Identity Resolution – How it Works
More and more businesses online are veering around to the fact that if they have to remain competitive, the one thing that will keep them ahead of the pack is a complete understanding of their customers. A customer’s identity resolution is like the Holy Grail for any business, especially in today’s competitive world. Know what is identity resolution and how it works.
A single unified customer profile is of paramount interest to businesses today. Identity resolution is the way of attributing customer interactions with your business across all touchpoints. Also, let’s not forget that in the last few years, the number of touchpoints customers can have with your business has simply exploded.
Want to know about the customer journey and customer identification? Fill up this short form and allow our experts to talk to you.
In a recent report, Forrester has described identity resolution as, “The process of integrating identifiers across available touchpoints and devices with behavior, transaction, and contextual information into a cohesive and addressable consumer profile for marketing analysis, orchestration, and delivery.”
So far, website cookies alone would do the trick where identity resolution was concerned but consumers are no longer shopping only on their desktops. In fact, more and more of them are now taking to buying from their mobile computing devices. That, and new data privacy laws have made the browser cookie passe.
What is “in” now instead is cross-device Identity resolution tracking. This is a process of collecting and combining data around a single customer in a way that links all his/her devices used back to him/her.
Why is cross-device tracking gaining importance? It is a fact that today, the average person owns multiple devices connected to the World Wide Web. Eg: laptops, smartphones, smart TVs, game consoles, etc.
Device tracking allows marketers to confirm which device and channel were used when a service or product was eventually purchased – whether it was a desktop, mobile, and whether it was from an e-commerce site or a social media channel.
With that, what has effectively happened is the e-commerce company or retailer has also managed to create a “unified” picture of its customer. The process is called “Identity Resolution”. Identity Resolution is the method in which certain unique “identifiers” are used to “connect” all the actions of a buyer to create a single, unified, real-time, customer identity.
So what are these identifiers? The location, the browser used, the device used, the platforms, and the channels, all of these are identifiers that help link the same person, irrespective of the device or channel used. Once a single customer’s identity is “resolved”, it becomes far easier for the marketer to serve up offers to him/her.
Because of Identity Resolution and customer data platform (CDP), each customer can be served up a personalized and unique brand experience.
All about cross-device tracking
So, for the reasons stated above, more than cookies, marketers these days prefer to use cross-device Identity Resolution tracking to maintain a complete profile of every client. Essentially, the latter allows the collection of data and its linking in a way that all devices used by one person are tracked back to him/ her.
The importance of cross-device tracking lies in the fact that it enables business teams to understand if a particular advertisement served on the desktop first resulted in a sale on a mobile; both devices are owned by the same person.
But here’s where things start to get a little tricky. There are two types of cross-device tracking – deterministic and probabilistic tracking – leading to two types of identity resolution – probabilistic and deterministic.
While both these methods involve complex matching across millions of data points as well as access to ALL the digital and device data, depending on the technology and data sets, they can deliver one of two types of matches.
Under deterministic matching, using a deterministic connector like perhaps a hashtag, email address, or username, the records of a particular customer are matched. This approach works best with first-party data.
Why first party? Because it involves the use of personally identifiable information of a customer like his/her email address, home or work address, telephone or credit card numbers, etc. At the end of this exercise, there remains no doubt about the identity of the customer.
Here’s an example: If John Doe uses his email address and logs into his email on his desktop, then uses the same on his Tablet, deterministic matching will show he is the same fellow. Deterministic matching is considered to be more perfect and takes the guesswork out of the tracking game.
But here’s the thing: although a more accurate matching method, it cannot be implemented by all businesses. One reason for that is because of the kind of resources it draws on, and because of the humungous amount of data, implementation is not possible for all the companies. So here’s where probabilistic matching comes in. The other factor that poses a hurdle in the implementation of deterministic matching is that it requires concrete identifiers like social security numbers or driver’s license numbers. But many a time, some companies do not need such information. Or, even if they do, consumers may refuse to submit such sensitive information.
This one tracks not a customer’s identifiable information but relies on algorithms. This method is more “anonymous” than a deterministic one. Essentially, probabilistic matching tracks anonymous data points from different digital elements. Some examples of the latter are device type, operating system, and browsing behavior. It uses a statistical approach to understand whether two customer records represent the same individual or not.
Probabilistic matching generates matching records that gives weightage to the frequency of the occurrence of a data value within a particular division.
But the obvious question is – why would you want to track anonymous data points? Isn’t this a very random way of tracking, especially because you want a near-accurate picture of your buyers?
Businesses must understand that to date, no matching system is perfect. Even for deterministic matching to work successfully you must be 100% sure of the veracity of your data points. If values are suspect or missing, the match may not be accurate.
Conclusion: Identity resolution is the process by which you can create an accurate customer profile. This helps in better personalization and an enhanced brand experience. Whether to use deterministic or probabilistic depends on what your company’s business goals are, what is your marketing trying to achieve, and the available resources at hand. But let’s also not forget that you could use both matching methods together. Since the probabilistic methodology is easily scalable, it can be used to add value and scale when applied to an identity solution that rests on a deterministic matching base.
An Engine that offers sophisticated ID resolution algorithms
Learn how Express Analytics customer data platform Oyster can help your business in customer identity resolution. This includes data cleaning, removal of siloed data, and optimization of customer identity markers.
Liked This Article?
Gain more insights, case studies, information on our product, customer data platform