Beyond the
Hype
Predictive Analytics and Machine Learning
Understanding the Utilization of
of respondents said they were considering or evaluating these tools
Meanwhile, around
The State of Adoption
Interest on the Rise
Under Consideration
These survey responses highlight some important trends. A very small number of organizations are operating with 1–2% forecast accuracy, and the majority don’t even know what the accuracy of their forecasts are. Lack of access to the necessary data and ineffective software tools are the among the top challenges.
10–20%
6–10%
3–5%
1–2%
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
Impacting Factors
Business Impact
Current Level of Accuracy
Top Use Cases for Predictive Analytics and Machine Learning
When asked about the use of Machine Learning tools, the results were similar—roughly 14% of respondents said they were currently using ML, while 35% said they were considering or planning to use ML.
No, but considering
Not Sure
No, and not considering
Yes
Predictive Analytics and AI/ML Usage Survey
July – August 2019
80% were from companies with over $500M in revenue
from Europe, the Middle East and Africa (EMEA)
of respondents
Represented IT
from North America
of respondents were
Finance Professionals
Survey Respondents
of survey respondents said they were currently using Predictive Analytics
Surprisingly, only
of respondents are still unsure about the value of Predictive Analytics
But, a whopping
And the results are somewhat surprising.
With all the buzz around Artificial Intelligence (AI) and Machine Learning (ML), you’d think that every organization was using these tools or planning for how they are going to use them. But is that really the case? What percentage of companies are actually using these tools? In what functions and for what applications? This was the focus of a recent OneStream Software market survey.
Machine Learning
When asked about the top use cases for Machine Learning, the answers were slightly different from those for Predictive Analytics. The #1 response was Intelligent Process Automation at 41%. This was followed by Sales/Revenue Forecasting (34%), Anomaly Detection (32%) and Demand Planning (32%).
High Impact
Medium Impact
Low Impact
Not Sure
Lack of Management Input
Lack of
Data Access
Ineffective Tools
Industry / Market Volatility
Not Sure
Poor Calculations and Drivers
Click or tap to view each graph
144
69%
76%
24%
18%
$500M+
IMPROVING FORECAST ACCURACY
Predictive
Analytics
16%
40%
21%
Click or tap each tile to view more info
MACHINE LEARNING
TOOL USAGE
Click or tap to view each graph
Organizations have now gotten beyond the “hype” stage and are seeing positive results from practical use cases for Predictive Analytics and Machine Learning. The next few years should be exciting as more are adopting these tools to achieve enterprise-wide success.
Predictive Analytics
When asked about the top use cases, the #1 response was Sales/Revenue Planning at 66% followed by Demand Planning (42%) and Customer Service (23%). These results are not surprising, since Predictive Analytics is all about using historic data to predict future outcomes—and sales/revenue and demand are the two areas that enterprises have the most trouble forecasting.
Predictive Analytics
Machine Learning
0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
Sales / Revenue Forecasting
Demand Planning
Fraud Detection
Dynamic Pricing
Customer Service
Call Center Staffing
Predictive Maintenance
Not Applicable
Other
Sales/Revenue Forecasting
Anomaly Detection
Data Center Optimization
Intelligent Process
Sales/Marketing Optimization
Customer Service
Cyber Security
Collaboration
Not Applicable
Other
Demand Planning
Click or tap to view each graph
Forecast accuracy is clearly a challenge for many organizations, and a large majority say improving forecast accuracy can have a medium to high impact on their business. To be prepared for the use of Predictive Analytics and Machine Learning, organizations should make sure they are selecting software partners, when for ERP or CPM, that are investing in these technologies and making them accessible and applicable to the business processes where that can most benefit customers.
Conclusion
OneStream Software
362 South Street
Rochester, MI 48307-2240
+1.248.650.1490
sales@onestreamsoftware.com
onestreamsoftware.com
DOWNLOAD THE FULL REPORT
Current Level of Accuracy
We asked respondents what their current level of forecast accuracy was for their organization. Not surprisingly, 46% of respondents weren’t sure. For those who did know, the largest grouping of responses was 22% saying 6–10% forecast accuracy. Another 17% claimed 3–5% accuracy, while only 6% claimed 1–2% forecast accuracy.
Business Impact
While 48% of respondents said they were satisfied with the accuracy of their forecasts, 83% said that improving forecast accuracy would have a medium to high impact on their business. Less than 10% believed improving forecast accuracy would have a low impact on their business.
Impacting Factors
When asked what factors were impacting their ability to produce more accurate forecasts, the primary responses included lack of line management input and accountability (39%), lack of access to necessary data (38%) and ineffective software tools (32%).
14%
35%
33%
18%
When asked about the use of Predictive Analytics tools, surprisingly, only 16% of respondents to our survey said they were currently using Predictive Analytics, but 40% said they were considering or evaluating these tools.
PREDICTIVE ANALYTICS
TOOL USAGE
21%
23%
40%
16%
Not Sure
No, and not considering
No, but considering
Yes
Click or tap to view each graph
What are Artificial Intelligence, Machine Learning and
Predictive Analytics?
Predictive Analytics
The practice of extracting information from existing, historical data sets to determine patterns and predict future outcomes and trends. Predictive Analytics forecasts what might happen in the future with an acceptable level of reliability and includes what-if scenarios and risk assessments.
Machine Learning
A branch of Artificial Intelligence that focuses on software that has capabilities based on recent experiences and past trends. Machine Learning leverages statistical algorithms to learn and get smarter over time (e.g., training), retraining itself the more it “experiences.”
Artificial Intelligence
The theory and development of computer systems to be able to perform tasks that typically require human intelligence—such as visual perception, speech recognition, decision-making and translation between languages.
BEYOND THE HYPE
Click or tap to view each graph
Not Sure
No, and not considering
No, but considering
Yes
onestreamsoftware.com
sales@onestreamsoftware.com
+1.248.650.1490
OneStream Software
362 South Street
Rochester, MI 48307-2240
