PM – Organisational Performance

Contents
Management Reports ................................................................................................................. 3
WHAT ARE MANAGEMENT REPORTS? .................................................................................... 3
CONTROLS OVER MANAGEMENT REPORTS ............................................................................ 3
SENSITIVE INFORMATION ........................................................................................................ 4
Sources of Management Information ......................................................................................... 5
USING MANAGEMENT INFORMATION AS A CONTROL ........................................................... 5
THE COSTS OF MANAGEMENT INFORMATION ....................................................................... 6
LIMITATIONS OF USING EXTERNALLY GENERATED INFORMATION ........................................ 7
Management Information Systems & Reports ........................................................................... 8
DEFINITION: ............................................................................................................................. 8
TYPES OF PLANNING: ............................................................................................................... 8
TYPES OF MANAGEMENT SYSTEMS: ....................................................................................... 9
SOURCES OF MANAGEMENT ACCOUNTING INFORMATION: ............................................... 11
INFORMATION SYSTEMS CONTROLS: .................................................................................... 12
COST VERSUS BENEFIT ........................................................................................................... 12
INTERNET, INTRANET, WIRELESS TECHNOLOGY AND NETWORKS ....................................... 13
Big Data and Data Analytics ...................................................................................................... 15
BIG DATA................................................................................................................................ 15
DATA ANALYTICS ................................................................................................................... 20
SUMMARY.............................................................................................................................. 22
Data Visualisation ......................................................................................................................... 23
What is Data Visualisation......................................................................................................... 23
Importance of data Visualisation .............................................................................................. 23
Examples of data Visualisation.................................................................................................. 24
Tools for Data Visualisation....................................................................................................... 27

1
2
MANAGEMENT REPORTS
WHAT ARE MANAGEMENT REPORTS?

They are internally produced reports detailing important company information such as the
profit and loss or costs etc. They can cover any period but are usually over one or four months,
depending on the company.

These reports are used for internal decision-making. Management reports should be
communicated through the appropriate channels, so, for example, don't make a report out of
what can be sent quickly and effectively in an email.

Management reports also contain good-quality information. These reports must:

− Be easy to use – if it's not easy to use, management may not bother looking at it.
− The report must be understood and be relevant to its users.
− The report must be complete – a half-finished report is not much use,
− The value of the information in the report must be more than the cost it took to produce,
be this the time, cost or resources invested in it.
− The report must be given in time to enable decision-makers to review it and make
decisions.

CONTROLS OVER MANAGEMENT REPORTS

There must be controls over the production of management reports. If not, time and money
could be wasted, and sensitive information could be leaked. Before a report is done:

− A cost/benefit analysis should be done. Do the benefits of having this information outweigh
the costs of putting this report together?
− The organisation's reports' formatting should be standardised and consistent. This saves
time you could have spent deciding on fonts, etc. It also ensures consistency so that the
report is less likely to be misinterpreted.
− The users' requirements must be precise and understood. If they are not, the report writing
team must verify them before the report is started. When the report is finished, it should be
rechecked against it.
− The report should clearly outline its recommendations and actions that need to be taken by
management. A report full of information is good but not helpful if it doesn't give me a few
indications of what I should do based on its findings.
− The writers must be identified – transparency and accountability will result in more
accurate information.

3
If a company writes a report, it uses its valuable resources and staff time to do that instead of
something else that could bring in the company revenue. Therefore, reports should only be
written if they will give new information.

SENSITIVE INFORMATION

The information in the managers' reports is often confidential to external and internal users.
For example, management would not want reports detailing new product designs being
accessed by competitors so they can be copied. Similarly, management may not want staff to
know of staffing issues or restructuring plans so as not to cause panic. What controls should be
put in place for sensitive information and to protect it?

− Information should only be given to the relevant parties, nobody else.

− Employees should sign confidentiality agreements preventing them from sharing sensitive
information with third parties.
− Secure all sensitive computer files with a secret password, not one that everyone knows.
And that password should be kept securely, not on your desk's post-it!
− Hold all sensitive reports in a safe place that can be locked, like a small room.
− The companies’ information systems should be protected with firewalls from people getting
in from the outside.
− As well as firewalls, encryption should be used to code files so hackers cannot read them.
Anti-virus and anti-spyware software should be installed to protect the company and its
information.
− Under personal data protection acts, companies must protect personal data and not give it
to third parties. Therefore personal data must be kept confidential.

The company should use sophisticated encryption techniques to secure confidential data.
Similarly, firewalls and VPNs (virtual private networks) can also ensure security and
confidentiality.

4
SOURCES OF MANAGEMENT INFORMATION
Companies need good-quality information to make decisions. This information is both internal
and external to the organisation. The key sources of management accounting information are:

− The company’s accounting records: The books of prime entry, such as the sales day book,
purchase day book and cash book.

− Payroll records: Wages, costs and quantity of people and time required to do a task.

− Departmental records of incidents, e.g. the accident log.

− Information received from customers, e.g. customer comments or complaints.

These records give management more than financial information. They give non-financial
information too. For example, quantities sold and their combinations provide valuable
information for the marketing department.

The management must also look at information outside the organisation for decision-making
because the company does not operate in a vacuum. Sources of external information are:

− The government: They provide publications on new legislation, taxation policies, etc.;

− The media: Information about changes in the organisation's industry, technological

changes, etc.;

− Financial institutions: They provide information on potential customers and foreign

currencies;

− Competitors' financial statements;

− The Internet; and

− Public databases.

Keeping up with new information from these sources will help the company make better
decisions.

USING MANAGEMENT INFORMATION AS A CONTROL

Management information, as well as being important for decision-making, is a powerful control

tool.

Both Internal and external sources are used. This information is compared to the budgeted or
forecasted results to find the variance and reasons for the variance.

5
Internal information, such as cost reports, can be monitored to measure the variances from the
budgeted or forecasted results.

External information is used to benchmark the company against its competitors.

The control process using internal and external information is as follows:

1. Management reviewed the previous months' sales volumes. This is internal information,
and they decided to gain 3% more market share. Market share information was obtained
from external sources.
2. The management now decides how this goal is to be achieved.
3. The following month's sales figures are examined to see if the target has been achieved.
This is called feedback. This is internal information.
4. The feedback demonstrates if the company is on target for achieving its goals. Depending
on the results, management will go back and amend their original controls to improve on
results.

This is sometimes called the feedback loop.

THE COSTS OF MANAGEMENT INFORMATION

These are the main costs associated with gathering and analysing management information:

1. Direct data capture costs: Direct data capture is a data input where there is no data entry
and where data is scanned into a data management system. For example, a till in a shop
that scans barcodes telling the system what has been bought. The costs associated with
direct data capture are:

a. The initial setup costs of installing the system; and

b. Linking up the information to the system.

2. Processing costs: Processing Costs are from processing and analysing the information.

a. The associated costs are mainly staff labour.

b. Data collected is only valuable when analysed, and time must be invested to analyse
thoroughly.

3. Indirect costs of producing information:

a. Information overload creates many indirect costs by:

i. Staff get so lost in the detail that they spend too much time analysing information,
thereby wasting time; and

6
 ii. Poor decisions that cost the company can be made from information overload.

b. Not looking at future trends: The company can become so focused on the past that it
fails to look to the future—for example, technology and legal changes.

LIMITATIONS OF USING EXTERNALLY GENERATED INFORMATION

− Quality: Information produced externally may not be accurate or reliable.

− The organisation's information needs may not be met: External information was not
created for their purpose.

− Difficulties in sourcing the information: For example, detailed breakdowns of competitors'

sales will be challenging because companies tend to keep this information confidential.

7
MANAGEMENT INFORMATION SYSTEMS & REPORTS
DEFINITION:

Management accounting and information systems are integral to producing the information
that managers use for performance measurement and management. These systems will
provide the information which enables performance measurement to take place.

TYPES OF PLANNING:

Planning, control, and decision-making can be classified into a three-tier hierarchy:

1) Strategic. Strategic planning is deciding on the business's longer-term objectives and the
high-level policies surrounding them. Management accounting information needed to
support strategic planning will be forward-looking, usually for several years, and have an
external orientation. Examples would be:

− Product profitability;
− Financial effects of competitor responses;
− The effect of acquisitions and mergers.

2) Tactical. Tactical planning focuses on the most efficient and effective use of resources to
support and achieve long-term strategic plans. Management accounting information
needed to support tactical planning decisions will have a much shorter time horizon, as
there will be much more precision with a much narrower focus on information. Most of the
information will be generated from within the organisation. Examples would include:

8
 − Annual budgets for sales and production;
− The levels of inventory to be held;
− Marketing and advertising campaigns.

3) Operational. Operational decisions focus on day-to-day tasks, ensuring maximum efficiency

and effectiveness. They are made to support tactical decisions and control. Management
accounting information here will be very detailed, have a very narrow focus, and have a
concise time frame. Examples would include:

− Day-to-day transaction data;

− Current inventory levels;
− Scheduling incredible work.

TYPES OF MANAGEMENT SYSTEMS:

Several types of management info systems can provide data to an organisation:

1) Transaction Processing System (TPS). A TPS system can collect, store, modify, and
retrieve large volumes of an organisation's data. The key characteristics of a TPS system
are:

− Rapid response. If fast performance is vital to a business, the input needs to

become the output in a matter of seconds;

− Reliability. If potential failure could stop business, then a backup and recovery
process must be in place;

− Inflexibility. A TPS system follows a standard process route. It is not able to adapt
or have any flexible response to the input it receives;

− Processing. A TPS can process data in batches or in real-time. With batch

processing, the data is collected throughout a designated period and processed
later, resulting in a time delay. Real-time processing is the immediate processing
of data.

2) Management Information System (MIS). MIS system can take data from the TPS system
and convert it into summary or exception reports for decision-making. An MIS system
generally doesn't have the analytical capability. MIS is usually used at the operational
and tactical levels.

3) Executive Information System (EIS). An EIS system will typically draw critical strategic
information data from an internal MIS system and allow communication with external

9
 sources of information (e.g., data from competitors, legislation, market research, and
information databases). An EIS will typically involve data analysis and modelling tools
and perform "what if" analysis to aid strategic decision-making.

4) Enterprise Resource Planning System (ERP). An ERP system is modular and will allow
integration between the fundamental processes of an organisation. It will typically
support sales and order processing, procurement, production, distribution, customer
service, human resources, and finance activities. There are many benefits from a fully
integrated ERP system, including removing inefficiencies and duplicated data, with
significant savings in time and effort.

A system can be open or closed:

a) A closed system is isolated from the external environment, and data will not be
provided to or received from the environment;

b) With an open system, a business can focus on the external factors critical to the
organisation's success and adapt to the changing environment as necessary.

Other management systems

− Knowledge Management Systems are information systems that can store and retrieve
knowledge for understanding and collaboration. Such systems can exist at numerous levels
and units. For example, such systems can exist at the entity, division, team, etc. Also, note
that a knowledge management system can be integrated for different units and levels. This
is called an enterprise-wide knowledge management system.
− A Customer Relationship Management System (CRM) is a system that stores and analyses
data relating to customer relations. These systems can be operational, analytical and
collaborative. These systems manage customer interactions and relations (for example,
keeping a record of data needed to contract customers, computing the churn rate, etc.).
− A Supplier Relationship Management System (SRM) is a system that stores data relating to
the suppliers of company. It helps give insights relating to the company's suppliers that can
help achieve cost savings and synergies.
− A Decision Support System (DSS) aims to support decision-making for semi-structured
problems (problems that still require the application of judgement).
Note: Formal decision support systems have the advantage of having been developed using
best practice development methods, whereas spreadsheets written by the user tend to be
full of errors.
− The expert system seeks to replace some human expertise to a degree. Such systems have
the following advantages:
• They are available instantly, 24 hours a day and seven days a week;

10
 • They make consistent decisions;
• They are less expensive than employing human experts.
• The disadvantages of such systems are:
• The installation and maintenance costs money;
• They are inflexible.
− A data warehouse is a data management system that supports business intelligence (BI).
This is also known as an enterprise data warehouse, and its primary purpose is to store and
analyse data. The company's actual data is 'dumped' or stored in the data warehouse from
different sources, where it is easily accessible. Such a warehouse can then be used for
techniques like data mining and data analytics.

SOURCES OF MANAGEMENT ACCOUNTING INFORMATION:

Sources of management accounting information can be both internal and external:

1) Internal information. This will include data captured within the financial accounting
records, such as data from the sales and purchase ledgers or payroll data. A significant
amount of this data will be extracted from company systems, and there could also be
informal communications between management and staff, interviews, or maybe
documented meetings.

2) External information. This information tends to be more relevant to strategic and

tactical decision-making than operational decision-making. External data can be either
primary or secondary:

Primary Data Secondary Data

− Tailored to a company's exact − Not collected by the organisation

needs; itself (e.g., government statistics,
internet databases, data from
− 100% relevant to the company;
trade journals);
− More expensive than secondary
− Cheaper than primary data;
data.
− It may not be relevant.

The costs of collection, processing and production of data can be divided into three categories:

a) Data capture costs;

11
 b) Process costs;

c) Other indirect costs.

INFORMATION SYSTEMS CONTROLS:

It is essential to consider the necessary controls over the information:

1) Procedures manual. It will set out the following controls for distributing internal data:

− Format of the report;

− Frequency of the report;
− Distribution list;
− Confidentiality;

2) Security system. It helps to regulate which staff members have access to different data
types. Passwords can limit access to certain areas within a system and will provide an
audit trail to establish any unauthorised attempted access;

3) Firewall. It protects the data from external unauthorised access;

4) Anti-virus software. It will scan files looking for known viruses and identify suspicious
behaviour from a computer program that could indicate infection;

5) Physical controls. These controls include door/cupboard locks and safes to prevent
hardware or paper documentation access.

COST VERSUS BENEFIT

As highlighted earlier, it is essential to assess the benefits derived from the information systems
against the costs that are being paid for organising and preparing the information. The benefits
should outweigh the costs. It is also essential to understand that not all benefits or costs will be
financial. The costs may incur, and the benefits are more strategic but non-financial.

Generally, the benefits derived can be of four main types:

1. Cost-reducing or saving benefits – for example, the information system helps us to

identify areas where we can save costs (like changing the suppliers to reduce input
costs).
2. Cost avoidance benefits – for example, the information system helps us identify areas
where we can avoid certain costs (like identifying duplication in processes that can be
avoided).

12
 3. Improved service/product benefits – for example, the information system helps us
identify areas where we can improve the quality of our products or services.
4. Improved information benefits – for example, timely and more accurate reports help
management make better and more informed decisions.

As highlighted above, these costs and benefits can be tangible and intangible (or non-financial).
Similarly, these can be direct or indirect and fixed or variable. To perform an effective cost-
benefit analysis, it is essential to understand the general nature and behaviour of the costs and
benefits.

Similarly, there can be many hidden costs for implementing and maintaining the information
system. If these are overlooked in the cost-benefit analysis, the management might be
overestimating the impact of the information system.

For example, training of the staff, licencing costs, and annual retainership fees are some
examples of more apparent, direct, and tangible costs. However, there can be hidden or
intangible costs like management resistance, inability to comprehend the reporting module
properly, generating incorrect reports, etc., that the management must identify and ascertain
to ensure an adequate cost-benefit analysis.

INTERNET, INTRANET, WIRELESS TECHNOLOGY AND NETWORKS

Internet

In straightforward terms, the internet is the global system of interconnected computer

networks that use the internet protocol suite (TCP/IP) to communicate. Information is easily
shared using the interest, for example, by sending emails or sharing files through the cloud (for
example, Microsoft OneDrive, Google Drive, etc.).

Intranet

Intranets are private networks created or companies using internet networking standards and
web technologies. This restricts their access but, at the same time, increases the integrity and
confidentiality of the information. Usually, these are set up to provide access to data across the
enterprise. An intranet can help create applications running on different computers or mobile
phones throughout the entity. Usually, the enterprise will store its internal database here, and
staff will access contents like online policy manuals, leave records, official calendars and
meetings schedules, etc.

Wireless technology

13
Wireless technology (for example, cloud computing, the internet of things (IoT), and mobile
technology) enables remote working and increased enterprise mobility. This can increase
productivity and reduce costs for the enterprise. However, additional costs are involved in
implementing wireless technology (for example, setting up equipment, maintenance, staff
training etc.). Despite these costs, these technologies are now more of a 'necessity' for modern
businesses. As such, the costs and benefits need to be carefully identified and managed by
businesses to remain competitive. Another critical consideration with wireless technology is the
security and confidentiality of the data. Concepts like firewalls, encryption, and VPN (a virtual
private network) become relevant here.

14
BIG DATA AND DATA ANALYTICS
BIG DATA

In the modern age, we are drowning in data. It is all over the place in a variety of different
sources. Consider, for a moment, your digital footprint.

• You may post what you’re up to today on Facebook, or a photo on Instagram, saying where
you are, what you’re doing, who you’re with, the clothes you’re wearing, etc.
• You may also have searched online for something to buy or download a book.
• You may have listened to some music on Spotify.
• You may even have ‘spoken’ to appliances in your home to turn the heating on.

In addition to all the loyalty cards you use, you tell vendors what your shopping habits are. All
these things leave a 'trail'.

Big data is information such as this scattered around the internet. What use could this be to an
organisation?

Primary sources of big data

In theory, they can be grouped under the headings:

1- Social (human) - this source is becoming more and more relevant to organisations. This
source includes all social media posts, videos posted etc.
2- Machine (sensor) - this data comes from what can be measured by the equipment used.
3- Transactional - this comes from the transactions which the organisation undertakes. This is
perhaps the most traditional of the sources.

Laney’s 5 Vs

A core area of theory about Big Data is Laney's 5 Vs. The 5 V's help us understand necessary
data's advantages and disadvantages to a business. Let's work with an example as we go
through the list. Imagine a business attempting to build an accurate profile of its customers to
help it plan its product range in the future.

1. The first of the 5 Vs is Volume.

There is an awful lot of this data out there. This is an opportunity because there is plenty to
find out. A detailed picture could be built up of our customers to help us plan successfully for
our new product range. However, this is also an issue. With so much data and information,
it's difficult to cope. We need to process all that data, which takes time and money. We need
to be smart in the way we approach this.

15
2. The second of the 5 V's is Velocity.

It changes constantly. You may have posted somewhere in the last hour where you are or
what you are doing. This is an excellent opportunity to build a remarkably current picture of
you. Regular updating means we can understand what you are like now and what you were
like. Trends can be mapped as a result. For example, we might see that you are planning to
go on a walking holiday this year and that, through analysing Facebook trends, lots of your
demographic are taking this up as a new hobby. This could help inform our choices for our
product range.

However, the fact it changes so quickly is also a challenge for us. Keeping pace is complex,
and pictures we pull together of you will quickly be outdated. Our ability to utilise current
information quickly is essential.

3. The third of the 5 V's is Variety.

The information is in an awful lot of different sources and different formats. Photographs,
social media postings, website cookies, GPS tracker information, emails, instant messages,
etc. This is good because the information is a rich palette - we can build up a complex,
multidimensional hologram of you to help us plan more accurately.

However, practically it's also a problem. Trying to name and access the sources and
summarise the data meaningfully is a considerable challenge.

Because not all of that data is readily usable in analytics, it is essential to differentiate
between the three types of big data:

a) Structured data is the easiest to work with. It is highly organised with dimensions
defined by set parameters. Think spreadsheets; every piece of information is grouped
into rows and columns. Specific elements defined by certain variables are easily
discoverable. It's all your quantitative data, for example, age, billing, expenses, credit
card numbers etc.

Structured data follows schemas: essentially road maps to specific data points. These
schemas outline where each datum is and what it means. It is the most accessible data to
analyse because it requires little preparation before processing. A user might need to
cleanse data and pare it down to only relevant points, but it won't need to be interpreted
or converted too deeply before a proper inquiry can be performed.

b) Unstructured data is all our unorganised data. You might be able to figure out why it
constitutes so much of the modern data library. Almost everything you do with your
computer generates unstructured data. No one transcribes their phone calls or assigns

16
 semantic tags to every tweet they send. While structured data saves time in an analytical
process, taking the time and effort to give unstructured data some level of readability is
cumbersome. The hardest part of analysing unstructured data is teaching an application
to understand the information it's extracting. More often than not, this means
translating it into structured data. It almost universally involves a complex algorithm
blending scanning, interpreting and contextualising functions.

c) Semi-structured data toes the line between structured and unstructured. Most of the
time, this translates to unstructured data with metadata attached to it. Let's say you take
a picture of your cat from your phone. It automatically logs the time the picture was
taken, the GPS data at the time of the capture and your device ID. If you're using any web
service for storage, like iCloud, your account info becomes attached to the file. Semi-
structured data has no set schema. This can be both a benefit and a challenge. It can be
more difficult to work with because effort must be put in to tell the application what
each data point means.

The second classification of data is based on its measurement levels. In this respect, there is
a difference between:

a) Categorical data describes categories or groups. One example is car brands like
Mercedes, BMW and Audi — they show different categories.
b) Numerical data, on the other hand, as its name suggests, represents numbers. It is
further divided into two subsets:
i. Discrete data can usually be counted in a finite matter. A good example would be
the number of children you want.
ii. Continuous data is infinite and impossible to count. For instance, your weight can
take on every value in some range.

4. The fourth of the 5 V's is Veracity.

It means accuracy and truthfulness and relates to the quality of the data. The data collected
must be accurate if we want any analysis to provide valuable decision-making findings. How
is this done in practice? To assess how accurate the data collected is, companies must
consider how accurate or reliable a data set might be and how trusted the data source is.
They must be able to trust the source of the data being collected and be confident that the
data is reliable and accurate if they base essential and often costly decisions on the findings
of its analysis.

17
 The difficulty that companies face here is that, by its very nature, the data collected comes
from many different sources. Take, for example, data from social media-they can be easily
manipulated, particularly given the recent increase in so-called 'fake news' and growing
reports of deliberately manipulated customer reviews on retail sites.

5. The final of the 5 V's is the value.

There is little point in going to the effort and expense of gathering and analysing the data if
this does not add value to the company. This is why companies need to consider the
potential of big data analytics and the value it could create if gathered, analysed and used
wisely.

Summary

Each of the 5V's is a considerable challenge. However, the prizes are potentially huge. It can be
a source of competitive advantage. If we can understand our customers better than others, we
stand a good chance of making better decisions as a business.

The challenges can seem overwhelming, but remember that no one said this needs to be
mastered perfectly, as, indeed, it cannot be mastered perfectly. You just need to do a better
job at it than your competitors.

Big Data (DIKW) Pyramid

Another concept related to this topic is the Big data (DIKW) pyramid. It is also known as the
knowledge pyramid and explains the relationships between data, information, knowledge and
wisdom:

18
− A range of raw data could be collected from various sources. For example, a company
collects a range of data from previous purchases, customer questionnaires, and social media
posts.
− The raw data can be analysed to look for trends or patterns. For example, there may be a
link between purchasing a particular product and a particular group of customers. This is
information.
− In the next step, the company analyses information further to establish how the identified
links are connected. Knowing precisely what types of customers buy a particular product or
favour particular product features is knowledge.
− Finally, the knowledge gathered can be used to make informed business decisions. This is
wisdom. For example, the company could adapt the packaging of its products to the tastes
of different customer groups.

Problems with big data

Big data raises difficult ethical questions - some people have a 'big brother' style concern about
businesses building detailed profiles of us - and potentially about legal concerns relating to data
protection. Both of these risks carry associated reputation risks if our use of big data becomes
public knowledge, and this isn't received well. Also, increasing linkages with external data
sources increases the risk of viruses infecting our systems.

So, big data is not without its risks and problems, but it has enormous upside potential.

19
DATA ANALYTICS

Data analytics is concerned with processing data to gain new insights that hopefully be used to
inform decisions in the organisation. It may inform decisions about what we make, how we sell
it, and who we sell it to, for example. This is not a new idea. Even before the days of Big Data,
organisations were exploiting their internal data resources by consolidating them into one place
- a 'data warehouse', and analysing it to their advantage.

Data mining

One such example of an analytical procedure is data mining. A data mine is a piece of software
that burrows into a set of data, notes apparent relationships between data items, and tells you
what it's found. A finding may or may not be helpful. For example, it may say, 'when sales
increase, costs tend to increase' or 'when sales increase, profits tend to increase'. Well, you
may well have known this anyway. However, on occasion, it may just tell you something you
DIDN'T know that you could exploit to your advantage.

Predictive analysis

Another thing to mention is predictive analysis. This type of data mining aims to predict future
events—for example, the chance of someone being persuaded to upgrade a flight.

Example

A classic example of this relates to Walmart in the US. This is such a well-worn example. I'm
unsure it's much more than 'urban legend', but in any case, it is well known and makes the
point well. Walmart owned a chain of 7-11 convenience stores in the US - high street
supermarkets designed for perhaps small purchases as you pass by. The store layouts were
altered during the day, so items shoppers were more likely to want to buy at different times of
day were moved perhaps to the front of the store.

Information was gathered, perhaps through a loyalty card scheme, that recorded what was
purchased, when and by whom. A data mine was ‘let loose’ on this data set and came up with a
surprising correlation.

The correlation it noted was that when young men purchased nappies on a Friday evening, a
high percentage of the time, beer was purchased in the same transaction. The two were then
co-located in the store, increasing sales of both.

In hindsight, you might imagine why the sale of beer and nappies might be correlated in this
way. A young man on a Friday night may historically have gone out with friends after work for a
social night. However, he has a young child at home, so he foregoes his Friday night to look

20
after his child. He may buy a drink of beer to have at home instead. This is all supposition, but
hopefully, it seems plausible.

The point is, it would be an odd question to ask before any analysis - 'I wonder if beer and
nappy sales are related, and if so shall we promote them jointly'. The chances are you probably
wouldn't have thought even to ask the question. This is the beauty of data mining. It tells you
things you didn't even know about yourself. It could be brand-new wisdom that can give you a
competitive edge.

The Walmart example was based on an internal data set. Apply the concept to big data, and the
opportunities (and difficulties, too, admittedly) increase dramatically.

Descriptive statistics and inferential statistics

When analysing data, we can use both descriptive and inferential statistics in our analysis.

1. Descriptive statistics

Descriptive statistics is the term given to the analysis of data that helps describe, show or
summarise data in a meaningful way such that, for example, patterns might emerge from
the data. It does not allow us to make conclusions beyond the data we have analysed or
reach conclusions regarding any hypotheses we might have made. They are simply a way to
describe our data.

For example, if we had the results of 100 pieces of students' coursework, we may be
interested in the overall performance of those students. We would also be interested in the
distribution or spread of the marks.

Descriptive statistics allow us to do this by using a combination of tabulated description

(i.e., tables), graphical description (i.e., graphs and charts) and statistical commentary (i.e., a
discussion of the results).

2. Inferential statistics

Inferential statistics allow us to use these samples to make generalisations about the
populations from which the samples were drawn. It is, therefore, important that the sample
accurately represents the population. The process of achieving this is called sampling.
Inferential statistics arise because sampling naturally incurs sampling error; thus, a sample is
not expected to represent the population perfectly.

21
SUMMARY

− Big data and how it is analysed can help improve the quality of decisions made in the
organisation, enhancing the value added by the business.
− Therefore, the organisational processes relating to big data become a dimension of
competition. If we can deal with it better than our competitors, we will gain an advantage
over them.

22
Data Visualisation
WHAT IS DATA VISUALISATION
Data visualisation means translating or transforming the information into a visual context. For
example, graphs or maps. This makes the data or information easier to understand, analyze,
and make inference.

The main idea here is to assist in understanding and comprehending data. This practice
highlight patterns, trends, and identifies outliers in the data.

Data visualisation is actually a step of the overall data science process, which is explained as
follows:

The importance of data sciences, and therefore data visualisation, has substantially increased
with the advent of technology and big data. As businesses now accumulate high volume, high
variety, and high velocity data (that is, big data), they needed to get this summarized quickly,
this is where data visualisation comes into play. As such, data visualisation is fundamental part
of the advanced data analytics.

IMPORTANCE OF DATA VISUALISATION

Data visualisation provides an efficient and effective way to communicate and share
information. As such, this is now an integral part of the overall decision making process of the
entities. Data visualisation plays a key role in the following:

• Identifying factors that affect customer behavior.

23
 • Identifying areas that need to be improved.
• Identifying areas that need special focus of attention of the management.
• Making data more memorable for all relevant stakeholders.
• Developing strategies and plans.
• Forecasting, budgeting, predicting future outcomes.

Other benefits of data visualisation include:

• Improving the ability to absorb information quickly.

• Improving improve insights and comprehension of data.
• Faster and more effective decision-making process.
• Increased understanding of operations and strategies in place.
• Maintaining the audience’s interest, as information is presented in an appealing fashion.
• Improving the overall distribution and communication of information.
• Making information more accessible.
• Helps in reducing mistakes and achieving success.

EXAMPLES OF DATA VISUALISATION

Following techniques are used to visualise data:

• Line charts
These display how variables can change over time.

Sales
1000
900
800
700
600
500
400
300
200
100
0
Q1 Q2 Q3 Q4

2014 2015 2016

24
• Area charts
This technique is an extension of line chart, where multiple values in a time series are
displayed with equally spaced points in time.

Sales
1400

1200

1000

800

600

400

200

0
Q1 Q2 Q3 Q4

2014 2015 2016

• Scatter plots
This technique displays the relationship between two variables. A scatter plot takes the
form of an x-axis and y-axis and represents data points on a graph.
0.9
0.8
0.7
0.6
Return

0.5
0.4
0.3
0.2
0.1
0
0 2 4 6 8 10 12
Risk

25
• Treemaps
This method shows hierarchical data in a nested format. Here, the size of the rectangles,
representing each category, is proportional to its percentage of the whole.

• Pyramid or funnel
This technique uses a stacked bar graph to display the complex narrative. It is best used
when displaying the distribution of a population.

26
• Bubble clouds
Bubble cloud charts are used for illustrating the relationship between elements in the data
set. For example, you can visualise sales data collected from different quarters, and
represent each year as a bubble whose size is proportional to the value for that quarter’s
sales.

Quarterly sales
1200

1000

800

600

400

200

0
0 1 2 3 4 5
-200

• Heatmaps
Heatmap is a method of representing data graphically where values are depicted by color,
making it easy to visualise complex data and understand it immediately.

TOOLS FOR DATA VISUALISATION

Some of the commonly used data visualisation tools include:

• Microsoft Excel

27
• Microsoft Power BI (Business Intelligence)
• Oracle Visual Analyzer
• SAP Lumira
• SAS Visual Analytics
• Zoho Analytics
• Jupyter Note Book and Python
• MicroStrategy
• Google Charts

28