The data and analytics market moves so fast, it’s hard to gauge what’s really trending and what’s just a passing fad. Luckily, data and analytics is our business and our experts have identified where you should place your focus in 2023.

What are the key concepts and solutions within data and analytics this year? What new approaches and tools are businesses adopting and why?

Our experts answer those questions to help you navigate the ever-changing data and analytics landscape so that you can transform your business and move at the pace needed to keep up.

Here are the 5 Key Data and Analytics Trends to Think About This Year

Trend #1: Data Products

Treat Your Data Like the Valuable Product It Is

Your data is more than tidbits of information—it has the potential to deliver tangible value for data consumers throughout your organization. It’s time to start treating it accordingly by adopting a data product mindset to meet your overall business strategy.

Think of data products in the way you might think of any other product or service you have that serves your customers today.

By adopting a data product mindset, you shift your focus from simply thinking of data as an isolated entity to considering the needs of the user and packaging the data in a way that will help them be successful. This involves creating a cohesive product that combines the data, along with the tools and resources necessary for the user to effectively utilize it. Examples of packaging your data product to make it more usable includes:

  • Identifying who to contact with questions or issues about the data.
  • Audience-appropriate documentation for how to use it—such as field definitions, example reports, dashboards, or queries.
  • Information about the data itself—such as freshness and where it comes from.

The goal is to improve the data consumer experience for everyone and to reduce complex dependencies and questions of ownership that often exist in legacy solutions. A data product could be used by itself or could be combined with other products to answer increasingly complex questions within the organization.

Brown cardboard box opened with four white graphics inside and orange arrows coming out of the box pointing to the text: A data product includes example reports and dashboards, data freshness and location, field definitions, contact information.

A data product allows you to package data along with the tools and resources necessary for the user to effectively utilize it—ultimately allowing them to make informed decisions and drive business growth.

To adopt a data product mindset, begin by identifying the needs of data consumers for a particular data product you have in mind, and look for commonalities in what they require to use it successfully. This can help you create data products that meet the needs of multiple teams. Once you have many data products, that’s when a data mesh (see below) becomes valuable.

The end-user experience should have always been central to any data offering. Data products ensure that’s the case—making it essential for a data-driven organization.

“A data product mindset is essential to any organization wanting to thrive in a complex data-driven world. Looking at data through the lens of your data consumers is the only way to get there.” – Tony Dahlager, Managing Director

Trend #2: Data Mesh

Clean up your Data Mess with a Data Mesh

Do you feel like your data teams never move fast enough, are always behind the curve of demand, and spending money without measurable impact? Maybe worse—are you noticing that when there are initial signs of progress your teams seem to slow down and can’t keep pace to maintain competitive advantages initially gained?

Don’t lose patience—there may be a better way to handle things!

Data mesh is a decentralized approach that enables teams to create their own data products and share them with others. It emphasizes the interaction between people and technology to improve collaboration and shared ownership of data—leading to faster decision making, increased agility and competitiveness, and continuous improvement and innovation.

To determine if data mesh is a right fit for your organization, assess whether your current approach to data is meeting the needs of your business. If you’ve reached a point of complexity where a purely centralized approach isn’t cutting it—then data mesh may be a viable solution.

To implement a data mesh strategy successfully, you need to:

  • Prioritize first creating data products that will serve consumers’ most pressing needs by empowering domain-oriented teams to create their own data products.
  • Provide a framework that makes it easy for teams to create, find, and share high-quality and consistent data products with others.
  • Ensure you have clear expectations and mechanisms in place to apply the necessary legal, compliance, and security requirements of your data products.
Orange cube with left side stating features before a data mesh implementation and the right side of cube stating benefits after implementation.

A data mesh enables organizations to decentralize their data, empowers cross-functional teams to share data, increases autonomy, accessibility, and connectivity.

Caution: A data mesh strategy is not for everyone, especially if you spread data teams too thin or do not have proper data governance. Without proper strategy, you can find yourself in a bigger mess than where you started.

“Implementing data mesh requires a commitment to treating data as a product and embracing a decentralized, self-service model for data management. While this approach can unlock the full potential of your data and turn it into a powerful asset that drives value for your organization, it’s important to make sure your team is ready for the change before diving in.” – Tony Dahlager, Managing Director


Have a data strategy that isn’t delivering?

Talk to an expert about your data strategy needs

Trend #3: Event-Driven Architecture

The Ultimate Approach to Orchestrate Decoupled Tools

Organizations are using many different cloud-based tools and services in their modern data architectures today. With the more tools you add to an architecture, the more contingencies you create, making the coordination of their jobs more difficult to manage.

The easiest way to manage the integration of your tools is to schedule jobs that run at particular intervals (as many organizations do)—but this can also be the hardest to maintain. Often, the best approach to orchestration is to create workflows driven by events.

An event-driven architecture is like a conductor of an orchestra—it helps all the different instruments work together in harmony. It uses event messages to trigger and communicate tasks between decoupled services, ensuring that everything runs smoothly and efficiently. There are many benefits to an event-driven architecture, including:

  • Data can be processed as soon as it’s available. The lag times of overnight batch loading are gone. This is the foundation of streaming workloads, and it also enables real-time and near real-time analytics. Dependencies between code and data are reduced, making it easier to roll out and maintain without disruption.
  • Costs and functionality of the cloud are optimized because on demand tasks only run when needed. Gone are the days of expenses services running full time just waiting for something to happen.
  • Data observability is increased and more intuitive. When a task fails to meet its intended purpose, an engineer can quickly pinpoint the proceeding event that failed, diagnose which component caused it, and fix it while other events are continuing in parallel.

Event-driven architecture is not an all or nothing pursuit. Before you develop any new workloads, start by asking yourself if it could be an event-driven architecture. If you’re looking to refactor existing workloads, address the ones that are most cumbersome to maintain.

Life is too short to be constantly managing broken data pipelines and workflows. The more you can implement an event-driven approach—the more time you’ll have to spend on finding ways to use the data for the good of your organization while saving money.

Five benefits of event-driven architecture. Blue icons with descriptions below describes the benefits.

An event-driven architecture makes it easier to orchestrate tools and services.

“An event-driven architecture helps your business adapt and react quickly to change. It makes your systems more flexible and easier to scale while reducing vendor lock-in.”
– Patrick Vinton, Chief Technology Officer

Trend #4: Machine Learning Operations (MLOps)

Maximize Your Potential to Scale

Machine learning (ML) isn’t new—especially for organizations with advanced analytical maturity. It has been on the scene for some time, and according to a McKinsey Global Survey, adoption of AI tools continues to grow, and the benefits remain significant. Organizations that invest in ML and AI technologies are expanding their competitive advantage and unlocking additional value across the organization.

The problem, however, lies in the ability to redeploy ML models at scale and to retrain them so that they remain relevant.

Enter machine learning operations (MLOps)—a trend you should pay attention to if you’re ready to treat data science not like a one-off pilot project, but rather an integrated part of your existing data analytics environment. It is the practice of building machine learning pipelines—rather than just a model—to ensure that you can automate and optimize your ML workflows.

Benefits of building a machine learning pipeline includes enabling machine learning operations (MLOps) to scale your workflow. Five circles in shades of blue connected by dots on singular line.

A machine learning pipeline enables MLOps workflows to extract data, train and retrain models, test and validate, serve into production—and most importantly monitor and evaluate progress.

Are you ready for MLOps? If you’re working with a modern data stack, using cloud-native AI tools, and have a need to scale your ML use cases—it makes sense to look at MLOps as part of your data strategy.

Start with qualifying your initial use case and assess for any gaps or barriers in technology, processes, or internal skillsets. Then look for opportunities to accelerate the development and deployment of your machine learning efforts in all areas of your business. With each use case, you’ll improve your model training and fine-tune your processes to gain valuable insights and experience, and ultimately, achieve longer term success with machine learning.

Don’t waste the time and effort you spend on machine learning just to let it turn into a short-lived one-off project. Scale your machine learning efforts and expand your competitive advantage.

“MLOps allows organizations to turn their data into a competitive advantage by automating and scaling their predictive models and processes. By leveraging the power of machine learning and data science, organizations can improve efficiency, accuracy, and speed—and make better, more informed decisions. MLOps is essential for any organization looking to drive innovation and stay ahead in a rapidly changing world.” – Kevin Lobo, Managing Director

Trend #5: Data Literacy

Empower Your Team with Data Skills

You’ve invested time, money, and resources in the latest technologies. You’ve developed a data strategy roadmap, implemented a successful data warehouse even explored new ways to analyze data—but your users still aren’t fully enabled.

What’s the problem?

If your business users are still struggling to understand when to use data and how to interpret it—your organization lacks data literacy.

You’re not alone. According to the Gartner Annual Chief Data Officer Survey, data literacy is the second biggest internal roadblock to success for a data-driven organization—and by 2023, it will become essential in driving business value.

So how do you know if you have a data literacy problem? Look for some common signs, including low user adoption of data and analytics products, glaring inefficiencies across the organization, lack of useful insights or immature use of data, etc. None of these reasons alone means you have a data literacy problem, but if all foundational pieces are in place and your organization is still not achieving business goals, it’s time to dig deeper.

Grey and black MacBook with blue colored bubble graphics with text describing data literacy: when and how to use data.

Data literacy will enable your business users to know what they are seeing, doing, and sharing when it comes to data.

Data literacy cannot be achieved with technology (although there are tools that can help get you on the right track)—it is a practice you need to adopt. But with certain things in place, you can begin to enable your business users with information like:

  • When to insert data and analytics in a business process
  • How to read/interpret charts and visualizations
  • What to do with the insights uncovered
  • How to leverage existing reporting and data analytics initiatives
  • How to use data responsibly and accurately

If you’re ready to invest in better data literacy, start by assessing the specific needs of your business users and evaluate their analytical maturity. This will help to identify areas where additional training and support are necessary. Any plans you create around improving data literacy must be accounted for and integrated into your data strategy.

Remember, the goal of data literacy is to enable and empower your business users to know what they are seeing, doing, and sharing when it comes to data.

“To stay competitive in today’s data-driven world, it is important for organizations to invest in data literacy so that business users can understand and use data effectively. This can lead to better business outcomes, better decision-making, and the ability to use data as a strategic asset for innovation and growth.” – Christina Salmi, Managing Director

Don’t Forget About the Core Principles of Data and Analytics

Before considering how you can incorporate these approaches into your data solutions, ensure you have a solid foundation in place: a current data strategy, a well-equipped data stack, and a well-documented roadmap for reaching your desired future state.

While there will always be something new and it’s easy to get caught up in the hype, don’t forget that there are core principles of data and analytics that will always serve you well. Every data initiative should be approached with usability, speed, security, stability, and scalability in mind. If you follow these core principles, you will be set up to navigate any worthwhile trend with ease while getting the most value out of your data initiatives.

Watch the live-panel discussion—Navigating the Data Landscape in 2023: 5 Key Trends to Watch


Get In Touch With a Data Expert Today

Sharon Rehana Sharon Rehana is the content manager at Analytics8 with experience in creating content across multiple industries. She found a home in data and analytics because that’s where storytelling always begins.
Subscribe to

The Insider

Sign up to receive our monthly newsletter, and get the latest insights, tips, and advice.

Thank You!