Internet of Things: Visualise The Impact
Internet of Things: Visualise The Impact
Internet of Things: Visualise The Impact
Things
Visualise the Impact
Table of contents
3 5 8 10
13 16 19 21
24 26 29 32
Executive
Summary
Governments and businesses are
expecting the Internet of Things (IoT)
to hit mainstream by 2020, which
suggests early adopters have already
acquired some experience from
deployments. While there have been
numerous studies exploring potential
and intention, we were interested in
lessons learned. The pace of change
in an increasingly digital economy
is forcing faster adoption cycles,
and embracing learnings from early
adopters is a critical enabler.
This study targeted organisations that recognised they were well on their way to integrating IoT in their operations. We spoke to 75 teams
who could draw from their experience of recent deployment or a sizeable proof-of-concept. These expert interviews were framed to draw
out how deployments were scoped and sponsored, resourced and delivered. Across all industries, we found four key lessons.
Scope is almost always under-estimated Data and governance skills are critical
The immediate opportunity for cost savings or revenue uplifts drive Although the immediate task of IoT is to connect the unconnected, the
specific IoT deployments. But as data streams through, potential value is realised only when data is analysed and consumed. The path
impact on other workflows attract new stakeholders. The definition from sensor to decision is fraught with data quality, data privacy, data
of success is under pressure as new needs are inserted. This is an management, analysis and distribution considerations. Data managers
inevitable trade-off between speed and distance. Project teams are being included in the initial project team by more experienced IoT
need to manage communications and stakeholder expectations, and programme offices.
aggressively introduce phasing.
TOC 4
IoT means
many things
Survey respondents were asked
what the Internet of Things (IoT) was
expected to mean for their customers,
business and competitors. Given its
wide applicability, responses fell into
17 different categories, which confirms
the view that the IoT has a wide variety
of meanings, and that its impact will
vary across different companies and
industry sectors.
5
A shared view of future benefits
However, there was quite a lot of consensus on the top
two results of introducing IoT technology. Over 40% of
respondents said that they thought that the IoT would result
in improved operational efficiency, and more than one third
(around 36%) that it would improve user experience. Whether
this is already the experience of these companies, or currently
represents their hopes and ambitions for the use of the
technology remains to be seen. It seems most likely that it is a
mixture of the two, given that the technology is still relatively
new and untried in many sectors. The shared scope of vision,
however, is remarkably consistent.
TOC 6
Improvements shared by customers and organisations
The suggestion that the same improvements may benefit both organisations, by improving operational processes, and customers, by
affecting user experience, is supported by the comments from several interviewees. One commented, for example, that it would give the
company the ability to improve either the way that they provided existing services, or deliver new services that were not possible before.
Combining services might also be easier.
It is also clear from the responses that the categories may not be clear cut. Better operational efficiency and resource management, for
example, is likely to result in improvements to user experience, and better value for money for consumers. One noted, that both clients and
end users would benefit from better service, and that reduced costs would benefit both company and clients.
The interaction between operational processes and customer experience was seen as key. Improvements in managing data would result in
better insights into customer behaviour, enabling provision of services that better met customer needs, at a better price. One respondent
noted that there was huge benefit to both business and consumer if organisations could build a more holistic view of customers.
TOC 7
An equally
wide range
of sponsors
Pursuing the connected
consumer
We found that there were, in
total, 15 classes of use case, with
the percentage of organisations
deploying them ranging from 20%
down to around 4%. The majority of
use case classes were in operation in
between 5% and 15% of companies
surveyed.
8
The most popular use case category, which is being deployed by 20% of those surveyed, is projects to support what we might call the connected
consumer. Close behind, and being used by between 15% and 20% of organisations, are auto-diagnostics on 17% and asset tracking on 16%. Those three
classes of use case lead the way, and are the only ones being used by more than 15%. There are, however, a further five classes of use cases which are being
deployed by between 10 and 15% of respondents. The first three, being used by about 14%, are healthcare, connected vehicles, and supply chain control.
Smart factory technology and other transport applications are both being used by about 11%.
It is clear from this that IoT technology is being deployed across a wide range of fields, reflecting its wide applicability and potential to improve both
processes and outcomes. It is clearly being used to improve outcomes for consumers in some cases, for example, via projects classified as the connected
consumer in the retail sector and in healthcare. It is also being used to improve processes, such as auto-diagnostics, asset tracking and connected transport
and vehicles.
Digging down into the detail of the responses to the survey, each category Connected customer
or class contains a wide range of use cases. In healthcare, for instance, Autodiagnostic
some organisations are working on wearables and sensors to monitor Asset tracking
health remotely. Others are working to improve operational processes and Connected vehicles
equipment maintenance by tagging hospital linen. Healthcare
Asset tracking and auto-diagnostics, which are often closely linked, both had Supply chain control
a variety of use cases. Some respondents provided examples of more than Smart factory
one IoT use case, depending on customer needs, whether those were real- Transport
time monitoring or predictive maintenance. This serves to emphasise the
Emergency & surveillance
wide range of use cases. There is, of course, considerable overlap between
Smart cities
use cases, and it can often be hard to allocate one example to a single use-
Smart communication
case category.
Building management
One notable exception to the general rule was a company using IoT
Energy management
technology to track people, not inanimate assets, to ensure their safety in
Smart retail
high-risk working environments. This has a wide variety of possible uses,
Smart agriculture
including manufacturing, warehouse and distribution and construction
sectors. Use cases clearly need to be both flexible and additive, and 0% 5% 10% 15% 20% 25%
organisations need to be open to alternative uses of the technology. Source: SAS IoT Impact Study, September 2016. N = 75. Multiple responses allowed. Q2: What are the
top three initiatives your organisation has embarked in the past 12 months to profit from IoT?
TOC 9
Design
thinking is
critical
What have organisations learnt
from their initial implementation
of IoT technology that will help
them in future? It is fair to say that
different companies had learnt quite
different lessons, with little consensus
on the most important category
among the responses. The range
of responses may be a reflection
of the great diversity among the
organisations represented, but there
are unmistakable commonality related
to soft skills associated with culture
and leadership, such as collaboration,
education and changes to various
approaches.
10
The importance of classical change
management skills
The three most popular lessons, each mentioned by more
than one fifth of respondents, will come as no surprise to
any student of classic change management theory. First was
the importance of applying design thinking. This is where
users, or user experience, is allowed to drive design, bringing
together behavioural theory with technological expertise. It is
a very human-centric process, emphasising the importance of
customers in developing technology, and was mentioned by
almost 30% of respondents.
TOC 11
Design thinking imperative
Respondents comments supported the importance of putting users at the heart of the change process: the design thinking approach. One respondent
crisply articulated how it was all about moving to a more proactive approach that would enable them to anticipate customer needs. This, in turn, would
reduce downtime, and improve operational efficiency. The key, though, was simple and e asy-to-use solutions that would be straightforward for customers to
understand and use.
In many cases, piloting was also a case of putting users first. Respondents noted that initiatives that did not resonate with users could and should be
abandoned quickly as part of the piloting process. This was important for both goals that were over-ambitious, and also prototypes that just did not engage
people.
Other respondents recognised that the real issue may not be introducing a new technology, but change, pointing out that the move to real-time analytics
and decision-making might be a big opportunity, but it was also a big change. Another said that speed of movement was never fast enough, a sentiment
that will be familiar to change managers throughout the world, and across all business sectors.
25%
20%
15%
10%
5%
0%
Apply design Scaling up IoT Lab & Advanced Education Cross organisation Third party Agile/Scrum Advanced data
thinking initiatives & develop testing analytics collaboration collaboration management mangement
new opportunities
Source: SAS IoT Impact Study, September 2016. N = 75. Multiple responses allowed. Q3: What were the main lessons you learnt from these initiatives that will guide how you plan future IoT ventures?
TOC 12
Challenges
span both
technology
and change
There are, of course, challenges
associated with being able to exploit
IoT technology fully. And like the
wide range of lessons learned, theres
a long list of challenges without
strong preponderance for any one
of them. In this case, the study
identified 16 distinct categories of
challenge, each reported by at least
5% of respondents, but none were
mentioned by more than one quarter
of those surveyed.
13
Data handling issues were identified
as the biggest challenges
The top three challenges were each identified by at
least one fifth of the organisations surveyed. The two
biggest issues, each discussed by 21% of those surveyed,
both related to data handling: they were real-time data
analytics, and security concerns. Both of these emerge
at least in part from the sheer quantity of data that is
available from IoT technology, and they reflect the very
real issues involved in extracting value from the data.
Data security has been a big issue for many years, with
few months passing without at least one major security
breach or leak involving a big company or organisation,
and many people are anxious about it.
TOC 14
Comments from respondents reflect the overlaps between categories
The comments from those surveyed clearly demonstrate that the categories utilised in the study are not exclusive, and also are not easily separated. One
noted, for example, that the challenge was to change the culture around information sharing and data security: a combination of three separate categories.
Another noted that the introduction of new technology required time for people to adapt, a classic change management issue.
Several respondents commented on the importance of analytics, and particularly the challenge of using the data well, and the issues of combining data
to generate new insights, or selecting which data to use. The real concern was how to get useful information out of the data, and particularly insights that
would affect outcomes for consumers or customers. One respondent in particular commented on the challenge of analysing massive amounts of data
in real-time, but emphasised the opportunities that would come from new data visualisation tools. One issue that cut across all the challenges was time.
Several respondents commented that implementation had taken much longer than expected. This, too, will be a familiar issue for anyone who has even
been responsible for managing a change programme.
Source: SAS IoT Impact Study, September 2016. N = 75. Multiple responses allowed. Q4: What were the challenges associated with exploiting the IoT fully?
TOC 15
Emerging
skills
Another objective of the study was to
understand what was needed to get
best value out of the IoT, so we asked
organisations about the skills and
resources that they had found most
useful in their IoT projects. The results
may be surprising, particularly for IT
specialists.
16
A surprising leader in the skills stakes
If you thought that the most important skill would be analytics, or data science, then you are in for disappointment. It seems that many of the early movers
have simply bypassed a need to acquire technology expertise by working with consultants. And while buying in skills is seen as a good way to manage
the skills gap, someone needs to be able to manage those providing the bought-in skills. The skill that was identified as crucial by the biggest group of
respondents was the ability to work collaboratively with external consultants.
Beyond the first identified skill, the next three most important skills identified by 12% or 13% of respondents in each case, all relate to technology, 13%
mentioned process automation. The next is engineering skills to address sensor performance and data reliability. The third is about adapting people tasks
to the new technology, and therefore relates both to technology and to cooperation. Below that, and important to 9% of respondents were both data
science and security expertise, much as we would have predicted.
Perhaps surprisingly, given the broad sweep of change management challenges identified, project management skills were not seen as key by more than
5%, and change management skills were not even mentioned. What about the skills gaps? The main skills gaps, identified by 7% of respondents, were in
data visualisation and storytelling and in interpreting results, although analysis was up there too on 6%.
TOC 17
Collaboration is crucial Fig 5: Skills surprise
The reason change management skills did
not surface in this part of the study could have Collaboration with external consultants
been the way that responses were classified. Process automation
One respondent explained that in their view, Engineering skills for sensor
collaboration and innovation are key, and performance / data reliability
another that facilitating change and providing Adapting people tasks to new technologies
leadership was hard. Both these would count Security expertise
as change management to many people.
Data scientist
Several respondents commented on the Algorithm development
importance of collaborating closely with Better understanding of customer
consultants, and buying in skills that were needs (to improve UX)
Prototype elaboration / POC / AI expertise
unavailable in-house. A willingness to engage
with and acquire new skills was seen as crucial Innovation
for success, with respondents using terms like Machine learning / AI expertise
go shopping, and buy in skills, as well as an
Deep project management
expressed need for support in particular areas.
Customer education about data privacy
In a neat fit with the design focus, several 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
respondents spoke about the importance of
putting customer and user needs at the heart The main skills gap lies in:
of design. This was not just in terms of the
end product but understanding customer 8
TOC 18
External
specialists
are an
important
resource
Closing the skills gap
And how are organisations closing
the skills gaps identified, especially
in data and analytics? The prevalent
response in this study is unlikely to
surprise anyone who has read the
previous section. The most useful skill
in exploiting the IoT was seen as an
ability to manage consultants; many
organisations are closing the skills
gap by hiring external consultants,
especially those with an expertise in
technology.
19
Borrow, buy or build? The thinking behind these choices
By far and away the most common response to the problem of The reasons for these choices make interesting reading. One
a skills shortage around data and analytics, indicated by 38% of respondent, for instance, was upfront about their inability to attract
respondents, was to form a partnership with a technology vendor. After or retain high quality staff with the necessary skills. Instead, they had
all, technology vendors have both the skills and the technology. They decided to engage effectively with external consultants. Another noted
can also be employed to train existing staff for the necessary skills to that a technology vendor could provide both the skills and the software,
manage the technology on an ongoing basis. and that having a single organisation providing both simplified the
issues involved in implementation.
14% of respondents mentioned both the classic borrow, buy or build
triumvirate, where consultancy is borrow. Buy was represented Several respondents already had an eye to the future in terms
by external hiring of skilled expertise, often on a project basis, and of growing skills. One, for example, was upfront about requiring
build by training existing employees through a variety of means. technology partners to engage in knowledge transfer, as part of the
Those mentioned included placements, workshops, online training, agreement. Others were using a hybrid approach. In the short term,
conferences, and certification. they were engaging skilled consultants, drawing on their technology
vendor. But in the longer term, they were developing a pool of
Other responses were generally a variation on these themes. Around
freelance analysts to help as and when necessary. Others were relying
10% said that they were building or accessing a trusted community of
on their staff to be willing to learn new skills. The borrow, buy or build
analysts or specialists. Again, this is effectively consultancy, but with a
approach turns out to be more nuanced than we might suspect.
group or individual other than a technology vendor. A few respondents
(around 9%) said that they were trying to understand the skills gap first,
and others that they were hiring aggressively among graduates.
Fig 6: Gap reduction strategies. Source: SAS IoT Impact Study, September 2016. N = 75. Multiple responses allowed. Q6: How are you filling these skills gaps?
TOC 20
Vision of IoT
impact
Maturity of business
functions supporting IoT
Drilling further down into the detail,
respondents were asked which
business functions were most mature
in their use of IoT technology. A
surprisingly large proportion, 20%,
said that maturity was the same across
all departments, although this may
reflect low maturing and very early
stage IoT deployment, rather than a
high level of confidence.
21
The customer priority
Priority for customers was between 5% and 10% for each
of the following areas; logistics and operations, sales and
marketing, customer service and user experience,
a dedicated IoT unit, manufacturing and product
development and the IT department. The low level of
mentions of a dedicated IoT unit may reflect the low
numbers of such units currently in existence rather than
any lack of maturity in these teams.
TOC 22
Expertise spreading across Fig 7: Breaking organisational silos
organisations
Across all departments / organisation
It is clear that a customer focus is what
is driving IoT development across many Operations / field / logistic / store
organisations. One respondent commented
that the technology was being used to assist Sales / marketing
customers, and was therefore by definition
in operations or customer service. This Customer service / user experience
customer focus is, in many cases, what has
Specific IoT / device business unit / lab
driven the whole organisation approach to
IoT use: departments have worked together Manufacturing / product development
to improve services for customers. This was
echoed by another respondent, who noted HR
that one of the challenges of IoT deployment
was that it spanned traditional silos. This IT / network
meant that whole organisation maturity was
Planning / finance
the only option.
TOC 23
Perceived
risks of not
engaging
As well as the challenges associated
with the exploiting IoT technology, it is
always helpful to know what is driving
change. The study asked about the
risks that were seen as associated with
not implementing IoT technology.
24
The top three risks were all Experiences shared across public and private sector
identified by the same proportion
of respondents, 18%, which related Fear of falling behind was not a fear confined only to the private sector. Public sector respondents,
to concerns about losing market including those representing smart cities, also expressed concern about being overtaken. While their
share to competitors, losing concern was perhaps more reputational than financial, a failing smart city will be less attractive to new
efficiency, and not being able to business and risks lower growth, with a very real economic impact.
provide a cost-effective service or Several respondents viewed the rise of the IoT as inevitable. As a result, one suggested that it was better
product to consumers. The first of to be the disruptor than the disrupted. Another commented that even market leaders cannot afford
these is probably inevitable: most to sit back, saying that to stay ahead, it is necessary to continue to innovate and disrupt. Phrases and
organisations rightly fear falling sentiments that kept recurring included the desire to stay ahead, the need to be among the first movers,
behind their competitors if they and the importance of constantly improving the customer experience.
miss an opportunity that others can
Others identified risks as the inverse of what they were already gaining. In other words, they were now
exploit. The second is also not too
seeing that the IoT enabled them to track goods, improve customer service through better visibility,
surprising given the high proportion
and reduce environmental impact, for example. Not using it would mean that they would lose all these
that identified the IoT as a way of
benefitsand, potentially, lose ground to their competitors.
improving efficiency.
than incremental improvements Not being able to reduce risk of running business
by other incumbents, because it is
much harder to predict or counter. Lower understanding of customer / pro-sumer needs
New technology always holds this
Losing momentum in the adoption curve
fear, although of course it is also
0 2 4 6 8 10 12 14 16 18 20
an opportunity for those who can
Source: SAS IoT Impact Study, September 2016. N = 75. Multiple responses allowed. Q8: What do you see as the biggest risks associated with not fully
exploit it. exploiting Internet of things opportunities?
TOC 25
Speed
Respondents were asked how they
assess if their appetite and initiatives
around IoT are moving as quickly
as the market demands. It is always
difficult to assess whether you are
moving quickly enough, especially in
a new area where the technology is
moving at such speed anyway. Our
respondents, however, showed that
they were pragmatic.
26
Practical efforts to stay ahead of the
curve
The focus of many of our respondents was simply
to keep moving. 15% said that they simply kept
developing innovative or disruptive solutions, in the
hope that they would remain in the lead. Nearly as
many, around 13%, said that they kept investing and
focused on expanding their use of the IoT technology.
This is a very pragmatic approach to a fast-moving area:
if it moves, you move with it.
TOC 27
A focus on customer demand and needs
We learned a lot about the companies approach to innovation from the detailed answers for the study. One respondent noted that innovation was a key
purpose of the company, and their partners. It was, effectively part of their DNA, and they were focused on pushing the boundaries and developing further.
Another respondent noted that customer demand was important. It was, they suggested, possible to ensure that you were keeping up by responding
to customer requests. This is, of course, only likely to work for incremental improvements following initial implementation, and for not a major disruptive
change such as initial adoption of IoT technology. Another respondent, however, pointed out that it was important to ensure that users came with you on
the journey: innovating for innovations sake would not help.
Other companies were focusing on rolling out their existing technology to new geographies or new users, expanding into more sectors as they went.
They were also adding new modifications, or identifying new uses of data. This was a particularly common use of an innovation lab or dedicated product
development team, and shows continuous incremental innovation at work.
Fig 9: Benchmarking
16
14
12
10
0
Developing Keep on Reviewing Conducting Conducting Talking to Checking the Promoting Others
innovating/disruptive investing/focus company/group research through market analysis analysts/consultant regulatory collaboration
IoT solutions on expansion IoT strategy internal resouces through a /advisors environment for across business
third party new product/service unit
offerings
Source: SAS IoT Impact Study, September 2016. N = 75. Multiple responses allowed. Q9: How do you assess if your appetite and initiatives around Internet of Things is moving as quickly as the market demands?
TOC 28
Changes
in the
addressable
market
Finally, respondents were asked how
they saw their addressable market
changing over the next five years.
While some might characterise this
as crystal ball gazing, it is clear that
many companies had thought long
and hard about this.
29
Incremental achievements and improvements
It will probably not come as a surprise to learn that most companies already active in using IoT technology saw the way forward as being through
incremental improvements. The top three response categories, each given by more than 15% of those surveyed, were all about improvements and
expansions, rather than big changes. The most popular of these, from 18% of respondents, was to scale up or extend current IoT projects. This is consistent
with the idea that most IoT projects start small, rather than going for a big bang.
Improved customer services and other offerings, and an enhanced digital offering, were both seen as the way forward by 17% of respondents. The reality is
that many companies see incremental improvements as an important way to expand, fitting with the requirement to take both staff and customers along on
the forward journey.
Fewer respondents, only 11%, saw the expansion of the connected customer being key, or that IoT would become a key industry focus in their sector.
Increased data traffic and use of analytics were mentioned, but again, not by many. It is, of course, always harder to predict disruptive innovations, or see
where the market is going in bigger ways. Who, after all, predicted the rise of the smartphone twenty years ago? Not many people, that is for sure.
TOC 30
More focus on customers
But whether or not respondents were predicting major
changes in their sector, there was a consistent core
message: customer demand was central. Respondents
talked about the rise of partnerships with customers,
and how the IoT would allow them to improve services
for customers.
TOC 31
Appendix
75
to identify organisations with such Healthcare & Life Sciences 12%
experiences, to better understand
Engineering 9%
challenges and opportunities from
the deployment process. We were Facilities & Utilities 9%
Respondents
specifically interested in the data Other Service Industries 10%
and analytics aspects. Our experts
interviewed 75 large organisations Public Sector 8%
during the summer of 2016, and Construction & Real Estate 8%
together with secondary research,
this report represents the first of our
analysis.
TOC 32
Appendix
SAS Contributors
Anand Chitale Caroline Hermon Federico Alberto Pozzi Rene van der Laan
http://blogs.sas.com/content/ www.blogs.sas.com/content/ www.blogs.sas.com/content/ www.blogs.sas.com/content/
sascom/author/anandchitale/ blog/author/carolinehermon/ blog/author/federicopozzi/ blog/author/renelaan/
TOC 33
SAS INSTITUTE INC. 100 SAS CAMPUS DRIVE CARY NC 27513-2414 USA
1-919-677-8000 www.sas.com
SAS and all other SAS Institute Inc. product or service names are registered trademarks or trademarks of SAS Institute
Inc. in the USA and other countries. indicates USA registration. Other brand and product names are trademarks of
their respective companies. Copyright 2016, SAS Institute Inc. All rights reserved. 1463549 UK0816