Powering the future of influencer marketing with SurrealDB context layers

Later built a context-aware knowledge graph on SurrealDB, unifying semantic search, graph traversal, and vector search in one engine to discover and activate creators faster.

IndustrySocial media & influencer marketing

Use caseContext-aware knowledge graph

ProductLater EdgeAI

01 |OVERVIEW

Reinventing how brands and creators connect

Later is a leading social media and influencer marketing platform that helps brands discover creators, launch campaigns, and measure performance at scale. As influencer marketing becomes increasingly strategic and data-driven, Later operates at the centre of a fast-moving ecosystem shaped by creators, audiences, cultural trends, and constantly changing context.

With millions of creators, campaigns, and social signals flowing through its platform, Later is investing heavily in AI-powered systems to reinvent how brands and creators connect. This is part of EdgeAI, a broader initiative that brings together inference, persistence, and agentic feedback loops to fundamentally change how campaigns are planned and executed. At the core of this evolution is a shift toward knowledge graphs with semantic entry points as the foundation for smarter decision-making.

To support this transformation, Later partnered with SurrealDB to build a context-aware knowledge graph that enables faster discovery, deeper understanding, and more accurate AI-driven campaign workflows.

02 |THE CHALLENGE

When static filters break down in a dynamic world

Influencer marketing is a dynamic problem. Every campaign introduces a radically different set of variables, from brand safety requirements and audience fit to content style, platform dynamics, and emerging social trends. Matching the right creators to the right campaigns requires an elegant way to represent and persist an understanding of how these factors relate to one another.

Before adopting SurrealDB, creator discovery at Later relied heavily on manual processes and rigid filters such as follower counts, categories, or platform-specific metrics. Defining the right creator set for a campaign was time-consuming and often took hours of effort, while still failing to capture the nuance and intent behind each brand's goals.

As Later expanded its AI ambitions, these limitations became more pronounced. Creator data is deeply interconnected and semi-structured. Social listening signals and historical performance data do not fit neatly into tables. Traditional databases made it difficult to unify graph relationships, semantic search, and predictive modeling into a single system. The result was growing complexity and fragmented infrastructure at the exact moment Later needed speed, accuracy, and scale.

03 |THE SOLUTION

Turning relationships and meaning into a competitive advantage

Later chose SurrealDB because it offers a single, multi-model database capable of combining graph relationships, document flexibility, vector search, and transactions in one cohesive system to support knowledge graphs. Often this kind of system is built by gluing together many specialised tools to get the same functionality.

Using SurrealDB, Later built a knowledge graph that connects creators, brands, campaigns, content topics, social trends, and performance signals. These connections are not static. Relationships carry context and metadata that differ between campaigns and are updated when new content is created. This enables a more accurate representation of the dynamics in influencer marketing.

This project enables a major shift in how creators are discovered and activated. Rather than relying on filtering by common metrics, Later now offers a performant AI-powered search across creator and brand information, which dynamically employs semantic similarity, conventional filters and graph traversal. Taken together, this mimics the kind of search a domain expert would perform. Teams can now get proactive suggestions, as well as explore the ecosystem using natural language chats that retrieve results which reflect intent, relevance, and logical next steps.

Kyle Chamberlain

EdgeAI Architect, Later

SurrealDB's ability to run vector similarity search and graph traversal in a single query has been a key differentiator. Later can combine semantic relevance with multi-hop relationships, brand safety constraints, and historical performance data without complex pipelines or fragile integrations. The simplicity and expressiveness of SurrealDB's query language make it possible to build powerful retrieval and reasoning workflows without introducing unnecessary complexity.

The new DISKANN index type in v3.1.0 has been critical to the success of this project. The strategy we use to represent creator content results in a very high volume of embeddings to store. This type of index makes it uniquely possible to keep embedding dimensions high and still find closest neighbours quickly as graph investigation entry points.

Kyle Chamberlain

EdgeAI Architect, Later

SurrealDB now serves as a core part of Later's EdgeAI persistence layer, enabling recursive knowledge loops where campaign insights continuously improve through feedback and learning. The graph is fed by an event-driven ingestion pipeline Later runs on Encore: as documents are uploaded, parsed, and turned into nodes and relationships, each step emits events that write back into SurrealDB. Together they form Later's AI stack, with Encore orchestrating the flow of work and SurrealDB persisting the meaning.

04 |AT SCALE

From manual execution to intelligent scale

By adopting SurrealDB's context graph architecture, Later has dramatically reduced the time required to plan and execute influencer campaigns. What once took days can now be accomplished in hours, and workflows that previously required hours of manual filtering can now be completed in minutes using semantic search and AI-assisted discovery.

SurrealDB gives us a foundation where we can unify semantic search, knowledge graphs, and AI-driven decision making without stitching together multiple systems. Collapsing responsibility into SurrealDB has become our default engineering posture.

Justin Foley

VP of Engineering, Later

This shift has transformed the customer experience. Brands can move from campaign ideas to activation much faster, discovering creators through intent-based search. Later's services teams save significant time, allowing them to run more campaigns without increasing headcount. Customers are able to launch more creator activations per quarter, improving performance through rapid iteration and experimentation.

The impact compounds over time. Running more campaigns brings in new data from the market, which is used to improve matching accuracy, which leads to better results and stronger customer confidence. As a result, Later is seeing higher win rates on larger deals, improved margins, and increased revenue driven by AI-powered campaign intelligence.

05 |WHAT'S NEXT

One platform for the next generation of Influence AI

Later views its knowledge graph initiative as the foundation of its long-term AI strategy. While parts of the platform still rely on Postgres and MySQL today, the roadmap is focused on consolidating more of the creator ecosystem, campaign intelligence, social trend data, and documents directly into SurrealDB.

By migrating from legacy relational databases and unifying graph, vector, and transactional workloads in one place, Later aims to build a true knowledge centre for influencer marketing. This foundation will power the next generation of Influence AI, enabling deeper context awareness, more capable AI agents, and a fundamentally new way to operate at scale.

With SurrealDB, Later is redefining how brands and creators connect through AI-native marketing intelligence.

MORE CASE STUDIES

See what other teams have shipped

From knowledge graphs to AI assistants - how enterprise teams are building on the context layer.

Samsung

Unlocking insights with knowledge graphs

Samsung Ads uses SurrealDB to build dynamic, real-time knowledge graphs for smarter campaign execution - collapsing three legacy data stores into one.

Read case study

Verizon

AI assistant empowering 10,000 technicians

Verizon uses SurrealDB to power a generative AI assistant for 10,000 field technicians, delivering instant access to documentation, outage updates, and workflows.

Read case study

Tencent