Most business challenges facing the enterprise today are in-fact data challenges.
Why do I say that?
Because data is at the core of all modern businesses. In today's world, organizations generate and collect vast amounts of data from various sources, including customers, suppliers, employees, and operational systems. This data can be used to gain deeper understanding of aspects of the business, such as customer behavior, market trends, and operational efficiency.
But we all know, it is increasingly complex, painstaking, and slow to manage and understand the data that’s created and stored within enterprise applications and systems. .
Take Caterpillar, for example – the world’s largest manufacturer of heavy equipment and machinery .
Caterpillar realized they were sitting on a gold mine of data, but it was hidden across a mountain of disparate maintenance and repair documents.
.They needed to find a way to unlock valuable insights hidden within this vast repository of technical documents – and understand patterns and relationships – in order to make equipment repair and maintenance more efficient and improve their supply chain management processes.
This is not a challenge that is unique to Caterpillar. At the heart of every enterprise challenge today is an explosion of data complexity.
And this is across ALL industries.
With the increasing amount of digitalization in the economy, all companies are data companies – generating, collecting, and processing vast amounts of data, often from a variety of sources, such as SaaS apps, social media, IoT devices, and online transactions.
Whether it’s Hästens pursuing a full 360-degree view of it’s customers while improving sales operations – managing 275 points of sale and over 200,000 different combinations of components and colors.
Or Price-waterhouse-Coopers tackling the growing threat of digital fraud and money-laundering disguised in complex networks of financial transactions.
Or Adobe managing and growing the world's largest creative network for showcasing and discovering creative work.
These are all data problems!
It’s not just that data volumes are growing exponentially – data is increasingly complex and connected..
The first number on the slide is almost unfathomable. It’s predicted we’ll have over 200 zettabytes in cloud storage by 2025.
So, 1 Zettabyte is a billion terabytes. Not helpful? Well, I asked ChatGPT and this is what it told me: “To put it into perspective, it would take approximately 250 billion DVDs to store a zetabyte of data, or it would take over 4.4 million years to watch all the videos contained within a zetabyte.”
But the second number is actually much more interesting..According to Okta’s 2023 Business @ Work report, the average European company now have 65 SaaS apps deployed. This is across marketing, sales, finance, HR, customer service, etc. – All producing valuable business data to guide business decisions, improve operations, and delivering better customer experiences.
Oh, and in case you’re curious, the average volume of apps in North America is 98!
But it’s not just people and apps that are creating data. Everyone is connected to everything. In 2025, we’re projected to have 41 billion – yes, 41 BILLION – IOT connected devices.
Organizations are drowning in data.
But we frequently hear from leaders that all that data is not delivering on the promise of deeper understanding of insights.
Salesforce just published a 2023 research study of more than 10,000 organizations that underscores this very point.
In the report, 80% of business leaders say that data is critical in decision-making.
Yet, 41% say that it’s challenging to understand the data they’re generating and collecting because it’s simply too complex.
We covered this earlier – organizations often have multiple systems and applications, which have a tendency to create data silos that are not integrated with one another.
This leads to data duplication, inconsistencies, and gaps that make it difficult to gain a complete view of the business.
On top of that, true business value is often only realized when data from multiple sources are integrated to provide a broader, fuller view.
The growth of data that is inherently connected continues to rise. And increasingly, business value is hidden in the structures and relationships of those data connections.
It’s the connections and the insights derived from seeing the points of intersection that have the most power. It is these relationships that unlock opportunities and open the doors to innovation and competitive advantages you didn’t know existed.
The reality for enterprises today: It’s increasingly complex, painstaking, slow and expensive to extract actionable insights from these interconnected data sets.
Business value is hidden not just in data points, but more so in the relationships and patterns across billions of data connections.
Traditional approaches to data management can't solve this problem for us..Relational databases struggle with the complexity and scale of these interconnected data sets.
It’s not unsurprising when you think about it –the relational database was originally architected for storage efficiency, not surfacing data patterns and relationships
Now, these systems are breaking down because the relationships in the data are becoming too complex.
If we look at relational databases, they aren’t well adapted to interconnected and semi-structured data
Connections in data contain context and meaning. But traditional DBs strip out connection
So, relationship queries require painful SQL and lots of lookups and joins
The schema rigidity makes it difficult to update the data model and limits quick expansion to new uses
Other NoSQL databases are built to scale simple data (store & retrieve)
And lack of guaranteed data validity (ACID) is a major issue with business-critical applications
A fundamentally new approach is needed.
To solve this problem, you must approach the technology in a fundamentally different way. Doing more of the same Is the wrong answer.
You've got to expand your thinking about data. It’s not just about the data points themselves, but about the connections between them.
[5 secs - snap!]
The solution is graph!
Graph databases are designed to manage and uncover the hidden patterns and connections common in today’s highly connected data. They think in relationships, just like the human brain. And they’re built for today’s fast-moving world in which datasets are in constant motion.
The fundamental idea was sketched on a napkin by Neo4j’s co-founder and CEO, Emil Eifrem.
The original napkin has been lost to time, but we made a copy from memory.
Let’s look closer at that napkin. It depicts what we today know as the property graph model.
The property graph model is very simple. It builds on only three concepts: you have nodes, you have typed relationships between them and you have key-value pairs that we call properties on both nodes and relationships.
The true power of the property graph model is that it’s the natural, human way approach data.
The Neo4j graph model allows you to model data in a way that everyone can understand. This ultimately makes the data more useful because it is easily understood, without any cognitive dissonance.
With the graph model, there is no need to translate data into a traditional data model with rows and columns, which can be difficult for the business to extract understanding from. Instead, the graph model is intuitive and easier to digest, making it easier to communicate across the business.
The graph model also allows for exploring data and understanding the context around it, which helps to expand curiosity and identify new insights. The graph model maps the real world as is, just like what you would draw on a whiteboard, and is completely flexible, allowing you to change and adapt as your needs change.
Overall, the graph model makes it easier to understand and use data, leading to more insights and better decision-making.
Graph technology’s talent for handling semi-structured and unstructured changing data makes it ideal for handling the complexities and challenges of Enterprises applications.
We believe this is why graph has been the fastest growing database category – by far – for the past 10 years.
The explosion in data complexity is driving adoption adoption of graph databases – the only system purpose-built to store and navigate relationships in connected data.
HAMMER HOME THIS SLIDE Graph enables organizations to quickly and easily uncover hidden relationships and patterns across billions of data connections
Powering applications that are impossible with other technologies
The limitations of traditional relational databases become apparent when it comes to exploring multiple levels of connections in data.
As the number of connections increases, graph continues to provide millisecond performance while relational databases struggle and collapse under the weight of data complexity.
And data challenges today are increasingly complex with many layers of connections. For example, most supply chains today have dozens and dozens of levels. Managing your supply chain in real-time is unfeasible with an app backed by a relational database. Or consider fraud rings where you need to explore multiple levels of connections to expose individuals with certain shared characteristics. Modern fraud detection is simply impossible with legacy relational systems.
We started talking about Caterpillar and them wanting to extract business value from their maintenance and repair documents.
To uncover hidden relationships, Caterpillar created a natural language processing (NLP) tool bulit on Neo4j that allows them them to find patterns, build hierarchies, and add ontologies. With the help of the solution, users within Caterpillar can extract deeper understanding from millions of documents, enabling efficient equipment repair and maintenance.
So what about Hästens? By using Neo4j, Hästens was able to automate and optimize the management of requests for its product catalogue, which dramatically reduced delivery time. Additionally, the company used deeper insights from Neo4j to optimize marketing to deliver more targeted campaigns in specific areas, driving significant revenue growth
PwC’s Financial Crime Practice now have a next-generation fraud and money laundering detection system backed by Neo4j, enabling the company to deliver rapid solutions for their clients.
Finally, Adobe Behance is the world’s largest professional network for creatives. Originally architected on MongoDB, the system ballooned to over 150 instances with a 50TB footprint. With Neo4j, Adobe could reduce to three Neo4j Graph Database instances with just a 40GB footprint. In addition to significant performance improvement, Adobe saves millions of dollars in AWS fees.
Don’t just take our word for it.
Gartner – and I think we can all agree that they’re a fairly conservative organization. Gartner, they predict that within the next two years, graph will be used in 80% of data and analytics innovations.
Because of our early entry into what is now the graph market, as well as our technology maturity, we've been engaged with leading companies around the world who are all aggressively adopting graph to build better data systems to help them serve the needs of their businesses and customers.
Today, Neo4j graph-powered applications are used by over 75% of the Fortune 500, including some of the world's largest pharmaceutical companies, aircraft companies, auto manufacturers, and banks.
As you can imagine all of these sectors have significant networks of people, places, goods and services that graph is helping them understand better.
First, we see how Neo4j can support both transactional and analytical workloads - this is non-trivial from an engineering perspective! We have overcome the typical trade-offs associated with having to design or tune a database to support one vs. the other, allowing you incredible flexibility to run analytics on your freshest data, and support real time decisions.
Next, we talked about graph processing in memory, highly compressed - essentially using graph projections for GDS, which gives you blazing fast speeds for your graph algorithms
To support all of these capabilities, we have what I believe to be the richest, deepest and broadest set of tools and integrations in the industry, giving you so many ways to plug into your technology stacks and operational processes
And we have new security controls like SSO VPC peering along with fine grained RBAC
Finally, we are bringing all of this to you where you need us - on-premises, on AWS, GCP and soon Azure, with a truly multi-cloud approach
Neo4j pioneered graph technology. And we continue to be the leading provider of graph-powered solutions to the world.