journey and plan for your future.

How to distinguish unique customers during information-sharing without disclosing their identities?

A sound data matching solution should match information within the organisation and match against outside attributes. To unravel this case, the provider (source) needs the power to process raw, dirty data in diverse formats, standardising and securing the information.

The information providers and consumers would fancy aggregating and sharing the information to unleash its marketing powerfully! Leading to the question of preserving anonymity: How to distinguish unique customers during information-sharing without disclosing their identities? It is like you and your designated partner trying to find each other in a masquerade ball without anybody else knowing who you are. And it ought to be accomplished at a large scale - for millions of pairs simultaneously.

Anonymous Data Matching

HorizonX Data Profiling Platform solves this predicament ethically, legally, and professionally. Each provider and consumer processes outgoing and incoming information in two steps:

1. Masquerade - An irreversible information processing operation that translates readable information to a format unreadable yet matchable
2. Exchange - sends known information and receives new information.

All participating (systems) on this platform are enforced to follow the agreed masquerading protocol. Consequently, the masqueraded information can be matched in a verbatim manner without disclosing its meaning (i.e. being translated back, actually there is no way of translating back.) Therefore, all meaningful information resides inside providers/consumers and remains masqueraded en route.

Execution at Scale

Another key driving factor for a robust data profiling approach is the ability to scale, primarily serving external systems. How fast can the information be de-identified and matched against the source dataset and information shared via an API

Our solution does not aim at finding a needle in a haystack. Instead, it seeks to instantly find millions of hays in a haystack, all with unique characteristics, made possible because masqueraded information is distinguishable and non-identifiable. It can be safely processed in large-scale database operations preserving anonymity.

Should you require any support with your customer identification journey, please feel free to reach out to our team:

• Huaxin Xu (Lead Data Scientist)
• Vinu Kumar (Chief Technologist) or
• Utsav Sinha (CEO - HORIZONX)

Need help?

Get in touch with our team of passionate, expert and customer-obsessed practitioners, focusing on innovation and invention on your behalf. We operate as the technical partner for your business, working as an extension of your digital teams. We follow a combination of the Lean and Agile methodologies and a transparent approach to deliver real value to our customers.