Draft:Identity graph
| Submission declined on 8 March 2025 by Caleb Stanford (talk). This submission reads more like an essay than an encyclopedia article. Submissions should summarise information in secondary, reliable sources and not contain opinions or original research. Please write about the topic from a neutral point of view in an encyclopedic manner.
Where to get help
How to improve a draft
You can also browse Wikipedia:Featured articles and Wikipedia:Good articles to find examples of Wikipedia's best writing on topics similar to your proposed article. Improving your odds of a speedy review To improve your odds of a faster review, tag your draft with relevant WikiProject tags using the button below. This will let reviewers know a new draft has been submitted in their area of interest. For instance, if you wrote about a female astronomer, you would want to add the Biography, Astronomy, and Women scientists tags. Editor resources
|  |
Comment: Can you provide context for where this concept came to be, why it is important? In what field is this studied? Who is using it and why? What other concepts does it relate to? "Graph databases" should be wikilinked. Caleb Stanford (talk) 21:22, 8 March 2025 (UTC)
An identity graph (also known as ID graph, identity spine, or identity network) is a database that links customer identifiers across different sources to create a unified profile in order for businesses to understand their customers holistically.[1] These databases are used by businesses to help personalize advertisements based on individual devices.[2] Identity graphs were also created because of the increasing number of platforms that customers have access to and for businesses to personalize all of these interactions with customers.[3] Some example identifiers that can be linked together are usernames, phone numbers, purchase histories, and loyalty card numbers.[4]
To accomplish identity resolution across different sources to generate the identity graph, either deterministic or probabilities methods, or a combination of both, are used.[5] Graph databases are typically used to support identity graphs.[1]
Identity graphs are one type of cookie alternative.[6]
Creation
[edit]Identity graphs are generally created in three steps:[7]
- Ingest event data from different identifiers
- Train and build a machine learning model
- Construct the graph
The identifiers are clustered at either the household or individual level. Deterministic and probabilistic identity resolution is then done to unify the identifiers.[7] Identity graphs are typically built using graph databases.[8][9]
Identifiers
[edit]Identity graphs are built up from a number of identifiers, such as:[4][5][7]
- Usernames
- Hashed email addresses
- Phone numbers
- Account numbers
- Credit card numbers
- Purchase histories
- Loyalty card numbers
- First-party cookies
- Third-party cookies
- Alternative IDs
- Personally identifiable information (PII)
- Customer ID
- IP addresses
- Identifiers for advertising
- Mobile ad IDs (MAIDs)
- Connected TV ID
Examples
[edit]Netflix and Amazon are able to recommend more relevant shows and products using browser history across devices.[1]
International shoe retailer Clarks used Wunderkind's identity network to deanonymize 32% of their website traffic, which brought in twelve times more visitors to the retailer's website and 5.5 times more revenue growth.[10]
Programmatic media partner MiQ collaborated with Experian to help their identity graph create a 64% increase in reaching audiences through universal IDs and adding 6.5 devices to each matched IP address.[11]
Applications
[edit]Using identity graphs, businesses are more likely to achieve the following:[1][4][5][12]
- Personalized and improved customer service
- Cross-device attribution
- Personalized in-app experiences
- Deliver effective promotions using context-aware messaging
- Precise and personalized marketing campaigns
- Reach non-logged-in audiences
- Increase customer engagement and revenue
- Marketing attribution
- Audience segmentation by brand
- Early adopter path to purchase insights
- Indentify look-alike customers
A more complete identity graph for customers may help machine learning algorithms to analyze seasonality, cross-category purchases, churn risk, price sensitivity, and in-store predictions.[4]
See also
[edit]References
[edit]- ^ a b c d "What is an Identity Graph? | Everything You Need To Know - Richpanel". www.richpanel.com. Retrieved 2025-01-29.
- ^ "What Is an Identity Graph? [The Plain-English Guide]". blog.hubspot.com. 2022-11-11. Retrieved 2025-03-20.
- ^ Sands, Mike (2016-11-16). "ID graphs: The path to identity resolution". MarTech. Retrieved 2025-01-29.
- ^ a b c d "Retail Identity Graphs: Identity Management Is The Foundation Of Accurate Customer Insights And Predictive AI | Martech Zone". 2024-10-14. Retrieved 2025-01-29.
- ^ a b c Jinturkar, Renu (2022-10-04). "What's an identity graph?". LiveIntent. Retrieved 2025-01-29.
- ^ "Identity Graph: What Is in It for Marketing?". Admixer.Blog. 2021-06-10. Retrieved 2025-01-29.
- ^ a b c "How identity graphs are built — present and future | The Trade Desk". The Trade Desk. Retrieved 2025-01-29.
- ^ "What is an Identity Graph? | Everything You Need To Know - Richpanel". www.richpanel.com. Retrieved 2025-03-20.
- ^ "Building a customer identity graph with Amazon Neptune | AWS Database Blog". aws.amazon.com. 2020-05-12. Retrieved 2025-03-20.
- ^ "Wunderkind's Massive Identity Graph Lifts Revenue 5.5X for Clarks". Adweek. 2023-11-07. Retrieved 2025-01-29.
- ^ Schneider, Hayley (2024-02-27). "MiQ's Identity Spine enhanced by Experian's Graph | Experian". Marketing Forward Blog. Retrieved 2025-01-29.
- ^ "Building a customer identity graph with Amazon Neptune | AWS Database Blog". aws.amazon.com. 2020-05-12. Retrieved 2025-01-29.