Wednesday, January 10, 2018

5 Most Influential Data Analytics Trends for 2018

2018 Data Analytics Trends

Data Analytics Trends 2018
Happy New Year!  As we enter the year, the question always comes up, “what are the data analytics trends that we should pay attention to in 2018?”  The technology trends we see are emerging to support a couple of large-scale business analytics trends.

First, there is a slow but steady movement away from process-driven towards data-driven decision-making.  Future thinking organizations understand that 1+1 does not always equal 2 if other factors are considered so following a linear decision-making process may not always result in the desired outcome.  Data provides a clear picture of internal and external influences that may impact outcomes, positively or negatively.  Getting up-to-date information in the hands of the right people up and down the organization, in a context they can use, is becoming critical to ensure future competitiveness. 

Second, there is widespread realization that data is an asset and a valuable one at that.  While analysts have been talking about data monetization for a while, most organizations are still trying to get their data to work for them internally.  In 2018 many will continue to focus their analytics thinking inwards, but those who have figured it out are starting to look outward.  They are changing how they think about analytics to include how their customers and partners might also benefit. It may be a while before we see a proliferation of data marketplaces, but organizations are starting to think about how they can use their data to create new data-driven products and services and open up new revenue streams.

With these high-level business drivers in mind, here is our take on the 5 Most Influential Data Analytics Trends for 2018. 

An analytics architecture rethink is underway.

Data Architecture Team
For most companies, single stacks data architectures just don’t fit anymore.  Modern data stacks optimize for different types of data and use case, so most enterprises have a mix of RDBMS, SQL, NoSQL and Cloud APIs.  Trying to funnel all the right data into a data lake, data warehouse or other reporting data store is increasingly more difficult and time-consuming, as business and data move faster than ever before.  To keep up with business demands for agility and rapid experimentation data analytics architectures will need to adapt and modernize to cope with an evolving future.

In 2018, this means analytics platforms that venture into this modern data environment will stop being simply data visualization tools.  The next wave of BI solutions will take on more data management (data plumbing) capabilities to eliminate the need for ETL and data prep tools and simplify the data analytics stack.  They will natively interact with structured, unstructured and semi-structured data sources, be highly extensible, embeddable, and include published APIs to insert into operational workflows.  These capabilities will allow for the merging of analytics and applications into data-driven applications and transform BI tools from monolithic single destination applications to analytics distribution frameworks that fuel business transformation.

Data science, data discovery and data engineering converge to support rapid experimentation.

Data science, data discovery and data engineering convergence
Is 2018 the year Big Data just becomes data?  I don’t know, but it seems that as NoSQL technologies become mainstream (note: MongoDB IPO), it makes less and less sense to distinguish Big Data from any other kind of enterprise data.  Business teams want to explore large data sets, add context from different sources and build advanced analytics.  For the most part, Big Data and AI are segmented off and use their own set of tools and resources which limits the ability of an enterprise to do interesting things beyond a few use case level experiments.

2018 will be the year that CTOs and CDOs realize that to transform into a data-driven enterprise they need a unified view of all their data as well as a data architecture and culture that encourages experimentation.  New technologies like AI need to be integrated so the work done by data scientists and data engineers can be shared with a broader audience and leveraged enterprise-wide.  This is more than just a shift away from old data architectures but requires rethinking how business teams, data engineers, and data scientist work together in an iterative, agile, development process that is better suited for rapid experimentation.

Data-driven applications and services create new revenue streams.

Data-driven application drive revenue growth
Forward-thinking companies understand that analytics can change experiences for the better.  By providing analytics as part of product or service offering, customers stay longer and do more, making these new data-driven applications more valuable to the consumer.  This value converts to a willingness to pay more, higher retention rates, and long-term strategic relationships which are the stuff of every product managers dreams.

In 2018, product managers will ride the wave of experimentation focused analytics architectures on driving top-line growth.  Product managers will need the ability to experiment with all available data, easily and securely embed analytics into applications and iterate quickly in a self-service environment. As an example, one of our customers, Boku, is monetizing their analytics by providing financial reporting to their merchant and carrier partners.  You can read more here.

Public cloud, private cloud, on-premise deployment?  Yes, please.

analytics architectures must support hybrid environments
Cloud-first strategies continue to be the favorite option for analytics, including Big Data analytics, because of the reduced onboarding friction and greater flexibility.  However, enterprise data stacks are not necessarily on the same path.  With regulatory, security, variable costs, and performance concerns, many enterprises have opted out of the cloud for some applications.  Additionally, NoSQL technologies are optimized to store certain types of data and serve specific use cases.  For example, enterprises use MySQL to store customer information, MongoDB to store customer interactions from their website, and Elasticsearch is used to enable customers to search large data sets very quickly.  Reporting platforms need to pull data from potentially all these sources to add context and answer even basic questions around how engaged a customer is with a product or service.  There is little indication that enterprises will standardize on a single stack or environment, in fact, the opposite is true.

In 2018, there will be a push to modernize analytics architectures to handle increasingly fragmented data and applications architectures without requiring data to be moved and transformed as this limits the ability to be agile and experiment.  Analytics platforms that play nicely in the cloud API’s and can navigate any variety of hybrid environments will have the advantage.

Reporting moves beyond single destination dashboards.

future analytics move beyond dashboardsMass adoption of analytics is often a barrier to transitioning into a data-driven enterprise.  The reasons are two-fold.  First, analytics platforms typically don’t serve different users with varying skills sets well.  Second, getting up-to-the-minute information in the hands of the right people is pretty hard.  These gaps leave groups of people either reliant on other team members for reporting while others just give up. 

In 2018, the push toward data-driven enterprises continues, and analytics platforms will start to move the boundaries of where reporting happens to challenge long-standing barriers to enterprise-wide adoption.  Logging into an analytics platform and looking at a set of dashboards isn’t going away anytime soon.  However, what you will see in 2018 is a broader use of highly contextualized analytics pushed to where the user lives.  This includes targeted analytics built into data applications and starting to leverage AI to make analytics more interactive.  While talking to your analytics platform or type a question into Slack and having a dashboard appear sounds a bit gimmicky today, did you think five years ago it would be normal to ask a device sitting in your kitchen to turn on the lights?  We are still in the early stages of this transition away from desktop dashboards to analytics everywhere, but 2018 will see a significant step forward on the journey.


Post a Comment

Please share your thoughts. If you have a question, please reference our docs section at