This site is not optimized for Internet Explorer 9 and lower. Please choose another browser or upgrade your existing browser in order get the best experience of this website.

Beyond Efficiency: The Surprising Wins of Collaborative Knowledge Graphs

June 09, 2022

Ben Nussbaum

Beyond Efficiency: The Surprising Wins of Collaborative Knowledge Graphs
You shouldn’t force your knowledge workers to keep everything they’ve learned about your organization in their heads or on papers scattered around their desks. Keeping their knowledge confined isn’t just ineffective—it sets a dangerous precedent for generating inaccurate insights about your organization, its processes, its customers, and more. Once misinterpretations are created, they travel from data silos, get miscommunicated via the lack of consistent data language, and start gnawing away at your foundation.

When organizations empower their people with a genuinely collaborative knowledge graph, the time-savings benefits are enormous. One financial consultancy, which helps its clients predict and navigate the complex global network of sanctions from the U.S. Treasury, turned a project that would have taken weeks of open-source intelligence gathering into hours of collaborative effort.

But when knowledge workers collaborate on a knowledge graph, the benefits don’t stop with getting work done faster.

Easier insights for non-technical employees

With a knowledge graph, data is represented simply, as people, places, things, and the relationships between each. Not ten competing lexicons from each of your business units or complex database schemas that knowledge workers can only understand with a few years’ worths of practical experience in writing SQL queries.

Investigation into unique insights becomes more accessible with simple data and underlying structures, opening it up to knowledge workers across your organization—especially those who might have shied away from technical work before.

And a suite of tools for controlling, managing, and collaborating across a knowledge graph, like GraphGrid CDP, comes with an intuitive search for full-text articles and unstructured data. That alone helps unlock unexpected truths across a sprawling corpus of online articles, social media interactions, customer interactions, or other forms your data might take. Graph data is also easily visualized to show the type and strength of relationships between nodes, unlocking new investigatory avenues for your entire organization.

Smoother and faster developer experience

Every organization faces a degree of lag (at best) and whiplash (at worst) in the repetitive cycle between identifying the need for new data, reports, or visualizations and fighting with technical staff over the bandwidth required to finalize the implementation.

And one of the main ways these cycles spiral into their worst qualities is inconsistency in the language used around data. By helping implement and validate a common language, knowledge graphs are “whiteboard-friendly,” allowing discrete teams to coordinate in the same room or over Zoom calls to discuss needs and scope development projects collaboratively.

Platforms that help organizations build and maintain collaborative knowledge graphs often come with integrations for event-driven architectures. The knowledge graph can react and respond to events, ingesting data from any number of sources, transforming it in real-time using natural language processing (NLP), or trigger dynamic indexing and search processes. Developers already put hard work into their automation and can now extend that into knowledge work.

Developers can focus on higher-value API implementations with simplified fundamentals and self-serve processes. In GraphGrid CDP, these are called Showmes—dynamic APIs that return targeted results of graph data. Showmes are queries, written in the Geequel graph query language, that can be tuned and optimized to perform specific tasks that developers know their knowledge worker peers will need regularly.

Once developers have deployed a Showme to GraphGrip CDP, all updates to its results are incorporated and displayed in real-time.

Greater durability of your data, relationships, and foundational knowledge

Every popular, modern database comes with a degree of data durability—they can roll back failed transactions to maintain a consistent data state.

But because knowledge graphs help people derive insights from relationships, it’s important to treat data with an extra layer of protection. For example, the native graph database in GraphGrid CDP is designed to guarantee the data elements on either side always agree upon relationships. When it writes to the relationship between a start and end node, also known as the edge, it locks both nodes for guaranteed durability.

By preserving the relationship, the graph database also preserves your knowledge workers’ ability to do their work without the threat of misinterpretation. When seeking insights from data, the last thing any organization needs is erroneously-edited relationships being disseminated across the knowledge graph, into applications and reports, and ultimately to incorrect decisions from stakeholders.

You can also extend durability with highly-available clustering with ONgDB, which GraphGrid CDP uses for its internal storage framework, to improve reliability, availability, and speed at the same time.

These traits of native graph databases give you certainty. The certainty that your underlying data is immutable, and the certainty that all the knowledge your people input and analyze today is based on a dependable source of truth. And over the years after deploying your first collaborative knowledge graph, certainty that your knowledge workers will be able to discover new relationships and insights on a truly accurate and complete source of organizational truth.

And that’s why GraphGrid CDP uses a native graph database at its core—a level of data integrity that gives you immutable certainty in your knowledge graph.

Accelerated return on data

Every organization must weigh the value of how much they spend on collecting, storing, maintaining, and evaluating its data. The calculus can be particularly painful when you consider that your organization is likely paying a large monthly bill to store its data but not getting any value from 90% of it, equating to a low return on data (ROD).

Knowledge workers can swing these numbers in your organization’s favor, but only when they have connected data and collaborative tools at their fingertips.

After implementing a knowledge graph that interconnects and automates all your data silos, these employees suddenly have access to swaths of previously inaccessible information without wasting time writing SQL queries whenever they want to explore or investigate. They can apply textual search, which indexes the connected data from their knowledge graph, to return the most relevant nodes and relationships in context.

But your data isn’t also isn’t a static asset—you’re constantly ingesting new data from existing sources and integrating new tools or applications. To prevent wasted time in data entry or manually connecting new data to the existing graph, natural language processing (NLP) automatically ingests new data from the sources you enable and surfaces potentially valuable relationships across your entire connected graph. Knowledge workers focus on the most useful results and relationships rather than fretting over complex queries.

Wherever you want to automate what were once manual reporting processes, you can implement Showmes, freeing knowledge workers to explore into that foggy 90% of data that you’ve been collecting and paying for but have been getting no value from.

All these tools and techniques improve your organization’s “time to knowledge,” the speed at which knowledge workers can create and share meaningful insights after beginning a new initiative.


To get started layering in these wins of using a collaborative knowledge graph, you can download GraphGrid CDP entirely for free. These freemium editions come with Manager, Search, and graph visualization features, plus the security and configuration capabilities your development teams need to deploy with both durability and development speed in mind.

The GraphGrid team has been developing enriched change data capture (CDC) tooling within GraphGrid CDP, expanding how you can automatically extract more value from the data you already have in your knowledge graph based on new changes. Stay tuned for a new series on the what, how, and why you should bring CDC to your knowledge graph in a future series!