Utilising an ID-Graph to enhance customer identification

In today’s digital landscape, companies face a recurring challenge: integrating diverse customer and prospect data to derive actionable insights. Despite seeming straightforward, this task is often complicated by the presence of multiple identifiers—hundreds of first-party cookie IDs, multiple connected devices, and several email addresses and customer IDs per person.

A customer data platform can help address these challenges by providing a cost-effective, scalable, secure, and highly available solution for real-time response to customer signals, advertising targeting, and marketing journey orchestration.

The necessity for lawful processing under GDPR based on consent and/or legitimate interest further complicates the situation. Enter identity graphs – a revolutionary tool in customer identification strategies. This blog will answer the question of what is an identity graph, explore the significance of identity graphs – or ID-Graphs –, their types, how to build and maintain them, and their benefits.


The essential in brief:

  • ID-Graphs integrate data from multiple sources, creating a unified customer view that enhances data accuracy, profile building, real-time activations, and multitouch attribution, leading to more effective marketing strategies.
  • There are different types of ID-Graphs—deterministic vs. probabilistic, first-party vs. third-party, real-time vs. non-real-time, and single domain vs. cross-domain—each with specific applications and benefits for customer identification and personalisation.
  • Effective identity graphs require regular updates, accurate data integration, and compliance with privacy regulations. Key methods include login identification, self-identification, click identification, cross-domain identification, and network identification.

Why is an ID-Graph relevant?

In the evolving landscape of customer identification, identity graphs have become indispensable. Traditionally, companies have employed various methods to tackle the complexities of identifying customers accurately.

The earliest approach, Entity Resolution (ER), focused on identifying and linking different records that referred to the same entity across various datasets, even when data was inconsistent or incomplete. This method laid the groundwork for more sophisticated identity resolution frameworks.

As technology advanced, Cookie-Based Identity Resolution emerged. This method tracked user behaviour across sessions and websites using cookies. However, the effectiveness of this approach has waned due to stringent privacy laws and the depreciation of third-party cookies, making it difficult to maintain a consistent view of the same customer across multiple devices.

To overcome these limitations, Profile-Based Identity Resolution was developed. This method combined customer data and behavioural data to create unified user profiles. While more effective, it often resulted in “monster profiles” filled with too many or inaccurate identifiers.

The most advanced method today is ID Graph-Based Identity Resolution. ID-Graphs manage relationships between individuals, devices, and groups like households or companies, maintaining accuracy by reflecting dynamic changes and interactions over time. Utilizing a graph database for these ID-Graphs allows for efficient management of interconnected entities, real-time data updates, and querying highly connected datasets with milliseconds latency, ensuring the most accurate and up-to-date identity resolution.





What Are different types of ID-Graphs?

Understanding the various types of identity graphs is essential for businesses looking to enhance their customer identification strategies. Here, we explore the differences between deterministic and probabilistic ID-Graphs, the distinctions between first-party and third-party ID-Graphs, the importance of real-time updates, and the capabilities of single domain versus cross domain ID-Graphs.


Deterministic vs. Probabilistic ID-Graphs

Deterministic ID-Graphs rely on explicit, direct identifiers to link devices or identifiers to the same individual. Using unique, unchanging data points such as email addresses, phone numbers, or login credentials ensures high accuracy. Deterministic matching is a vital component of identity resolution solutions, providing a reliable and precise way to map customer identities.

Unlike deterministic graphs, probabilistic ID-Graphs use statistical algorithms to infer connections between devices and identifiers likely belonging to the same person. By analysing behavioural data and patterns, these graphs can link individual devices and offline identifiers with a high degree of confidence. Signals such as similar personal identifiers or actions originating from the same IP address, location, or Wi-Fi network are utilized in the process of 'fuzzy matching' identity resolution.

While probabilistic matching is useful for scenarios where direct identifiers are unavailable, such as offline interactions, the inherent uncertainty makes them less reliable for real-time personalisation.


First-Party vs. Third-Party ID-Graphs

Built and maintained using data directly collected by a company from its own customers and prospects, First-Party ID-Graphs leverage first-party data gathered through direct interactions with the company’s digital properties. This method ensures a high level of accuracy and reliability, as the data points are collected first-hand.

These ID-Graphs often function as an online database, providing a comprehensive collection of personal and anonymous data related to individual customers. This helps businesses personalise advertisements and better understand and target customers based on updated information.

On the other hand, created by external entities using data from various sources, Third-Party ID-Graphs can be less reliable and often come with privacy concerns due to the lack of a direct consumer relationship. These graphs are typically built using data aggregated from different websites, apps, and other digital platforms, making it challenging to ensure data accuracy and compliance with privacy regulations.


Real-Time vs. Non-Real-Time ID-Graphs

Real-Time ID-Graphs are updated instantaneously when new identification events are detected, allowing for accurate and up-to-date identity graph solutions. Real-time ID-Graphs enable businesses to deliver real-time personalisation and timely marketing efforts, adapting quickly to the evolving customer journey. This capability is particularly beneficial for industries like digital advertising and streaming services, where immediate responsiveness is critical.

In contrast, Non-Real-Time ID-Graphs process identification events with a delay, which can lead to inaccuracies and limit their effectiveness for real-time personalisation. While still useful for historical data analysis and audience segmentation, the lag in updates can hinder the ability to provide a seamless, personalised experience to individual customers.


Single Domain vs. Cross Domain ID-Graphs

Single domain ID-graphs resolve identities within a single domain, such as a website or app. This approach is straightforward and effective for companies with a singular online presence. For instance, a retail website can use a single domain ID-graph to track user interactions, from browsing to purchasing, providing a cohesive customer experience within that domain.

Cross-domain ID-graphs, however, link identities across multiple domains, offering a comprehensive view of customer interactions. Typical examples are: 

  • multicategory retailers with own sites for groceries, fashion, electrical appliances, and furniture,
  • multi-brand manufacturers or media houses with different sites for each of their brands,
  • multi-site companies like hotel groups or companies operating leisure parks that have individual websites for each location or
  • companies that separate a content-based media site from their e-commerce site.

A company with multiple websites and apps can use a cross-domain ID-graph to recognise users across all these touchpoints and perform overlap analysis. Sharing identification events significantly increases user identification and personalisation, helping companies target customers better and identify opportunities for cross-selling.


Building and maintaining a First-Party ID-Graph

Building and maintaining a robust first-party identity graph is crucial for businesses aiming to enhance their customer identification strategies. This section provides an overview of the various methods used to link customer identities through first-party data and explains the importance of regular updates to ensure the integrity and accuracy of the identity graph.

Choosing a reliable identity graph vendor is essential for obtaining accurate customer data. A trustworthy vendor should be transparent in sharing all analysed data sets and campaign insights. This transparency helps businesses understand their customers on a more personal level and trust the vendor's process.


Types of identification on a user-level

Linking device/browser combinations or apps where logins occur with the ID-Graph via first-party cookies is referred to as login identification. Common examples include app logins, self-service portals, and e-shop logins.

On a more personal level, an individual customer might also provide identifiers like emails or phone numbers, linking their devices or apps to the ID-Graph. This self-identification can happen through newsletter subscriptions, form submissions, or pop-ups. Digital identity can also be transferred between channels via click-throughs, or click identification, linking devices or apps to the ID-Graph.


Identification through domains and networks

On a more technical level, cross-domain identification recognises users across multiple domains, creating a comprehensive multi-domain profile. This allows shared identifications across domains, enhancing user recognition and personalisation. Cross-domain identity graphs work to provide an encompassing view of customer interactions across platforms.

In a process called network identification or network signal, network providers use IP address resolution to link devices to the ID-Graph without user input. Non-network providers can use services like Utiq’s Authentic Consent Service to leverage network signals.


Maintaining an ID-Graph

Regular updates are crucial for maintaining an accurate ID-Graph. As identifiers change, the graph must realign device associations and purge outdated profiles. Effective maintenance ensures a consistent and reliable view of users, supporting identity resolution frameworks and real-time personalisation.

In summary, building and maintaining a first-party identity graph involves login identification, self identification, click identification, cross-domain identification, and network identification. These methods collectively enhance identity resolution, improve customer data accuracy, and support personalised marketing efforts.


Willing to learn more? Schedule a meeting with our staff:

Free Assessment


What are the key benefits of an ID-Graph?

ID-Graphs play a pivotal role in identity resolution by consolidating data from multiple sources to create a unified view of each customer. Below you find an overview on the key benefits of identity graphs and how they enhance data accuracy, profile building, activations, and attribution:

Data Enabler

An identity graph acts as a crucial data enabler by integrating data from various sources into a single, unified customer view. This approach eliminates duplicate records, leading to more accurate insights and a comprehensive analytics tool.


Identity graphs enable accurate multitouch attribution by tracking customer interactions across multiple touchpoints throughout the customer journey. This capability helps identify the most effective marketing tactics, allowing businesses to optimise their marketing strategies.


An accurate identity graph enhances multichannel activations by linking more identifiers to customer data profiles, thereby increasing reach across various channels. The detailed profiles enable precision targeting, reducing marketing waste and ensuring consistent messaging and cross-device experiences. Moreover, identity graphs support behavioural triggers, allowing for real-time personalisation and timely campaign activations based on real-time customer interactions.

Profile Building

Identity graphs build rich, detailed customer data profiles by integrating diverse customer data points, including behavioural data and offline identifiers. This detailed profiling supports personalised experiences, predictive analytics, and lookalike modelling, enhancing the effectiveness of marketing efforts.


Overcoming challenges and ensuring success

Successfully building and maintaining an ID-Graph involves addressing several key challenges while ensuring compliance, data quality, and strategic integration. Here’s an overview of how to navigate these challenges to enhance your identity resolution framework.


Challenge: Privacy concerns and compliance

When creating an ID-Graph, it is crucial to obtain user consent and clearly communicate its purpose. Users should understand the benefits of giving consent, such as receiving personalised experiences and relevant marketing content.

Additionally, it’s important to provide options for those who prefer to remain anonymous. Ensuring transparency and offering choices helps build trust and compliance with privacy regulations.


Challenge: Data quality and integration

Ensuring the quality and seamless integration of customer data is fundamental to the success of an ID-Graph. Poor data quality can lead to inaccurate identity resolution, resulting in fragmented customer profiles and ineffective marketing strategies. There are several strategies that can help:

  • Inconsistent Data Formats: Standardise data formats across all sources to maintain consistency.
  • Duplicate Profiles: Implement de-duplication processes to eliminate duplicate records and achieve a unified customer view.
  • Incomplete Data: Use data enrichment techniques to fill in missing information and enhance customer profiles.
  • Data Accuracy: Regularly update and validate data points to maintain accuracy.
  • Data Freshness: Ensure real-time or near-real-time updates to keep data current.
  • Data Silos: Break down data silos by centralising data management, enabling a holistic view of customer identities across multiple devices and platforms.


Challenge: Stand-alone or embedded?

Deciding whether to implement a stand-alone ID-Graph or embed identity resolution within existing applications depends on your specific needs. A stand-alone ID-Graph offers centralised control, consistency, and scalability, making it easier to manage and update.

On the other hand, embedding identity resolution within existing applications can simplify integration and reduce costs. Evaluate your current and future needs to determine the best approach for your business.


Conclusion: Embrace the ID-Graph Revolution!

Identity graphs should be considered foundational to your business strategy rather than an optional luxury. Identity graphs enhance basic capabilities such as analytics and support sophisticated use cases over time. Investing in identity graph solutions early can create a self-reinforcing loop, improving customer engagement, trust, and overall marketing effectiveness.

For companies pursuing a first-party data strategy, an identity graph is essential for creating consistent user journeys and tailored customer experiences. Start now with a first-party, deterministic, real-time identity graph. Consider a stand-alone solution for comprehensive capabilities or an embedded approach for simpler integration. Develop a data and marketing strategy to fully utilize the potential of an ID-Graph.

Embrace the ID-Graph revolution and transform your customer identification and engagement strategies!


Get advice now:

 Free Assessment 



FAQ: ID-Graph

Here, we have compiled some frequently asked questions and answers about ID-graphs. 

How does an ID-Graph differ from other data management solutions?
An ID-Graph focuses on identity resolution by consolidating data from multiple sources to create a unified view of each customer. Unlike traditional databases, which store data in silos, an ID-Graph links data points through unique identifiers, enabling accurate tracking of customer interactions across devices and channels.
How can businesses benefit from utilising ID-Graphs in their marketing strategies?
ID-Graphs offer precise targeting and personalised experiences, enhancing customer engagement and loyalty. They enable cross-device attribution, providing insights into the customer journey and optimising marketing efforts. Real-time personalisation is also possible, improving the effectiveness of marketing campaigns and return on investment.
What are common challenges associated with implementing ID-Graph solutions?
Common challenges include ensuring data quality by standardising formats, eliminating duplicates, and enriching incomplete data. Integrating data from multiple sources and maintaining data accuracy and freshness are also crucial. Additionally, businesses must address privacy concerns and ensure regulatory compliance.
Are there regulatory considerations I need to know when using ID-Graphs?
Yes, businesses must comply with privacy laws such as GDPR, which require explicit user consent for data collection and processing. Companies must clearly explain the purpose of the ID-Graph, provide anonymity options, and implement robust data protection measures to safeguard customer information and ensure compliance.

About the author

Dirk Rohweder

Dirk Rohweder: COO and Founder | Teavaro