IoT intelligence: Attitudes towards big data and advanced analytics

October 31, 2017
Dresner Advisory Services, LLC
2017 Edition
IoT Intelligence®
Wisdom of Crowds®
Series
Licensed to Information Builders

2017 IoT Intelligence® Market Study
http://www.dresneradvisory.com Copyright 2017 – Dresner Advisory Services, LLC
2
Disclaimer
This report should be used for informational purposes only. Vendor and product selections should be made
based on multiple information sources, face-to-face meetings, customer reference checking, product
demonstrations, and proof-of-concept applications.
The information contained in all Wisdom of Crowds
®
Market Study Reports reflects the opinions expressed in
the online responses of individuals who chose to respond to our online questionnaire and does not represent
a scientific sampling of any kind. Dresner Advisory Services, LLC shall not be liable for the content of
reports, study results, or for any damages incurred or alleged to be incurred by any of the companies
included in the reports as a result of the content.
Reproduction and distribution of this publication in any form without prior written permission is forbidden.

3
Business Intelligence: A Definition
Business intelligence (BI) is ―knowledge gained through the access and analysis of
business information.
Business intelligence tools and technologies include query and reporting, OLAP (online
analytical processing), data mining and advanced analytics, end-user tools for ad hoc
query and analysis, and dashboards for performance monitoring.‖
Howard Dresner, The Performance Management Revolution: Business Results Through
Insight and Action (John Wiley & Sons, 2007)
Internet of Things (IoT) Definition
The network of physical objects or "things" embedded with electronics, software,
sensors, and connectivity to enable objects to collect and exchange data.
Source: Wikipedia

4
Introduction
This year we celebrate the tenth anniversary of Dresner Advisory Services and our first-
ever conference, Real Business Intelligence, which took place on July 11-12 on the MIT
campus in Cambridge, Massachusetts!
Our thanks to all of you for your continued support and ongoing encouragement!
Since our founding in 2007, we have worked hard to set the ―bar‖ high—challenging
ourselves to innovate and lead the market—offering ever greater value with each
successive year.
Our first market report in 2010 set the stage for where we are today. Since that time, we
expanded our agenda and added new research topics every year. For 2017, we plan to
release 15 major reports, including our original BI flagship report—in its eighth year of
publication.
In previous years, we added new topics to our agenda, and 2017 is no exception.
Earlier this year, we published our inaugural Analytical Data Infrastructure report and
added data catalog to the existing lineup during Q2.
This publication marks our third IoT Intelligence®
(formerly Internet of Things and BI)
market study report. In it, we review interest and demand for business intelligence in an
IoT world. We also examine key related technologies such as location intelligence, end
user data preparation, cloud computing, advanced and predictive analytics, and big data
analytics.
We hope you enjoy this report!
Best,
Howard Dresner
Chief Research Officer
Dresner Advisory Services

5
Contents
Business Intelligence: A Definition .................................................................................. 3
Internet of Things (IoT) Definition.................................................................................... 3
Introduction ..................................................................................................................... 4
Benefits of the Study ....................................................................................................... 8
Consumer Guide.......................................................................................................... 8
Supplier Tool................................................................................................................ 8
External Awareness.................................................................................................. 8
Internal Planning....................................................................................................... 8
About Howard Dresner and Dresner Advisory Services.................................................. 9
About Jim Ericson ......................................................................................................... 10
Survey Method and Data Collection.............................................................................. 11
Data Collection........................................................................................................... 11
Data Quality ............................................................................................................... 11
Executive Summary ...................................................................................................... 13
Study Demographics..................................................................................................... 14
Geography ................................................................................................................. 15
Functions ................................................................................................................... 16
Vertical Industries ...................................................................................................... 17
Organization Size....................................................................................................... 18
Analysis and Trends...................................................................................................... 20
IoT Importance........................................................................................................... 20
Technologies and Initiatives Strategic to Business Intelligence ................................. 28
Drivers of Business Intelligence................................................................................. 29
Business Intelligence Objectives................................................................................ 30
Business Intelligence Targets .................................................................................... 31
BI Penetration ............................................................................................................ 32
Business Intelligence Technologies and the Internet of Things..................................... 34
Advanced and Predictive Analytics ............................................................................ 34
Advanced and Predictive Analytics and IoT............................................................ 34

6
Importance of Advanced and Predictive Analytics.................................................. 35
Types of Users for Advanced and Predictive Analytics........................................... 36
Deployment Plans for Advanced and Predictive Analytics...................................... 37
Feature Requirements of Advanced and Predictive Analytics ................................ 39
Data Preparation Plans for Advanced and Predictive Analytics.............................. 40
Usability for Advanced and Predictive Analytics ..................................................... 41
Scalability of Advanced and Predictive Analytics.................................................... 42
Cloud Business Intelligence....................................................................................... 43
Cloud BI and IoT..................................................................................................... 43
Importance of Cloud BI........................................................................................... 44
Current and Future Plans for Cloud Business Intelligence ..................................... 45
Plans to Use Public Cloud Business Intelligence.................................................... 46
Cloud Business Intelligence Feature Requirements ............................................... 47
Cloud Business Intelligence Architecture................................................................ 48
Cloud Business Intelligence Security...................................................................... 49
Third-Party Cloud BI Data Connectors ................................................................... 50
End User Data Preparation........................................................................................ 51
End User Data Preparation and IoT........................................................................ 51
Importance of End User Data Preparation.............................................................. 52
Effectiveness of Current Approach to End User Data Preparation ......................... 53
Frequency of End User Data Preparation............................................................... 54
Frequency of End User Data Preparation Enrichment with Third-Party Data ......... 55
End User Data Preparation Usability Features ....................................................... 56
End User Data Preparation Integration Features.................................................... 57
End User Data Preparation Manipulation Features ................................................ 58
End User Data Preparation Supported Outputs...................................................... 59
End User Data Preparation Deployment Features.................................................. 60
Location Intelligence .................................................................................................. 61
Location Intelligence and IoT.................................................................................. 61
Importance of Location Intelligence ........................................................................ 62

7
Level of Geographic Detail Required...................................................................... 63
Prioritized Geocoding Features .............................................................................. 64
Targeted Users for Location Intelligence ................................................................ 65
Location Intelligence Features................................................................................ 66
Location Intelligence User Penetration ................................................................... 67
Mobile Location Intelligence Features .................................................................... 68
Required Integration with GIS Vendors .................................................................. 69
Big Data ..................................................................................................................... 70
Big Data and IoT..................................................................................................... 70
Big Data Adoption................................................................................................... 71
Future Adoption of Big Data ................................................................................... 72
Big Data Use Cases ............................................................................................... 73
Big Data Infrastructure............................................................................................ 74
Big Data – Data Access.......................................................................................... 75
Big Data Search...................................................................................................... 76
Big Data Analytics / Machine-Learning Technologies............................................. 77
Big Data Distributions ............................................................................................. 78
IoT Intelligence®
Vendor Ratings .................................................................................. 80
Other Dresner Advisory Services Research Reports .................................................... 81

8
Benefits of the Study
The DAS Internet of Things market study provides a wealth of information and
analysis—offering value to both consumers and producers of business intelligence
technology and services.
Consumer Guide
As an objective source of industry research, consumers use the DAS IoT market study
to understand how their peers leverage and invest in business intelligence and related
technologies.
Using our trademark vendor performance measurement system, users glean key
insights into BI software supplier performance, enabling:
 Comparisons of current vendor performance to industry norms
 Identification and selection of new vendors
Supplier Tool
Vendor licensees use the DAS IoT market study in several important ways. For
example:
External Awareness
- Build awareness for the business intelligence market and supplier brand, citing DAS
IoT market study trends and vendor performance
- Create lead and demand generation for supplier offerings through association with
the DAS IoT market study findings, webinars, etc.
Internal Planning
- Refine internal product plans and align with market priorities and realities as
identified in the DAS IoT market study
- Better understand customer priorities, concerns, and issues
- Identify competitive pressures and opportunities

9
About Howard Dresner and Dresner Advisory Services
The Business Intelligence Competency Center Market Study was conceived, designed
and executed by Dresner Advisory Services, LLC—an independent advisory firm—and
Howard Dresner, its President, Founder and Chief Research Officer.
Howard Dresner is one of the foremost thought leaders in business intelligence and
performance management, having coined the term ―Business Intelligence‖ in 1989. He
has published two books on the subject, The Performance
Management Revolution – Business Results through Insight
and Action (John Wiley & Sons, Nov. 2007) and Profiles in
Performance – Business Intelligence Journeys and the
Roadmap for Change (John Wiley & Sons, Nov. 2009). He
lectures at forums around the world and is often cited by the
business and trade press.
Prior to Dresner Advisory Services, Howard served as chief
strategy officer at Hyperion Solutions and was a research fellow at Gartner, where he
led its business intelligence research practice for 13 years.
Howard has conducted and directed numerous in-depth primary research studies over
the past two decades and is an expert in analyzing these markets.
Through the Wisdom of Crowds®
Business Intelligence market research reports, we
engage with a global community to redefine how research is created and shared. Other
research reports include:
- Advanced and Predictive Analytics
- Big Data Analytics
- Business Intelligence Competency Center
- Cloud Computing and Business Intelligence
- Collective InsightsTM
- Embedded Business Intelligence
- End User Data Preparation
- Enterprise Planning
- Location Intelligence
Howard (www.twitter.com/howarddresner) conducts a weekly Twitter ―tweetchat‖ on
Fridays at 1:00 p.m. ET. The hashtag is #BIWisdom. During these live events the
#BIWisdom community discusses a wide range of business intelligence topics.
You can find more information about Dresner Advisory Services at
www.dresneradvisory.com.

10
About Jim Ericson
Jim Ericson is Vice President and Research Director with Dresner Advisory Services.
Jim has served as a consultant and journalist who studies end-user management
practices and industry trending in the data and information management fields.
From 2004 to 2013 he was the editorial director at Information Management magazine
(formerly DM Review), where he created architectures for user and
industry coverage for hundreds of contributors across the breadth of
the data and information management industry.
As lead writer he interviewed and profiled more than 100 CIOs,
CTOs, and program directors in a 2010-2012 program called ―25
Top Information Managers.‖ His related feature articles earned
ASBPE national bronze and multiple Mid-Atlantic region gold and
silver awards for Technical Article and for Case History feature
writing.
A panelist, interviewer, blogger, community liaison, conference co-chair, and speaker in
the data-management community, he also sponsored and co-hosted a weekly podcast
in continuous production for more than five years.
Jim’s earlier background as senior morning news producer at NBC/Mutual Radio
Networks and as managing editor of MSNBC’s first Washington, D.C. online news
bureau cemented his understanding of fact-finding, topical reporting, and serving broad
audiences.

11
Survey Method and Data Collection
Data Collection
As with all of our Wisdom of Crowds market studies, we constructed a survey
instrument to collect data and used social media and crowdsourcing techniques to
recruit participants.
We expanded data collection to include our own research community of over 3,500
organizations as well as vendors’ customer communities.
Data Quality
We carefully scrutinized and verified all respondent entries to ensure that the study
includes only qualified participants.

12
Executive
Summary

13
Executive Summary
 Respondents in 2017 and over time continue to assign relatively weak interest to IoT
initiatives. Fewer than 15 percent consider IoT "critical" or "very important," and
nearly half say it is "not important.‖ BICC and R&D respondents and very large
enterprises are the most interested (pp. 20-27).
 IoT advocates show the most interest in initiatives including location intelligence,
streaming data analysis, and cognitive BI. They report standout interest in supply
chain and manufacturing. IoT advocates see the technology more in terms of
revenue and competitive opportunities (pp. 28-30).
 IoT advocates are more likely to see greater BI penetration in their organization and
more likely to target external constituencies with BI (pp. 31-33).
 IoT advocate interest in advanced and predictive analytics is flat; overall enterprise
adoption of APA is also low (pp. 34-42).
 IoT advocate interest in cloud BI/analytics is higher than in the overall sample, and
45 percent of advocates are likely to be current cloud users (pp. 43-50).
 IoT advocates consider end user data preparation more important than does the
entire population of our 2017 sample, but advocate sentiment toward data prep
declines more than among the sample at large year over year (pp. 51-60).
 IoT advocates are considerably more interested in location intelligence than the
overall population of our survey. Seventy-four percent say it is "critical" (pp. 61-69).
 Compared to all other parameters, IoT advocate sentiment toward big data is
strikingly high. Advocate sentiment grew year over year with more "critical" scores
and fewer "not important" or "somewhat important" scores (pp. 70-78).
 Vendor ratings are on page 80.

14
Study
Demographics

15
Study Demographics
For our third Internet of Things study, we gathered a cross-section of data across
geographies, functions, organization sizes, and vertical industries. We believe that,
unlike other industry research, this supports a more representative sample and is a
better indicator of true market dynamics. We constructed cross-tab analyses using
these demographics to identify and illustrate important industry trends.
Geography
About 57 percent of respondents work at North American organizations (including the
United States, Canada, and Puerto Rico). EMEA accounts for about 30 of respondents;
the remainder are distributed across Asia Pacific and Latin America (fig. 1).
Figure 1 – Geographies represented
56.9%
30.5%
8.6%
4.0%
0.0%
10.0%
20.0%
30.0%
40.0%
50.0%
60.0%
North America Europe, Middle East and
Africa
Asia Pacific Latin America
Geographies Represented

16
Functions
In 2017, Respondents by function are led by IT (29 percent) and Executive
Management (23 percent). Finance is the next most represented, followed by the BICC
and Operations.
Tabulating results across functions helps us develop analyses that reflect the
differences and influence of different departments within organizations.
Figure 2 – Functions represented
29%
23%
12%
7%
5% 5%
4%
3% 3%
11%
0%
5%
10%
15%
20%
25%
30%
35%
Functions Represented

17
Vertical Industries
In 2017, vertical industry sample participation is led by high-technology organizations
(15 percent). Financial Services, Healthcare, and Consulting are the next most
represented, followed by Higher Education, Retail/Wholesale, Energy and Business
Services (fig. 3).
Tabulating results across industries helps us develop analyses that reflect the maturity
and direction of different business sectors.
Figure 3 – Vertical industries represented
15%
9%
8% 8%
7%
5%
5%
3% 3%
2% 2% 2% 2% 2% 2% 2%
2% 2%
19%
0%
2%
4%
6%
8%
10%
12%
14%
16%
18%
20%
Vertical Industries Represented

18
Organization Size
Participation by organizations of different sizes (global employee head count) is
somewhat balanced in 2016. Small organizations (1-100 employees) represent about
28 percent of respondents, mid-size organizations (101-1,000 employees) represent 31
percent, and large organizations (>1,000 employees) account for the remaining 41
percent (fig. 4).
Tabulating results by organization size reveals important differences in practices,
planning, and maturity.
Figure 4 – Organization sizes represented
28%
31%
8%
10%
6%
17%
0%
5%
10%
15%
20%
25%
30%
35%
1-100 101-1,000 1,001-2,000 2,001-5,000 5,001-10,000 More than 10,000
Organization Sizes Represented

19
Analysis and
Trends

20
Analysis and Trends
IoT Importance
Across three years of our focused IoT study, our respondents report relatively weak
interest in Internet of Things technologies and initiatives. Interest in 2017 is almost
identical to 2016 levels and down from our first IoT study in 2015 (fig. 5). Fewer than 15
percent consider IoT "critical" or "very important." Beyond some select users, most
respondents clearly place IoT well behind more pressing issues of business intelligence
and data management. We believe wider adoption will remain low until more widely-
applicable use cases for IoT become apparent. On the positive side, a majority (53
percent) say IoT is, at minimum, ―somewhat important.‖
Figure 5 – Internet of Things importance 2015 to 2017
0%
5%
10%
15%
20%
25%
30%
35%
40%
45%
50%
Critical Very important Important Somewhat
important
Not important
Internet of Things Importance 2015-2017
2015 2016 2017

21
By geographic region, Latin American and Asia-Pacific respondents report the highest
perceived importance of IoT (fig. 6). While Asia Pacific is known as a region of early
adoption of new technologies, there is no obvious reason to explain why North America
or EMEA are regional laggards in measures of IoT importance. Nearly half of North
American and EMEA respondents say IoT is "not important," and close to 70 percent of
both groups say IoT is, at most, "somewhat important.‖
Figure 6 – IoT importance by geography
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Latin America Asia Pacific Europe, Middle
East and Africa
North America
IoT Importance by Geography
Not important
Somewhat important
Important
Very important
Critical

22
By function, BICC respondents (along with R&D) are the most likely to say IoT is, at
minimum, "important," reflecting project-centric rather than wide institutional interest in
the technology (fig. 7). Marketing/Sales, where a good bit of IoT marketing focuses,
(sales analytics/metrics) reports the next highest minimally "important" scores. Strategic
planning also shows above-average interest in IoT, and, along with R&D, is least likely
to say IoT is "not important.‖
Figure 7 – IoT importance by function
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Business
Intelligence
Competency
Center
Research and
Development
(R&D)
Information
Technology
(IT)
Executive
Management
Marketing &
Sales
Finance Operations Strategic
Planning
Function
Other - Write
In
IoT Importance by Function
Critical Very important Important Somewhat important Not important

23
In 2017, the response from a spectrum of industries indicates that IoT is applicable (if
still not undertaken) in many kinds of business environments (fig. 8). Manufacturing
respondents are least likely to say IoT is "not important. "Overall, Manufacturing,
Consulting, Business Services, Distribution/Logistics, and Energy report the greatest
number of, at minimum, "somewhat important" scores. Curiously, Federal and State
Government authorities, which oversee many diverse moving assets, are the least likely
to consider IoT important. Still, pluralities or majorities of respondents in most industries
find IoT less than important.
Figure 8 – IoT importance by industry
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
IoT Importance by Industry

24
As we often observe in complex technologies, IoT importance tends to increase as
organization headcount rises (fig. 9). In our 2017 sample, interest is highest in
organizations with10,000 employees and decreases as until headcount falls to 1,001-
2,000 employees. Thereafter, importance again rises at mid-sized organizations (101-
1,000 employees) and small organizations of up to 100 employees. (Interest among
small companies with one to 100 employees may be attributable in part to the
consulting respondents in our sample.) Despite this linearity, we do not believe scale
will be a hard-and-fast rule in a time when Internet-born organizations are capable of
using IoT to manage and monitor large populations of sensors or other devices.
Figure 9 – IoT importance by organization size
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
IoT Importance by Organization Size
Not important
Somewhat important
Important
Very important
Critical

25
Organizations that report success with business intelligence are more likely to attach
greater importance to the IoT (fig. 10). Among organizations that "completely agree" or
"agree somewhat" that their BI efforts have been successful, more than half say that IoT
is, at minimum, "somewhat important." While "critical" and "very important" scores fall as
BI success diminishes, a steady 31-35 percent of all groups say IoT is, at minimum,
"important."
Figure 10 – IoT importance by success with BI
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Completely agree Agree somewhat Disagree
somewhat
Disagree
IoT Importance by Success with BI
Not important
Somewhat important
Important
Very important
Critical

26
Enterprises with highly organized and accessible data are more likely to attach greater
importance to IoT (fig. 11). In 2017, this is primarily the case among organizations with
the highest state of treating "data as truth." Again, we cannot easily attach causation to
this finding (i.e., whether good data organization lends itself to IoT or follows it). It is not
surprising to us that organizations with their data well in place are more apt to
contemplate and become earlier adopters across a wide range of BI-related initiatives
and, in this case, IoT.
Figure 11 – IoT importance by state of data
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Data as "truth" - A
common view of
enterprise data is
available with common
application of data,
filters, rules, and
semantics
A common view of
enterprise data is
available. However,
parochial views and
semantics are used to
support specific
positions
Consistent data is
available at a
departmental level.
Conflicting, functional
views of data causes
confusion and
disagreement
We have multiple,
inconsistent data
sources with conflicting
semantics and data.
Information is generally
unreliable and
distrusted
IoT Importance by State of Data
Not important
Somewhat important
Important
Very important
Critical

27
Action on Insight is Dresner Advisory's high-level self-assessment of best (and worst)
practices in organizational use of data.
In our 2017 study, we find less correlation between measures of IoT importance and
actionable insight than in previous studies (fig. 12). Where previously, more evolved
organizations that work through "closed loop" processes were most likely to attach the
greatest importance to IoT, organizations with "uncoordinated/parochial" insight in 2017
are equally or more likely to value IoT. This finding tends to question the idea that IoT is
more important in highly collaborative workplaces. That said, the perceived importance
of IoT diminishes greatly in organizations where "insights are rarely leveraged."
Figure 12 – IoT importance by action on insight
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
"Closed loop" -
Information is
shared, teams
work to process
and act in a timely
fashion. No formal
boundaries
Ad hoc (informal)
action on insights
across functions
Uncoordinated/
parochial action
(sometimes at the
expense of others)
Insights are rarely
leveraged
IoT Importance by Action on Insight
Not important
Somewhat important
Important
Very important
Critical

28
Technologies and Initiatives Strategic to Business Intelligence
"IoT advocates,‖ respondents that rate IoT as either ―critical‖ or ―very important,‖ are
supportive of IoT in the context of a wide range of business intelligence activities and
processes (fig. 13). Areas of the greatest differing interest between advocates and the
broader audience not surprisingly include prime use cases such as location intelligence,
streaming data analysis, and cognitive BI. Interestingly, advocates do not see higher IoT
opportunities in the context of data warehousing. IoT advocates also do not show
outsized interest in BI staples (reporting and dashboards), which are highly ranked by
both groups; they see more importance in advanced visualization and end-user self-
service.
Figure 13 – Tech importance—IOT advocates versus overall sample
0
0.5
1
1.5
2
2.5
3
3.5
4
4.5
Reporting
Dashboards
Advanced visualization
End-user "self-service"
Data warehousing
Data discovery
Data mining, advanced algorithms, predictive
Data storytelling
Integration with operational processes
Mobile device support
Enterprise planning / budgeting
Embedded BI (contained within an
application, portal, etc.)
Governance
Collaborative support for group-based
analysis
End-user data preparation and blending
Data catalog
Search-based interface
Software-as-a-Service and cloud computing
In-memory analysis
Big Data (e.g., Hadoop)
Location intelligence / analytics
Ability to write to transactional applications
Streaming data analysis
Prepackaged vertical / functional analytical
applications
Cognitive BI (e.g., Artificial Intelligence-based
BI)
Text analytics
Natural language analytics (natural language
query/ natural language generation)
Social media analysis (Social BI)
Open source software
Complex event processing (CEP)
Edge computing
Video analytics
Tech Importance—IoT Advocates versus Overall
Sample
Overall 2017 IoT Advocates 2017
Social media analysis

29
Drivers of Business Intelligence
IoT advocates assign higher importance to all functional drivers of BI in their
organizations (fig. 14). Their greatest differing interest arises in the lower prioritized but
highly IoT-relevant areas of supply chain and manufacturing. Advocates are the next
most likely to see the importance of the BICC as a driver of IoT. Marketing, Strategic
Planning, and Sales are other strong drivers among IoT advocates.
Figure 14 – BI drivers—IoT advocates versus overall sample
0
0.5
1
1.5
2
2.5
3
3.5
4
Operations
Executive Management
Finance
Sales
Information Technology
(IT)
Strategic Planning
Function
Marketing
Research and
Development (R&D)
Competency
Center/Center of
Excellence
Human Resources
Supply Chain
Manufacturing
BI Drivers—IoT Advocates versus Overall Sample
Overall 2017 Advocates 2017

30
Business Intelligence Objectives
In 2017, IoT advocates rank all measured BI objectives higher than the overall sample
(fig. 15). The overall sample places the second-greatest emphasis on improved
operational efficiencies, but IoT advocates rank both increased competitive advantage
and revenue growth slightly ahead of operational gains. This likely indicates that
advocates look at IoT more opportunistically than does the broader audience.
Figure 15 – BI objectives—IoT advocates versus overall sample
0
0.5
1
1.5
2
2.5
3
3.5
4
4.5
5
Better decision-
making
Improved
operational
efficiencies
Increased
competitive
advantage
Growth in revenues Enhanced customer
service
BI Objectives—IoT Advocates versus Overall
Sample

31
Business Intelligence Targets
IoT advocates display higher affinity for targeting all audiences for BI and also drive BI
through the organization and beyond to external parties (fig. 16). Though internal
audiences still draw the greatest attention, the greatest differences between advocates
and the overall sample apply to customers and suppliers, which have always been the
lowest-ranked target audiences for BI deployment.
Figure 16 – BI targets—IoT advocates versus overall sample
0.00%
20.00%
40.00%
60.00%
80.00%
100.00%
120.00%
Executives Middle
managers
Line managers Individual
contributors
and
professionals
Customers Suppliers
BI Targets—IoT Advocates versus Overall Sample

32
BI Penetration
Compared to the overall audience, IoT advocates claim higher levels of business
intelligence penetration in their organizations (fig. 17). IoT advocates indicate fewer
instances of lower BI penetration (<20 percent) and more instances of mid-level
penetration (between 21-60 percent). Estimations of the highest penetration levels (>60
percent) are equal or slightly lower than in the overall population. Situationally, we
expect that many organizations with greater BI adoption are likely to find IoT more
useful.
Figure 17 – BI penetration—IoT advocates versus overall sample
0.00%
5.00%
10.00%
15.00%
20.00%
25.00%
30.00%
35.00%
40.00%
Under 10% 11 - 20% 21 - 40% 41 - 60% 61 - 80% 81% or more
BI Penetration—IoT Advocates versus Overall
Sample

33
IoT advocates describe higher 12-month mid-level and high-level penetration of
business intelligence to all potential users compared to the full sample of respondents
(fig. 18). Most dramatically, IoT advocates describe penetration between 41-60 percent
at almost twice the rate as the overall survey base. As a simple observation (also see
fig. 17), the entire sample is more likely to describe low (0-20 percent) and very high (81
percent or more) levels of penetration.
Figure 18 – BI 12-month penetration—IoT advocates versus overall sample
0.00%
5.00%
10.00%
15.00%
20.00%
25.00%
Under 10% 11 - 20% 21 - 40% 41 - 60% 61 - 80% 81% or more
BI 12-Month Penetration—IoT Advocates versus
Overall Sample

34
Business Intelligence Technologies and the Internet of Things
Advanced and Predictive Analytics
Advanced and predictive analytics includes statistics, modeling, machine learning, and
data mining to analyze facts to make predictions about future or otherwise unknown
events.
Advanced and Predictive Analytics and IoT
IoT advocates currently use or will use predictive analytics in the future to a greater
extent than the overall survey sample (fig. 19). Compared to the full survey base, IoT
advocates are more likely (27-23 percent) to be current users of advanced analytics or
about twice as likely (31-16 percent) to be currently evaluating APA.
Figure 19 – Use of advanced and predictive analytics—IoT advocates versus overall sample
0%
5%
10%
15%
20%
25%
30%
35%
40%
Yes. We use Advanced and
Predictive Analytics today
We may use Advanced and
Predictive Analytics in the
future
We are currently evaluating
Advanced and predictive
Analytics software
No. We have no plans to use
Advanced and Predictive
Analytics at all.
Use of Advanced and Predictive Analytics—IoT
Advocates versus Overall Sample

35
Importance of Advanced and Predictive Analytics
In 2017, perceived criticality of advanced and predictive analytics declines slightly
compared to both 2016 and 2015 (fig. 20). While not precipitous, this weakening
contrasts with the relatively high importance attached to the topic overall. That said,
advanced and predictive analytics remains, at minimum, "important" to more than 70
percent of respondents across four years of study.
Figure 20 – Importance of advanced and predictive analytics 2014-2016
0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
2014
2015
2016
2017
Importance of Advanced and Predictive Analytics
2014-2017
Critical Very important Important Somewhat important Unimportant

36
Types of Users for Advanced and Predictive Analytics
In 2017, the role of "BI expert" emerges as the most likely title for enterprise users of
advanced and predictive analytics (fig. 21). "Business analyst" is the second most cited
title, followed by "statistician / data scientist," "citizen data scientist," "financial analyst,"
and "marketing analyst." Executives and third-party consultants are the least likely users
of advanced and predictive analytics in 2017.
Figure 21 – Users of advanced and predictive analytics
0% 20% 40% 60% 80% 100%
BI Expert
Business Analyst
Statistician / Data Scientist
Citizen Data Scientist
Financial Analyst
Marketing Analyst
Executive
Third-Party Consultant
Users of Advanced and Predictive Analytics
Constantly Often Occasionally Rarely Never
Statistician/data scientist

37
Deployment Plans for Advanced and Predictive Analytics
Despite widespread regard for analytic proficiency and pockets of strong interest, we
observe that organizations are not broadly clamoring to adopt advanced and predictive
analytics in the current time frame. Penetration and use of advanced and predictive
analytic tools remains low, with current users accounting for just 23 percent of our
sample (organizations reporting current use of APA declined 1 percent year over year).
Thirty-nine percent of organizations are either using or evaluating the technologies, but
35 percent are only considering advanced analytics. Twenty-six percent have no plans
for use (up 3 percent from 2016) (fig. 22).
Figure 22 – Current deployment of advanced and predictive analytics
We may use
advanced and
predictive analytics
in the future, 35%
No. We have no
plans to use
advanced and
at all., 26%
Yes. We use
advanced and
today, 23%
We are currently
evaluating advanced
and predictive
analytics software,
16%
Current Deployment of advanced and predictive
analytics

38
Among respondent organizations evaluating or planning the use of advanced and
predictive analytics, fewer than half (45 percent) are adopting APA in 2017 or 2018 (fig.
23). Fifty-five percent say they will adopt beyond 2018. Coupled with current users, we
conclude that APA shows no signs of becoming a common or widespread practice in
the near future.
Figure 23 – Deployment plans for advanced and predictive analytics
Will adopt beyond
2018, 55%
Will adopt in 2017,
19%
Will adopt in 2018,
27%
Deployment Plans for Advanced and Predictive
Analytics

39
Feature Requirements of Advanced and Predictive Analytics
Respondents expressed interest in a broad range of feature requirements for advanced
and predictive analytics (fig. 24). The most popular among these support traditional
statistical methods: regression models, textbook statistical functions, and hierarchical
clustering. These three leading features are, at minimum, "somewhat important" to more
than 90 percent of 2017 respondents. Geospatial analysis (highly associated with
mapping, populations, demographics, and other Web-generated data), recommendation
engines, Bayesian methods, and automatic feature selection are the next most required
features.
Figure 24 – Features for advanced and predictive analytics
0% 20% 40% 60% 80% 100%
Range of regression models, from linear,
logistic to nonlinear
Textbook statistical functions for
descriptive statistics
Hierarchical clustering, expectation
maximization, k-Means, and variants of…
Geospatial analysis
Recommendation engine included
Bayesian methods, including Naive Bayes
and Bayesian Networks
Automatic feature selection like principal
component analysis (PCA)
Text analytic functions and sentiment
analysis
Various approaches to CART (e.g. ID3,
C4.5, CHAID, MARS, random forests,…
Neural networks supported
Vector machine (SVM) approaches for
classification and estimation
Ensemble learning
Video analysis
Features for Advanced and Predictive Analytics

40
Data Preparation Plans for Advanced and Predictive Analytics
For a fourth year, our study addresses a detailed set of data preparation features that
support advanced and predictive analytic activities and processes.
As in our previous studies, all six data preparation features earn very respectable
attention, led only slightly in 2017 by the simplest: set operations and detection of
duplicates (fig. 25). But 60-69 percent of respondents say every feature measured is, at
minimum, "very important," a reflection of the criticality of preparing and manipulating
data in the context of applying advanced and predictive analytics.
Figure 25 – Data preparation for advanced and predictive analytics
0% 20% 40% 60% 80% 100%
Set operations (e.g., joins, aggregations or
pivot tables)
Detection of duplicates or outliers
Complex filtering
Cleansing and enrichment of source data
Support for data type conversions
Support for cutting, merging, and replacing
of values
Data Preparation for Advanced and Predictive Analytics

41
Usability for Advanced and Predictive Analytics
Our study addresses a detailed set of usability benefits that support APA activities and
processes. Based on our specified criteria, we expect our respondent audience to be
fairly sophisticated in usability criteria compared to a mainstream business intelligence
audience.
Usability features generally address process or activity automation and streamlining.
Across nine usability criteria sampled, all but one are at least "important" to 80 percent
or more of respondents (fig. 26). Thus, at the bottom of this list, "specialist not required"
is important, but less so than access other features that are led by support for "easy
iteration," "simple process for continuous modification of models," and "access to
advanced analytics for predictive and temporal analysis."
Figure 26 – Usability for advanced and predictive analytics
0% 20% 40% 60% 80% 100%
Support for easy iteration
Simple process for continuous
modification of models
Access to advanced analytics for predictive
and temporal analysis
Support for entire process in a single
application/user interface
Fast cycle time for analysis with data
preparation functions
Support/guidance in preparing data
analytical models
Pre-built drag and drop macros and tools
from R that require no scripting or
programming
Automatic creation of models from data
A specialist NOT required to create
analytical models, test and run them
Usability for Advanced and Predictive Analytics

42
Scalability of Advanced and Predictive Analytics
In 2017, two features, in-memory analytics and in-database analytics, stand out most
sharply to respondents, after which interest notably declines (fig. 27). Both technologies
attract the highest "critical" and "very important" scores in our study and are, at
minimum, ―somewhat important" to more than 90 percent of respondents. In-Hadoop
analytics are the third most cited scalable analytic platform, followed by support for
massively parallel processing architecture and PMML support.
Figure 27 – Scalability of advanced and predictive analytics
0% 20% 40% 60% 80% 100%
In-Memory analytics
In-Database analytics
In-Hadoop analytics (on file system)
Support for MPP architecture
PMML Support
Scalability of Advanced and Predictive Analytics

43
Cloud Business Intelligence
Cloud business intelligence includes the technologies, tools, and solutions that employ
one or more cloud deployment models, including public cloud. Public cloud
infrastructure is provisioned for open use by the general public. It may be owned,
managed, and operated by a business, academic, or government organization, or some
combination of them. It exists on the premises of the cloud provider.
Cloud BI and IoT
IoT advocates currently use or will use cloud-based BI/analytics in the future to a
greater extent than the overall survey sample (fig. 28). Compared to the full survey
base, IoT advocates are more likely (45 percent compared to 31 percent) to be current
users of cloud-based BI analytics or more likely (14 percent compared to 10 percent) to
be currently evaluating cloud.
Figure 28 – Cloud BI—IoT advocates versus overall sample
0%
5%
10%
15%
20%
25%
30%
35%
40%
45%
50%
Yes. We use Cloud-based
BI/ Analytics today
We may use Cloud-based
BI/ Analytics in the future
We are currently
evaluating Cloud-based
BI/ Analytics software
No. We have no plans to
use Cloud-based BI/
Analytics at all.
Cloud BI—IoT Advocates versus Overall Sample

44
Importance of Cloud BI
Across six years of study, the perceived importance of cloud BI remains mostly
consistent (fig. 29). Cloud deployments continue to increase over time and, for most
organizations, cloud is an important but not burning issue for users. The small peak in
2014 might well reflect a peak of inflated expectations that settled cloud into the
infrastructure mix of topics. Certainly, smaller and younger "straight-to-cloud"
organizations might be most interested, familiar, and aware; though we believe by now
most organizations are familiar with the opportunities of cloud computing and software
as a service.
Figure 29 – Cloud BI importance 2012-2017
1.1
1.6
2.1
2.6
3.1
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
2012 2013 2014 2015 2016 2017
Cloud BI Importance 2012-2017
Not important
Somewhat important
Important
Very important
Critical
Mean

45
Current and Future Plans for Cloud Business Intelligence
Fig 30 shows year-over-year changes in the use of and plans for cloud BI (without
regard for public, private, or hybrid cloud BI models). As we would expect, cloud use
increases over time, a trend that should continue as companies retire older
infrastructure. In 2017, 31 percent of respondents say they currently use cloud BI /
analytics, a 6 percent increase from 2016. The number evaluating is mostly steady
while the number of those considering the use of cloud BI/analytics fell 5 percent to
offset the increase in current users. In both 2016 and 2017, more than one-third of
respondents say they still have no plans to use cloud BI / analytics.
Figure 30 – Plans to use cloud BI 2016-2017
0%
5%
10%
15%
20%
25%
30%
35%
40%
Yes. We use cloud-based
BI / Analytics today
We are currently
evaluating cloud-based BI
/ Analytics software
We may use cloud-based
BI / Analytics in the
future
No. We have no plans to
use cloud-based BI /
Analytics at all.
Plans to Use Cloud BI 2016-2017
2016 2017

46
Plans to Use Public Cloud Business Intelligence
Across six years of data, actual use and favorable attitudes toward future use of public
(multitenant) cloud BI increases in near linear fashion (fig. 31). In 2017, one-third of all
respondents currently use public cloud (compared to 13 percent in our inaugural 2012
study). However, year-over-year, 12-month, and 24-month plans decrease slightly (as in
2016), perhaps indicating a near-term plateau. We believe generic definitions of public
cloud BI are by now well understood to imply single instance, off-premises, multitenant
services. While the majority of respondents still resist public cloud, we expect
Marketing/Sales, Executive Management, and other underserved functions with
relatively urgent goals and time frames will remain first movers in public cloud BI use.
Figure 31 – Plans for public cloud BI 2012-2017
13%
17% 20%
24%
29%
33%7%
6%
6%
9%
9%
6%
8%
6%
6%
10%
9% 7%
73% 71% 68%
57%
53% 51%
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
2012 2013 2014 2015 2016 2017
Plans for Public Cloud BI 2012-2017
No plans
Next year
This year
Today

47
Cloud Business Intelligence Feature Requirements
Respondents show interest in a wide range of cloud BI features. In 2017, conventional
BI functionality, led by ad-hoc query, advanced visualization, dashboards, DI/DQ, end-
user self-service, and reporting lead the list of the most-required cloud BI features (fig.
32). These top features are, at minimum, somewhat important to more than 90 percent
of respondents. Data mining, data discovery, search interface, and end-user data
blending are respondents’ next most important features. Overall, we believe cloud BI
feature requirements closely mirror traditional BI preferences.
Figure 32 – Cloud BI feature requirements
0% 20% 40% 60% 80% 100%
Ad-hoc query
Advanced visualization
Personalized dashboards
Data integration / data quality…
End-user "self-service"
Production reporting
Data mining and advanced…
Data discovery
Search interface
End-user data blending or…
Data catalog
Location intelligence/analytics
Collaborative support for group-…
Big data (e.g., Hadoop) support
Prepackaged vertical/functional…
In-memory support
Ability to write to transactional…
Natural language analytics
Complex event processing (CEP)
Text analytics
Social media analysis (Social BI)
Cloud BI Feature Requirements

48
Cloud Business Intelligence Architecture
Architectural requirements for cloud BI are spread across several features of interest
(fig. 33). The most important architectural feature for cloud BI in 2017 (as in 2016) is
relational database support, which is either "critical" or "very important" to 71 percent of
respondents. Automatic upgrades, open client connectors, and connectors to on-
premises apps are the next most important requirements. The breadth of architectural
requirements is notable with every category, except virtualization, at or above 50
percent "critical" or "very important."
Figure 33 – Cloud BI architectural requirements
0% 20% 40% 60% 80% 100%
Relational database support
Automatic upgrades
Open client connector (e.g., ODBC,
JDBC)
Connectors to on-premises
applications and data (e.g., ERP,…
RESTful/Web Services API
Multi-dimensional database
support
Real-time query to third-party
cloud applications, cloud data…
Cloud database connectors (e.g.,
database.com, Redshift,…
Data virtualization
Cloud application connections (e.g.,
Salesforce, NetSuite)
Multi-tenancy (single executable
supporting multiple customers)
Cloud BI Architectural Requirements

49
Cloud Business Intelligence Security
We asked respondents to identify cloud BI security requirements in 2017 and, for a third
year, the most popular response is "don't know" (fig. 34). This finding likely represents a
gap between general awareness and those in positions of responsibility. Among
standards, ISO 27001 remains the most required security standard across all six years
of our study. HIPAA, SAS 70 AICPA, and PCI/DSS are the standards next most cited as
requirements.
Figure 34 – Cloud BI security requirements
0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
TRUSTe
FIPS 140-2 (Federal Information Processing Standard)
FERPA (Family Educational Rights and Privacy Act)
FISMA (Federal Information Security Management Act )
None
SOC 2 TYPE 2
PCI DSS (Payment Card Industry Data Security Standard)
SAS 70 AICPA Auditing Standard (now SSAE 16)
HIPAA (Health Insurance Portability and Accountability
Act)
ISO 27001 (Specification for Information Security
Management System)
Don't know
Cloud BI Security Requirements

50
Third-Party Cloud BI Data Connectors
Beginning in 2015, we asked respondents to specify preferred third-party connectors for
cloud BI. Since then, the top two choices remain Google Analytics followed by
Salesforce.com (fig. 35). In 2017, Dropbox falls to fifth place, surpassed by Google
Drive and Facebook. Twitter also falls behind LinkedIn and SurveyMonkey. We believe
these connectors (though clearly not in universal demand) reflect both the popularity of
off-premises data platforms and the value of data within. They also reflect demand for
cloud BI integration with a variety of Web-based sources versus conventional practices
addressing in-house resources.
Figure 35 – Third-party data connectors for cloud BI
0%
10%
20%
30%
40%
50%
60%
70%
Cloud BI Third-Party Data Connectors

51
End User Data Preparation
End user data preparation is a "self-service" capability for end users to model, prepare,
and combine data prior to analysis. This may complement traditional IT-driven data
quality/ETL processes or may be used independently.
End User Data Preparation and IoT
IoT advocates, by all measures, consider end user data preparation more important
than does the entire population of our 2017 sample (fig. 36). However, year-over-year
data prep sentiment declines more among IoT advocates than among the sample at
large. In 2017, IoT advocates are 50 percent likely to consider end user data prep
"critical" or "very important" (compared to 61 percent in 2016). Fewer IoT advocates
consider end user data prep to be, at minimum, "important" in 2017 versus 2016.
Overall sentiment toward end user data prep is more consistent year over year, perhaps
indicating that advocates are turning their attention away from data prep in the context
of IoT.
Figure 36 – Importance of end user data preparation—IoT advocates versus overall sample
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Overall 2016 Advocates 2016 Overall 2017 Advocates 2017
Importance of End User Data Preparation—IoT
Not important
Somewhat important
Important
Very important
Critical

52
Importance of End User Data Preparation
In our third year of focused study of end-user data preparation, respondents’ perceive
importance of end user data preparation is very high and in line with user demands for
self-service business intelligence and user autonomy (fig. 37). Sixty-seven percent of all
respondents say end user data preparation is either ―critical‖ or ―very important.‖ About
88 percent of respondents say end user data preparation is, at minimum, ―important.‖
Just 3 percent say end user data preparation is ―not important.‖
Figure 37 – Importance of end user data preparation
Critical, 34%
Very important, 33%
Important, 21%
Somewhat
important, 10%
Not important, 3%
Importance of End User Data Preparation

53
Effectiveness of Current Approach to End User Data Preparation
In 2017, a large majority of organizations say their current end user data preparation
approach is "highly effective" or "somewhat effective" (fig. 38). Just 4 percent say their
current approach is "totally ineffective." These results imply good levels of interaction
and experience with end user data preparation, likely in the context of self-service and
user autonomy, which are prime drivers for data preparation and business intelligence
generally.
Figure 38 – Effectiveness of current approach to end user data preparation
Highly effective, 18%
Somewhat effective,
56%
Somewhat
ineffective, 23%
Totally ineffective,
4%
Effectiveness of Current Approach to End User
Data Preparation

54
Frequency of End User Data Preparation
Sixty-seven percent of respondents say they "constantly" or "frequently" make use of
end user data preparation (fig. 39). We cannot distinguish whether end user efforts are
one-off or regular practice, but overall use of end user data preparation is high. Twenty-
nine percent say they only "occasionally" require end user data preparation; the
remaining 6 percent "rarely" or "never" do.
Figure 39 – Frequency of end user data preparation
Constantly, 26%
Frequently, 41%
Occasionally, 29%
Rarely,
4%
Never, 1%

55
Frequency of End User Data Preparation Enrichment with Third-Party Data
The majority of organizations "constantly," "frequently," or "occasionally" enrich end
user data preparation with third-party data (fig. 40). Still, just 6 percent are "constant"
users of non-proprietary data. Overall, we see a fairly broad spectrum of third-party data
use; 22 percent are "frequent" users while another 22 percent "rarely" use third-party
data.
Figure 40 – Frequency of end user data preparation enrichment with third-party data
Constantly
6%
Frequently
22%
Occasionally
38%
Rarely
22%
Never
12%
Enrichment with Third-Party Data

56
End User Data Preparation Usability Features
Respondents have high interest in a full range of end-user data preparation usability
features, all of which they consider "important" to "very important" (fig. 41). We believe
this reflects good understanding of needs and high expectations for data preparation
features associated with BI/analytics usage. A feature we added for 2017, "immediate
preview and feedback," debuted as a top requirement and is at least "very important" to
almost 80 percent of respondents. "Visual user interface" and "technical expertise not
required" are also very important to large majorities of respondents. Together, these
features reflect user demand for easy and intuitive guided and visual environments for
data preparation.
Figure 41 – End user data preparation usability features
0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
Immediate preview and feedback for end user
Visual interface for users to view and explore in-
process data sets, interactively profile and refine…
Technical expertise/programming is *NOT* required
to build/execute data transformation scripts
Automated detection of anomalies, outliers, and
duplicates
Visual highlighting of relationships between
columns, attributes, and datasets
Support for entire data transformation process in a
single application/user interface
Automated recommendations for data relationships
and keys for combining data across multiple data…
Automatically generate data transformation
code/scripts for execution
Machine learning and recommendations based on
usage data gathered across users, groups, or…
End User Data Preparation Usability Features
Critical Very important Important Somewhat important Not important Don't know
Automated detection of anomalies,
outliers and duplicates

57
End User Data Preparation Integration Features
Though not quite as pronounced as usability, demand for end-user data preparation
integration features is nonetheless quite strong in 2017 (fig. 42). The top three features,
"access to multiple common file formats," "access to traditional databases," and "ability
to combine data through joins/merges" (the most conventional integration modes) are,
at minimum, "very important" to about 80 percent or more respondents. Trailing these,
the ability to infer metadata is at least ―important‖ to 78 percent of respondents. Big data
and NoSQL demand are notably lower, but we can conclude user expectations for
integration are undeniably high overall.
Figure 42 – End user data preparation integration features
0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
Access to file formats (e.g., log files, CSV, Excel)
Access to traditional databases (e.g., RDBMS)
Ability to combine data across multiple data sets
and sources through joins and merging data
Ability to infer metadata by introspecting the data
elements
Access to Big data (e.g., Hadoop)
Access to NoSQL sources
End User Data Preparation Data Integration
Features
Access to big data (e.g., Hadoop)

58
End User Data Preparation Manipulation Features
We asked organizations to gauge their interest in specific data-manipulation features
and once again found a very high and broad level of interest. The top two features,
"ability to aggregate and group data" and "ability to pivot data," stand out as most critical
to users (fig. 43). The top seven manipulation feature priorities are all "critical" or "very
important" to 60-81 percent of respondents.
Figure 43 – End user data preparation manipulation features
0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
Ability to aggregate and group data
Ability to pivot (convert table to matrix) and reshape
(convert matrix to table) data
Simple interface for imposing structure on raw data
Ability to derive new data features from existing data (text
extraction, math expressions, date expressions, etc.)
Ability to normalize, standardize and enrich data
Support for cutting, merging and replacing of values
Ability to manipulate the order of data transformation
steps
Window and time series functions
Custom user-defined functions
Ability to unnest data (e.g., json/xml parsing)
Session-ize log or event data
End User Data Preparation Manipulation Features

59
End User Data Preparation Supported Outputs
Respondents say the most important data prep outputs are to flat files formats, outputs
to databases, and direct to business intelligence tools (fig. 44). Newer, proprietary and
more exotic outputs are, by comparison, unimportant to respondents. For example,
users are about four times more likely to seek flat file outputs than outputs for Hadoop,
a chasm that only becomes more dramatic in the case of Redshift, Azure, and other
formats.
Figure 44 – End user data preparation supported outputs
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Bizp/gizpAvroParquetAzureRedshiftHadoopPopular
(third-party)
business
intelligence
tool formats
Traditional
relational
database
(e.g., SQL
Server)
Excel, CSV
End User Data Preparation Supported Outputs

60
End User Data Preparation Deployment Features
We asked respondents about their preferences for scheduling, monitoring, and testing
aspects that make end user data preparation more of a formal and ongoing process (fig.
45). While this resonates less than other end user data preparation capabilities, the two
most popular features, "ability to schedule execution/replay of data transformation" and
"ability to iteratively sample data," are either "critical" or "very important" to more than 60
percent of respondents. Among other deployment features, interest in "API support" and
"support for multiple execution environments" is less than we expected for data
preparation deployment.
Figure 45 – End user data preparation deployment features
0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
Ability to schedule the execution/replay of data
transformation processing
Ability to iteratively sample data to provide an
interactive testing of transformation logic
Ability to monitor ongoing data transformation
processing to alert on anomalies or changes in the
structure
Push-down processing of data transformations into
the native data source for script execution (SQL, Pig,
etc)
API support (e.g., REST)
Support for multiple execution environments (e.g.,
MapReduce, Spark, Hive) based on volume and scale
of data sets
End-User Data Preparation Deployment Features

61
Location Intelligence
Location Intelligence is a form of business intelligence where the dominant dimension
used for analysis is location or geography. Most typically, though not exclusively,
analyses are conducted by viewing data points overlaid onto an interactive map
interface.
Location Intelligence and IoT
IoT advocates are considerably more interested in location intelligence than the overall
population of our survey (fig. 46). Also, advocate and overall sentiment remains
consistent year over year. Seventy-four percent of IoT advocates say location
intelligence is ―critical‖ or "very important, almost identical to the prior year. We are not
surprised at this finding since location is often a critical use for IoT data analysis. Only
26 percent of the overall sample shares this "critical" or "very important" view in 2017.
Figure 46 – Importance of location intelligence—IoT advocates versus overall sample
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Overall 2016 Advocates 2016 Overall 2017 Advocates 2017
Importance of Location Intelligence—IoT
Not important
Somewhat important
Important
Very important
Critical

62
Importance of Location Intelligence
The perceived importance of location intelligence increases across four years of study
and experiences a notable spike in interest in 2017 (fig. 47). While earlier studies reflect
a level of "stickiness," our latest finding indicates location intelligence might be gaining
new traction among BI adherents as more sources of data come online. Nonetheless,
we expect prospects for location intelligence will vary strongly from organization to
organization. In 2017, a mean of 2.75 places overall importance toward a level of
"importance."
Figure 47 – Location intelligence importance 2014-2017
0.00
0.50
1.00
1.50
2.00
2.50
3.00
3.50
4.00
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
2014 2015 2016 2017
Importance of Location Intelligence
2014-2017
Not important
Somewhat important
Very important
Critical
Mean

63
Level of Geographic Detail Required
The granularity or level of geographic detail that respondents find important speaks to
the business processes or analysis related to location intelligence. Postal codes or
countries, for example, are useful to supply chain or fulfillment but also to sales region
performance or demographics. Latitudinal/longitudinal or custom geographies (less
related to a physical address) are important assets for discovery/recovery, natural
resources, wellheads, or unmarked boundaries.
Respondents are interested in and/or pursuing activity across a range of marked,
unmarked, and virtual location parameters (fig. 48). In 2017, conventional demarcations,
led by postal code and province/state, are considered critical or very important to more
than 70 percent of respondents. Country-level detail, like postal codes are "critical" to
about 45 percent of respondents. Latitude/longitude is the next most important level of
geographic detail, after which criticality drops among custom demarcations.
Figure 48 – Level of geographic detail
1
1.5
2
2.5
3
3.5
4
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Level of Geographic Detail
Critical
Very important
Somewhat important
Not important
Mean

64
Prioritized Geocoding Features
Consistent with our earlier studies, 2017 respondents place the greatest importance (44
percent "critical" and 39 percent "very important") on built-in or native geocoding (fig.
49). (This feature presumes that software understands geography detail "baked in.")
About 67 percent of respondents say automated geocoding (in which software
automatically recognizes geographic data elements, for example, address) is ―critical‖ or
―very important.‖ Street-level geocoding is ―critical‖ or ―very important‖ to 60 percent of
respondents. (This requirement assumes the software can cross-reference
latitude/longitude and street-level address.) While interest in customer extensions,
offline, and worldwide support trails off, these features are nonetheless ―critical‖ or ―very
important‖ to large minorities of respondents.
Figure 49 – Prioritized geocoding features
0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
Worldwide geocoding support
Offline geocoding support
Customer extensions to map data (e.g., custom
POIs)
Street-level geocoding support
Automated geocoding support
Built-in (native) geocoding (e.g., country, region,
postal code, CBSA)
Prioritized Geocoding Features
Critical Very important Somewhat important Not important

65
Targeted Users for Location Intelligence
Managers are the most targeted users of location intelligence in 2017, which in part
speaks to the operational nature of the technology (fig. 50). Middle Managers are
frequent or occasional targets more than 80 percent of the time. Individual Contributors
and Line Managers are frequent or occasional users 70-80 percent of the time. This
targeting comes at the expense of Executives, traditionally a first-served audience (and
the most-served group in our earlier studies), which may indicate increasing/maturing
penetration of location intelligence. Customers and Suppliers are relatively low priorities.
Figure 50 – Targeted users for location intelligence
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Middle
managers
Individual
contributors
and
professionals
Line
managers
Executives Customers Suppliers
Targeted Users for Location Intelligence
Frequently
Occasionally
Rarely
Not at all

66
Location Intelligence Features
As we found in our earlier studies, the most important location intelligence features
identified by respondents in 2017 are also the most conventional: map-based
visualization of information, drill-down navigation that includes zoom and pan, and maps
that are embedded in dashboards or other displays (fig. 51). These top three categories
are at least ―somewhat important‖ to more than 90 percent of respondents. Layered
visualizations and value/range shading are the two next most important features and
are considered ―critical‖ or ―very important‖ by close to 70 percent or more of
respondents. Integration with third-party GIS systems (e.g., Google, Esri) gained
momentum compared to earlier studies and is "critical" or "very important" to nearly 70
percent of respondents.
Figure 51 – Location Intelligence prioritized features
0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
Support for interior spaces (e.g., retail stores,…
Off-line mapping
Syndicated demographic/psychographic data…
Animation of data on maps
Support for location calculations (e.g., drive time,…
Use of symbols to depict values
Choropleths (area fill) maps
Custom region definition and selection (e.g.,…
Integration with third-party GIS systems (e.g., ESRI,…
Value/range-based shading of maps
Layering of visualizations on top of maps (e.g.,…
Dashboard inclusion of maps
Drill-down navigation through map interface
Map-based visualization of data/information
Location Intelligence Prioritized Features
Not important Somewhat important Very important Critical
-

67
Location Intelligence User Penetration
Location intelligence penetration remains modest in 2017, though respondents describe
significant growth plans (fig. 52). Today, 59 percent of respondents report less than 10
percent penetration, and only 6 percent report the highest level of penetration. Twelve-
month plans (the most dependable and likely to be budgeted) call for reducing the
lowest penetration to about 30 percent and more than doubling 11-20 percent and 41-
60 percent penetration levels. Forecasts for 24 and 36 months call for extended growth
of higher levels of penetration.
Figure 52 – Location Intelligence user penetration 2017-2020
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Today In 12 months In 24 months In 36 months
Location Intelligence User Penetration
2017-2020
81% or more
61 - 80%
21 - 40%
41 - 60%
11 - 20%
Under 10%

68
Mobile Location Intelligence Features
For a fourth year, we measured the perceived criticality of three mobile location
intelligence feature priorities: location-based query filtering, reverse geocoding, and geo
fence alerting. In 2017, respondents strongly choose mobile location-based query
filtering (the software’s ability to query data using current location as a filter parameter)
as their highest priority (fig. 53). This feature is "critical" to 36 percent of respondents
and very important to another 42 percent. Less than 10 percent say location-based
query filtering is "not important." While less critical, the other two features sampled—
reverse geocoding (creating a physical address or place name from coordinate
information) and geo fence alerting (alerting when a device crosses a defined boundary)
—are, at minimum, "somewhat important" to 85 and 78 percent of respondents
respectively.
Figure 53 – Mobile location intelligence feature priorities
1
1.5
2
2.5
3
3.5
4
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Location-based query
filtering
Reverse geocoding Geo fence alerting
Mobile Location Intelligence Features
Not important
Somewhat important
Very important
Critical
Mean

69
Required Integration with GIS Vendors
Across four years of data, we see some notable trending in location intelligence
integration with GIS technologies requirements (fig. 54). Most notably, sentiment toward
Google integration declined 14 percent compared to 2014, interest in Esri grew 9
percent year over year, and database extensions interest grew by a similar amount. All
remaining GIS technologies are down or grew minimally for 2017.
Figure 54 – Required Integration with GIS vendors 2014-2017
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Google Esri Database
extensions
(e.g., Oracle,
Postgres)
Microsoft
(Bing)
MapInfo WMS (Web
Map
Service)
TomTom NAVTEQ
Location Intelligence Integration with GIS
Technologies 2014-2017
2014 2015 2016 2017

70
Big Data
Dresner Advisory Services defines big data analytics as systems that enable end user
access to and analysis of data contained and managed within the Hadoop ecosystem.
Big Data and IoT
Compared to all other parameters, IoT advocate sentiment toward big data is strikingly
high compared to overall respondent sentiment toward big data (fig. 55). IoT advocate
sentiment also grew year over year with more "critical" scores and fewer "not important"
or "somewhat important" scores. In 2017, IoT advocates are more than twice as likely to
consider big data "critical" and about one-fifth as likely to say big data is "not important."
As we noted in earlier studies, some advocates argue that the IoT is a core justification
for big data technologies and architectures.
Figure 55 – Importance of big data—IoT versus overall sample
0%
5%
10%
15%
20%
25%
30%
35%
40%
Critical Very important Important Somewhat
important
Not important
Importance of Big Data—IoT Advocates versus
Overall Sample
Overall 2016
Advocates 2016
Overall 2017
Advocates 2017

71
Big Data Adoption
Current adoption and future plans for the use of big data analytics reach a level of
significance we did not see last year. Forty-one percent of organizations say they are
already using big data analytics (fig. 56), which we define as "systems that enable end
user access to and analysis of data contained and managed within the Hadoop
ecosystem.‖ Even more respondents (46 percent) say they may use big data in the
future. Just 14 percent have no plans for future use of big data analytics.
Figure 56 – Adoption of big data
Yes. We use big
data today, 41%
We may use big
data in the future,
46%
No. We have no
plans to use big
data at all, 14%
Adoption of Big Data

72
Future Adoption of Big Data
Among organizations that have not yet adopted big data but have future plans, 14
percent say they will adopt in the current calendar year (fig. 57). This horizon grows
rapidly in 2017 when 47 percent plan to adopt. Unlike 2015, only a minority of non-users
of big data adopters are postponing plans beyond 2017. Though we often find big data
plans compartmentalized to projects or departments, future adoption also will hinge on
current investment budgets for more "conventional" technologies.
Figure 57 – Future adoption of big data
Will adopt in 2016,
14%
Will adopt in 2017,
47%
Will adopt beyond
2017, 40%
Future Adoption of Big Data

73
Big Data Use Cases
The top big data use case in 2016 is data warehouse optimization, which is considered
critical or very important to 65 percent of respondents (fig. 58). As data warehouse
deployments are mostly confined to large institutions, this reinforces our view that big
data is predominantly a large-organization pursuit meant to lower cost and complexity.
That said, customer / social analysis is the next most likely use case and is, at
minimum, "very important" to a majority of respondents.
Figure 58 – Big data use cases
0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
Data warehouse optimization
Customer/ social analysis
Clickstream analytics
Fraud detection
Internet of Things
Big Data Use Cases

74
Big Data Infrastructure
To gather baseline data on big data infrastructure awareness/adoption, we assembled a
list of relevant frameworks, databases, and other technologies in the Hadoop / open
source orbits of interest (fig. 59). Spark is the preferred mechanism followed by
Map/Reduce, Yarn, Oozie, Tez, Mesos, and Atlas. Spark and Map/Reduce notably
stand out across multiple grades of importance. All but the top three choices (Spark,
Map/Reduce, Yarn) are "not important" or only "somewhat important" to the majority of
respondents.
Figure 59 – Big data infrastructure
0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
Spark
Map/Reduce
Yarn
Oozie
Tez
Mesos
Atlas
Knox Gateway
Alluxio (formerly Tachyon)
Big Data Infrastructure

75
Big Data – Data Access
We asked organizations which big data structure access they prefer and which is
more/most important to them. This includes indirect access to Hadoop and other related
engines. In our latest study, Spark SQL is the most cited and considered, at minimum,
―important‖ to close to 80 percent of the sample (fig. 60). Hive and HDFS, perhaps more
familiar to the conventional data warehousing audience, follow closely and elicited even
more "critical" responses than Spark.
Figure 60 – Big data—data access
0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
Spark SQL
Hive/HiveQL
HDFS
HBase
Google BigQuery
Redshift
MongoDB
Impala
Pivotal HAWQ
Big Data - Data Access

76
Big Data Search
We asked respondents to rank interest in big data search facilities, which in Hadoop
include indexing and natural language textual search (fig. 61). In our latest sample,
Elasticsearch resonated most strongly followed by Apache Solr and Cloudera Search.
Despite shifting over time (which we will expand on in the following figure) there is no
clear first choice in big data search; all three technologies are, at minimum, "important"
to 65 percent to 74 percent of respondents.
Figure 61 – Big data search
0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
Elasticsearch
Apache Solr
Cloudera Search
Big Data Search

77
Big Data Analytics / Machine-Learning Technologies
We asked respondents to rank their interest in a variety of big data analytics and
machine-learning technologies (fig. 62). The leader, Spark MLib (here and throughout
this category), is considered, at minimum, ―important‖ by more than 60 percent of
respondents and ranks well ahead of all competitors. As we see in the following figure,
this is a stark improvement over the previous year. Still, Spark MLib is considered
"critical" to just 15 percent of respondents, reflecting an early-stage market response to
machine learning.
Figure 62 – Big data analytics / machine learning
0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
Spark MLib
Rhipe (R)
Mahout
Oryx
Myrrix
Big Data Analytics / Machine Learning

78
Big Data Distributions
We asked respondents to rank the most important big data distributions by order of
importance (fig. 63). Cloudera leads in measures of "critical" and is the strongest overall
performer, followed by Hortonworks, Amazon, and MAP/R. Cloudera, Hortonworks, and
MAP/R are all seen as, at minimum, "important" to 63-68 percent of respondents.
Figure 63 – Big data distributions
0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
Cloudera
Hortonworks
Amazon
MAP/R
Big Data Distributions

79
Vendor
Ratings

80
IoT Intelligence®
Vendor Ratings
In determining vendor ratings for IoT Intelligence (fig. 64), we include scores for
location intelligence, end user data preparation, cloud BI, advanced and predictive
analytics, and big data analytics—all key capabilities for business intelligence in an IoT
context. Vendors having at least four of five feature sets were considered for inclusion.
Top vendors include Domo (1st
), Information Builders (2nd
), Pentaho (3rd
), SAP (4th
), and
TIBCO (5th
).
Figure 64 – IoT Intelligence vendor ratings
0
5
10
15
20
25
30
35
40
Domo
Information Builders
Pentaho
SAP
TIBCO Software
Qlik
Microsoft
Tableau
RapidMiner
Logi AnalyticsSalesforce
GoodData
Looker
OpenText
Jedox
Sisense
Pyramid Analytics
MicroStrategy
Birst
IoT Intelligence® Vendor Ratings
Advanced and Predictive Analytics Cloud BI
Location Intelligence Big Data Analytics
End User Data Preparation

81
Other Dresner Advisory Services Research Reports
- ―Flagship‖ Wisdom of Crowds®
Business Intelligence Market Study
- Advanced and Predictive Analytics
- Big Data Analytics
- Business Intelligence Competency Center
- Cloud Computing and Business Intelligence
- Collective Insights®
- Embedded Business Intelligence
- End User Data Preparation
- ―Flagship‖ Wisdom of Crowds®
Enterprise Planning market Study
- Location Intelligence
- Small and Mid-Sized Enterprise BI

IoT intelligence: Attitudes towards big data and advanced analytics

Recommended

Recommended

More Related Content

What's hot

What's hot (18)

Similar to IoT intelligence: Attitudes towards big data and advanced analytics

Similar to IoT intelligence: Attitudes towards big data and advanced analytics (20)

More from Abhishek Sood

More from Abhishek Sood (20)

Recently uploaded

Recently uploaded (20)

IoT intelligence: Attitudes towards big data and advanced analytics