Over the past year, there have been a number of trends that have impacted digital transformation, including artificial intelligence (AI), big data, and robotic process automation (RPA), among many others.
At the beginning of 2018, G2 Crowd released a series of posts around these important digital trends and how they would transform business in the year ahead.
One of those trends was centered around AI and artificial intelligence software, but specifically discussed the solution of machine learning as a service (MLaaS). Similar to software as a service (SaaS) or infrastructure as a service(IaaS), MLaaS is when software vendors supply prebuilt machine learning technology solutions to other businesses to embed within their business or applications.
MLaaS solutions are sold on a subscription or usage basis and benefit customers because they don’t need to build their own machine learning models. Instead, they can benefit from prebuilt AI solutions
1. Machine Learning as a Service Will Become Important to Wall Street
These cloud providers are extremely competitive with their revenues and market share, so with each earnings report they like to boast about their quarterly or yearly sales numbers and how they are growing their cloud businesses. While cloud computing has become much more of a common business practice over the past decade, it is still a space that is growing very rapidly.
We believe that these same cloud computing providers will begin to tout their MLaaS sales the same way that they promote their cloud growth.
By Q4 2019, the traditional leaders in the highly competitive cloud computing market will begin to disclose the amount of revenue they are generating from their machine learning as a service (MLaaS) products. Because of the benefits MLaaS solutions can provide, its adoption will become a leading factor in businesses deciding which cloud computing provider to choose when starting a business or migrating to the cloud.
In the coming years, nearly all IT leaders will be tasked with implementing some type of machine learning to improve a huge variety of business functions. Example of these business critical technologies include natural language processing (NLP), natural language generation (NLG), and computer vision, among others. By implementing these AI solutions, businesses will be able to automate tasks that do not require human intervention and augment other tasks to assist employees with their day-to-day assignments.
As these machine learning technologies become ingrained into business processes, the need for easy-to-implement and easy-to-use MLaaS will only increase, which is why it will be a crucial aspect of selecting which cloud computing provider to use.
Additionally, software companies looking to build intelligent applications will want to choose the best AI platform available to build their product, whether that be with Google Machine Learning Engine, Amazon SageMaker, Microsoft Azure Machine learning Studio, or IBM Watson Studio.
These solutions do not work across vendors. It is not possible to select Amazon SageMaker and Google Compute Engine, so depending on which platform the business chooses to build their app on will ultimately impact the other various cloud computing components they need to run their business. That means revenue growth, which leads to market share growth, which leads to stock share growth, the ultimate end goal of each of these enterprises. That’s why they will want to publicize their revenue numbers surrounding MLaaS.
2. Robotic Process Automation Will Help Small Businesses Excel
Robotic process automation (RPA) software has become a very trendy solution over the course of 2018, and for good reason. The ability to automate tedious, time-consuming tasks, or assist employees with the use of supervised automation, is a massive benefit to businesses. Companies can save employees time and ensure that processes are conducted properly.
RPA is often placed in the bucket of AI and machine learning, and growth in that market has been booming for some time now. However, when segmented out of that very general bucket, RPA is projected to have some amazing growth as a standalone market. According to a Forrester report, the RPA market was worth $250 million back in 2016, but is estimated to grow to $2.1 billion by 2021.
On G2 Crowd alone, traffic to the RPA category page has grown substantially over the course of 2018. From the time the category was added in February 2018 through the end of October 2018, organic traffic to the RPA category page has grown almost ten times over and is now the tenth-highest category in terms of organic traffic on all of G2 Crowd, which means that buyers are very actively looking for automation solutions.
Traditionally, the companies leveraging these digital workforce solutions are enterprise-sized businesses. These companies have not only the resources to purchase and implement the software, but the need to automate a large number of processes. However, moving forward into 2019, small businesses will begin to implement these tools at a much higher rate.
Robotic Process Automation (RPA) vendors will begin to target small businesses, not just enterprise companies, and those smaller companies will leverage the advantages of a digital workforce. In turn, small businesses will become at least 30% of the entire RPA market.
As of November 27, 2018, only 16% of G2 Crowd reviews in the RPA category have come from users at companies of 50 employees or fewer. We believe that number nearly doubles over the course of 2019, to 30% or more.
Small businesses can benefit from employing a digital workforce to perform tasks that are necessary for rapid scaling, as opposed to hiring human labor to conduct those same tasks. This can save them a good deal of money in the long run. Additionally, RPA vendors will begin to embrace this segment as prospects and strive to sell to smaller, growing businesses.
Enterprises have long been implementing automation solutions and will continue to do so, but as the ROI becomes faster and more apparent, small businesses will see the value in deploying RPA software.
3. Machine Learning Data Catalogs (MLDCs) Will Become a Necessity
Big data is a critical aspect of digital transformation, as is leveraging data to improve business decision-making. Because companies have put such an emphasis on becoming data-driven organizations, they have empowered their employees to access huge data sets. Often, businesses do this through self-service business intelligence - orself-service BI, applications.
Self-service BI tools are great for those looking to explore and analyze neatly organized data that has previously been cleaned and prepared by a data scientist, data analyst, or IT team. If a data set that the end user is searching for is not already connected to the BI tool, then they will not be able to find it.
If the everyday employee can’t find the data they need, they fall short and ultimately abandon leveraging data to make their decisions. The company probably has the data somewhere in a master data management software tool, or maybe it lives in its raw state inside a data preparation software. Neither of these tools are the simplest solutions for everyday users to maneuver.
Now, if an employee is savvy enough to find the data set they are looking for within these tools, they may have stumbled upon much more than what they were intending. Often, these tools house all of a business’ data, including sensitive data that they might not want to share with all employees.
To better handle the growth of big data, businesses will rapidly increase implementations of machine learning data catalogs to better organize, govern, and allow self-service access of data to end users. The implementation of machine learning data catalogs will increase at least 50% year over year.
Machine learning data catalogs help businesses organize all of their data in an intuitive manner that allows for easy access to end users and non-technical employees. Additionally, it provides security features, like dynamic data masking, to help prevent users from finding potentially sensitive data.
Related: Learn how AI and embedded intelligence work together to advance technologies by integrating software with small devices in order to improve daily lives.
These aspects of data management are becoming necessities to businesses that are embracing the data-driven business world of 2018, and will only continue to do so even more in 2019.
Rob is a research principal focused on enterprise technology vendors and their continuous battle for market share in the age of digital transformation. Rob's work digs into competitive trends for enterprise giants, such as Amazon, Microsoft, Oracle, and IBM, among others. In addition, he highlights acquisitions, innovative product releases, and unique differentiators between enterprise vendors. He has been with G2 since 2015, and has shaped the direction of G2’s report and research offerings. While the enterprise is professional passion, in his free time Rob enjoys watching as many films as possible and even dabbles in some amateur screenwriting. His coverage areas include enterprise technology and strategy.