Why context is critical

In this lesson, you have been learning about the importance of context in business
intelligence. As a refresher, context is the condition in which something exists or happens. For
example, in a previous video you considered this data visualization:

This line graph just shows five different lines on a grid, but we don’t have any information
about what the lines of the graph represent, how they’re being measured, or what the
significance of this visualization is. That’s because this visualization is missing context. Check
out the completed version of this visualization:

This visualization has all of the information needed to interpret it. It has a clear title, a legend
indicating what the lines on the graph mean, a scale along the y axis, and the range of dates
being presented along the x axis. Contextualizing data helps make it more meaningful and
useful to your stakeholders and prevents any misinterpretations of the data that might
impact their decision-making. And this is true for more than just visualization! In this reading,
you’ll explore a business case where context was key to a BI project’s success.
The scenario

The CloudIsCool Support team provides support for users of their cloud products. A customer
support ticket is created every time a user reaches out for support. A first response team is in
charge of addressing these customer support tickets. However, if there is a particularly
complex ticket, a member of the first response team can request help from the second
response team. This is categorized as a consult within the ticketing system. The analytics
team analyzes the ticket and consults data to help improve customer support processes.

Usually, the consultation request is fulfilled successfully and the first response team is able to
resolve the customer’s ticket, using guidance from the second response team. However,
sometimes even the second response team isn’t able to fully answer the question or new
details about the case require additional insight. In that case, the first response team might
ask for another consultation, which is labeled as a reconsult.

This is all important context for a BI professional working with stakeholders who are
interested in how well current support processes are working and how they might be
improved. If they build reporting tables and dashboards that only track consults and not
reconsults, they might miss key insights about how effective the consultation system truly is.
For example, a high reconsult rate would mean that more cases aren’t being resolved in the
first or second attempts. This could lead to customers waiting longer for their issues to be
resolved. The leadership would want to evaluate these processes.

Knowing this context, the BI professional working on this project is able to build out
appropriate metrics, reporting tables, and the dashboard that tracks that metric in a way that
helps stakeholders make informed decisions about this process. By understanding the
business context, BI professionals can create more meaningful reports.

Conclusion

Context is the who, what, where, when, and why surrounding data that makes it meaningful.
Knowing this background information helps us interpret data correctly and visualize useful
business intelligence insights for stakeholders. When BI professionals understand the
context, choose the right data, and build contextualized visuals to share with stakeholders,
they can empower businesses and leadership to make successful decisions.
Data ethics and the importance of
data privacy
Recently, you’ve been learning about the importance of context in business intelligence. You
discovered that, when you contextualize, you put something into perspective by considering
its origin and other relevant background information; the motivation behind it; the larger
setting in which it exists, such as a particular time period; and what it might have an impact
on. Contextualization also supports fairness and reduces the chance of bias when your users
seek to gain useful insights from the data you’re presenting.

Likewise, as a BI professional, you have a responsibility to treat data ethically. Data ethics
refers to well-founded standards of right and wrong that dictate how data is collected,
shared, and used. Throughout your career you will work with a lot of data. This sometimes
includes PII, or personally identifiable information, which can be used by itself or
with other data to track down a person's identity. One element of treating that data ethically
is ensuring that the privacy and security of that data is maintained throughout its lifetime. In
this reading, you will learn more about the importance of data privacy and some strategies
for protecting the privacy of data subjects.

Privacy matters

Data privacy means preserving a data subject’s information and activity any time a data
transaction occurs. This is also called information privacy or data protection. Data privacy is
concerned with the access, use, and collection of personal data. For the people whose data is
being collected, this means they have the right to:

 Protection from unauthorized access to their private data

 Freedom from inappropriate use of their data
 The right to inspect, update, or correct their data
 Ability to give consent to data collection
 Legal right to access the data

In order to maintain these rights, businesses and organizations have to put privacy measures
in place to protect individuals’ data. This is also a matter of trust. The public’s ability to trust
companies with personal data is important. It’s what makes people want to use a company’s
product, share their information, and more. Trust is a really big responsibility that can’t be
taken lightly.

Protecting privacy with data anonymization

Organizations use a lot of different measures to protect the privacy of their data subjects, like
incorporating access permissions to ensure that only the people who are supposed to access
that information can do so. Another key strategy to maintaining privacy is data
anonymization.

Data anonymization is the process of protecting people's private or sensitive data by

eliminating PII. Typically, data anonymization involves blanking, hashing, or masking
personal information, often by using fixed-length codes to represent data columns, or hiding
data with altered values.

Data anonymization is used in just about every industry. As a BI professional, you probably
won’t personally be performing anonymization, but it’s useful to understand what kinds of
data are often anonymized before you start working with it. This data might include:

 Telephone numbers
 Names
 License plates and license numbers
 Social security numbers
 IP addresses
 Medical records
 Email addresses
 Photographs
  Account numbers

Imagine a world where we all had access to each other’s addresses, account numbers, and
other identifiable information. That would invade a lot of people’s privacy and make the
world less safe. Data anonymization is one of the ways we can keep data private and secure!

Key takeaways

For any professional working with data about actual people, it’s important to consider the
safety and privacy of those individuals. That’s why understanding the importance of data
privacy and how data that contains PII can be made secure for analysis is so important. We
have a responsibility to protect people’s data and the personal information that data might
contain.

If you’re interested in learning more about data privacy and ethics, you can check out the
Google Data Analytics Certificate program’s section on bias, credibility, privacy, ethics, and
access.

Anticipate data limitations

We live in a world where data is constantly being generated. There is so much information out
there to learn from. But we also live in a world that is constantly changing, and often the data
that we encounter has certain limitations we need to consider as we analyze data and draw
insights from it.

Factors of data availability

Previously, you learned about the importance of data availability, which is the degree or
extent to which timely and relevant information is readily accessible and able to be put to
use. The factors that influence data availability are:

 Data integrity: The accuracy, completeness, consistency, and trustworthiness of data

throughout its life cycle.

 Data visibility: The degree or extent to which information can be identified, monitored,
and integrated from disparate internal and external sources.

 Update frequency: How often disparate data sources are being refreshed with new
information.

 Change: The process of altering data, either through internal processes or external
influence.
Next, you are going to consider the limitations of data that might change the availability and
how you can anticipate those limitations as a BI professional.

Missing data

If you have incomplete or nonexistent data, you might not have enough data to reach a
conclusion. Or, you might even be exploring data about a totally different business problem!
Understanding what data is available, identifying potential other sources, and filling in the
gaps is an important part of the BI process.

Misaligned data

As a BI professional, you will often use data from different sources. Some of these might be
internal sources to the business you’re working with, but they might also include external
sources. These sources might define and measure things in completely different ways. In
cases like these, establishing how to measure things early on standardizes the data across the
board for greater reliability and accuracy. This will make sure comparisons between sources
are meaningful and insightful.

Dirty data

Dirty data refers to data that contains errors. Dirty data can cause errors in your system,
inaccurate reports, and poor decision-making. Implementing processes for cleaning data by
fixing or removing incorrect, corrupted, incorrectly formatted, duplicate, or incomplete data
within a dataset is one way you can prepare for this limitation.

Conclusion

As a BI professional, you’ll need to understand that sometimes the data you work with will
have limitations. This could mean that it doesn’t fit within a certain time range, or it only
applies to specific situations, or there are challenges identifying the data you need. Being
able to anticipate those issues and consider them when you build tools and systems for your
business will allow you to ensure that those limitations don’t stop your stakeholders from
getting the data they need to make great decisions and ensure project success.
How to identify key metrics for
a project
Choosing your metrics

In a previous video, you learned how business intelligence professionals determine which
metrics to include in their dashboards to deliver relevant and actionable data to their
stakeholders. In this reading, you’re going to consider how choosing the right metrics can
determine the success of a project. You’ll do this by exploring an example of a BI professional
identifying key metrics for their project.

There are five key points BI professionals take into account when choosing metrics:

1. The number of metrics: More information is not always better. BI

professionals limit the number of metrics on dashboards to focus specifically on the
ones that are key to a project’s success. Key metrics are relevant and actionable. For
instance, if metric X drops, is this good or bad? What action would a user take if it
dropped that would be different if it rose instead? Too many metrics that aren’t
relevant to the project can be confusing and make your dashboard less effective. The
goal isn’t to overload the dashboard to account for every single use case, but 80% of
the common use cases.
2. Alignment with business objectives: Understanding the business
objectives can help you narrow down which metrics will support those goals and
measure their success. For example, if the business objective is to increase sales,
include revenue in your dashboard. You will most likely not want to include a metric
such as customer satisfaction because that is not directly related to the business
objective of increasing sales.
3. The necessary technologies and processes: It’s important to confirm
that the necessary technologies and processes are in place for the metrics you’re
choosing. If you can’t obtain and analyze the necessary data, then those metrics aren’t
going to be very useful.
 4. The cadence of data: You have to consider how frequently the data becomes
available. If a lot of metrics are delivered at a different cadence and frequency, it
becomes difficult to schedule a review.
5. Use SMART methodology: If you earned your Google Data Analytics
Certificate, you know the SMART methodology is a useful tool for creating effective
questions to ask stakeholders. It can also be used to identify and refine key metrics by
ensuring that they are specific, measurable, action-oriented, relevant, and time-
bound. This can help you avoid vague or super-high-level metrics that aren’t useful to
stakeholders, and instead create metrics that are precise and informative.

An integrated view

In the BI world, data requires a dynamic and thoughtful approach to detect and respond to
events as they happen. An integrated view of the whole business is required. In some cases,
metrics can be straightforward. For example, revenue is fairly unambiguous: Revenue goes
up, and things are going well! But other metrics are a little more complicated.

In an earlier reading, you discovered the importance of context for the CloudIsCool Support
team when measuring their ability to effectively answer customer support questions. As a
refresher, a customer support ticket was created every time a customer reached out for
support. These tickets were addressed by the first response team at CloudIsCool. Sometimes
the first response team needed help answering more complex tickets. They would then reach
out to the second response team. This was marked as a consult on the support ticket.

Imagine that the BI professionals working with this team now are trying to decide which
metrics are useful in a dashboard designed to increase customer satisfaction ratings for
support tickets. Perhaps their stakeholders are interested in monitoring consults to ensure
that customers are getting the help they need in a timely manner. So the BI team considers
adding consult rate, which is the rate at which customer support agents are asking for help
from internal experts, as a metric in their dashboard.

Note that an increasing consult rate could be good or bad. It might mean that customer
support agents are being more customer-centric and trying to ensure each customer gets the
best answer. But it could also mean that agents are being overwhelmed with complaints and
having to offload them onto internal experts in order to keep up. Therefore, consult rate is a
metric that doesn’t have a clear direction; nor does it have an obvious influence on the
decision-making process on its own. So, it’s not a useful metric for this dashboard. Instead,
the BI professionals select metrics that indicate success or failure in a more meaningful way.
For instance, they might decide to include a metric that tracks when a support agent
experiences missing support documentation. This will help leaders decide whether to create
more documentation for agents to reference. Notice how this metric has a clear line of action
that we can take based on how high or low it is!

Conclusion

The ability to choose metrics that inform decision-making and support project success is a
key skill for your career as a BI professional. Remember to consider the number of metrics,
how they align with your business objectives, the technologies and processes necessary to
measure them, and how they adhere to SMART methodology. It’s also important to maintain
an integrated view of the entire business and how the information your metrics deliver is
used to guide stakeholder action.

North star metrics

So far, you have been learning about how BI professionals choose the right metrics to
measure the success of their projects. BI professionals also use another specific kind of metric
to measure the long-term success of the entire business or team; this metric is often referred
to as a north star metric. In this reading, you will learn more about north star metrics, how BI
professionals choose them, and how they can help a business’s growth over time.

The guiding star

A company’s north star metric goes beyond short-term goals– it’s intended to capture the
core measurable value of a business’s product or services over its entire lifetime. These
metrics are a guiding light that drive a business forward. That’s why it’s called a north star
metric– like the north star can be used to navigate the wilderness, these metrics can be used
to navigate business decisions and lead a business to growth.

Having this metric as the guiding light for the entire business is useful in three primary ways:
 1. Cross-team alignment: Different teams have different specialties and focuses
that help a business function. They aren’t always working on the same projects or with
the same metrics, which can make it difficult to align across the entire business. A
north star metric allows all of the teams to have a consistent goal to focus on, even as
they work on different things.
2. Tracking growth: It can be difficult to understand and track the growth of an
entire organization over time without understanding the driving metrics that
determine growth. A north star metric provides a long-term measurable data point
that stakeholders can focus on when discussing overall performance and growth in a
business.
3. Focusing values: A north star metric is primarily a guiding principle for a
business– it determines what is important to the organization and stakeholders. This
means that choosing the right metric to guide a business can help keep the values in
check– whether that’s customer satisfaction, number of customers completing the
sales cycle, or customer retention.

Choosing a north star metric

Because north star metrics are so key to a business’s ongoing success, choosing the right
metric is a foundational part of a business intelligence strategy. The north star metric has to
measure the most essential part or mission of the business. And because every business is
different, every business’s north star metric is going to be unique. In order to determine what
the most useful north star metric might be, there are a few questions you can ask:

 What is essential to this business’s processes?

 What are the most important KPIs being measured?
 Out of those KPIs, what captures all of the necessary information about this business?
 How can the other metrics be structured around that primary metric?

Real north star metrics

Because more businesses have begun using north star metrics to guide their business
strategies, there are a lot of examples of north star metrics in different industries:

 E-commerce:
 Weekly number of customers completing the sales cycle
  Value of daily purchases
 Social media:
 Number of daily active users
 Messages sent per day
 Streaming and media services:
 Number of new sign-ups
 Total reading time
 Total watching time
 Monthly subscription revenue
 Hospitality:
 Number of nights booked
 Number of repeat customers

These are just a few examples– there are a lot of potential north star metrics for businesses to
choose from across a variety of industries, from tech to finance!

Key takeaways

As a BI professional, one of your responsibilities will be to empower stakeholders to make

business decisions that will promote growth and success over the long term. North star
metrics are a great way to measure and guide a business into the future because they allow
you to actually measure the success of the entire business, align teams with a single goal, and
keep the business’s values at the forefront of their strategy.

Bridge the gap from current state

to ideal state
Bridge the gap
Business intelligence professionals continually monitor processes and systems to determine
if it’s necessary to make updates for greater efficiency and optimization. These professionals
explore ways to bring the current state closer to the ideal state. They do this through a
process called gap analysis, which is a method for examining and evaluating the current state
of a process in order to identify opportunities for improvement in the future.

Gap analysis involves understanding where you currently are compared to where you want to
be so that you can bridge the gap. BI uses gap analysis to do all kinds of things, such as
improve data delivery systems or create dashboard reports.

For example, perhaps a sales team uses a dashboard to track sales pipeline progress that has
a six-hour data lag. They use this dashboard to gather the most up-to-date information as
they prepare for important meetings. The six-hour lag is preventing them from accessing and
sharing near-real-time insights in stakeholder meetings. Ideally, the delay should be one hour
or less.

Setting direction with stakeholders

The first step in bridging the gap is to work with stakeholders to determine the right direction
for this BI project. Establishing stakeholder needs and understanding how users are
interacting with the data are important for assessing what the ideal state of a system actually
is. What needs do stakeholders have that aren’t being met or could be addressed more
efficiently? What data is necessary for their decision-making processes? Working closely with
stakeholders is necessary to understand what they actually need their BI tools to do.

The BI professionals collect information and learn that, as the company grew, it opened
offices across the country. So, the sales teams are now more dispersed. Currently, if a team
member from one office updates information about a prospective client, team members from
other offices won't get this update until the workday is almost over. So, their goal is to reduce
the data delay to enable better cross-team coordination.

Context and data quality

In addition to identifying stakeholder needs, it’s also important for the BI professional to
understand the context of the data they interact with and present. As you know, context is the
condition in which something exists or happens; it turns raw data into meaningful
information by providing the data perspective. This involves defining who collected it or
funded its collection; the motivation behind that action; where the data came from; when;
the method used to collect it; and what the data could have an impact on. BI professionals
also need to consider context when creating tools for users to ensure that stakeholders are
able to interpret findings correctly and act on them.

It’s also critical that BI professionals ensure the quality and integrity of the data stakeholders
are accessing. If the data is incorrect, the reporting tools won’t be accurate, and stakeholders
won’t be able to make appropriate decisions — no matter how much context they have been
given.

Now, the sales team's BI professional needs to identify data sources and the update
frequency for each source. They discover that most of the key data sources update every 15
minutes. There are a few nonessential data sources that rarely get updated, but the team
doesn’t actually have to wait until those data sources are updated to use the pipeline. They’re
also able to confirm that the data warehouse team will verify these data sources as being
clean and containing no duplicates or null fields that might cause issues.

Building structures and systems

A large part of a BI professional’s job is building structures and systems. This means designing
database storage systems, organizing the data, and working with database governance
specialists to maintain those systems. It also involves creating pipeline tools that move and
transform data automatically throughout the system to get data where it needs to go to be
useful.

These structures and systems can keep data organized, accessible, and useful for
stakeholders during their decision-making process. This empowers users to access the data
they need when they need it — an ideal system should be organized and structured to do just
that. To address the sales team’s needs, the BI analyst in this case designs a new workflow
through which data sources can be processed simultaneously, cutting down processing time
from 6 hours to less than an hour.

Sharing findings
If you are coming to this course from the Google Data Analytics Certificate, you may already
be familiar with the share stage of the data analysis process. This is the point at which a data
analyst creates data visualizations and reports and presents them to stakeholders. BI
professionals also need to share findings, but there are some key differences in how they do
so. As you have been learning, creating ways for users to access and explore data when they
need it is a key part of an ideal BI system. A BI professional creates automated systems to
deliver findings to stakeholders or dashboards that monitor incoming data and provide
current updates that users can navigate on their own.

In the sales team dashboard example, the final output is a dashboard that sales teams across
the country use to track progress in near-real time. In order to make sure the teams are aware
of the updates, the team’s BI analyst shares information about these backend improvements,
encouraging all sales teams to check the data at the top of the hour before each meeting.

Acting on insights

BI focuses on automating processes and information channels in order to transform relevant

data into actionable insights that are easily available to decision-makers. These insights
guide business decisions and development. But the BI process doesn’t stop there: BI
professionals continue to measure those results, monitor data, and make adjustments to the
system in order to account for changes or new requests from stakeholders.

After implementing the backend improvements, the sales team also creates system alerts to
automatically notify them when data processes lag behind so they're prepared for a data
delay. That way, they could know exactly how well the system is working and if it needs to be
updated again in the future.

Conclusion

A large part of a BI professional's work revolves around identifying how current systems and
processes operate, evaluating potential improvements, and implementing them so that the
current system is closer to the ideal system state. Throughout this course, you’ll learn how to
do that by collaborating with stakeholders, understanding context, maintaining data quality,
sharing findings, and acting on insights.
Case study: USDM - Selecting key
project metrics
In this part of the course, you have been focusing on how business intelligence professionals
identify effective metrics for a project. A key part of this process is working with stakeholders
to understand their data needs and how those interests can be measured and represented
with the data. In this case study, you will have the opportunity to explore an example of how
the BI team at USDM worked with stakeholders to develop metrics.

Company background

USDM, headquartered in Santa Barbara, California, collaborates with life science companies
across a variety of industries, including biotechnology, pharmaceutical, medical device
technology, and clinical. USDM helps its customers, from large-scale companies to small
businesses, ensure that their database systems are compliant with industry standards and
regulations, and work effectively to meet their needs. USDM’s vision is to bring life sciences
and healthcare solutions to the world better and faster—starting with its own company
values: customer delight, accountability, integrity, respect, collaboration, and innovation.
The challenge

In this case study, you’re going to explore an example of USDM’s work with one of their
clients. The client for this project researches and develops antibody treatments for cancer
patients. The client needs analytics that measure the effectiveness and efficiency of their
products. However, with the client’s existing database, to get the types of reports they need,
they have to access many systems, including facility data, licensing information, and sales
and marketing data. All of this data exists in various places, and as a result, developing
analysis reports creates issues for the client’s stakeholders. Also, it makes it harder to
compare key metrics because so many KPIs needed to be brought together in one place.

To help better understand how effective their product is and forecast demand, the client
asked USDM to help architect a data storage system that could address their specific needs.
They needed a system that could bring the data their team needs together, follow industry
regulations, and allow them to easily create reports based on key metrics that can be used to
measure product effectiveness and market trends. A significant part of this initiative started
with the basics: what were the actual key metrics for the client’s team and what data systems
did they come from?

The approach

To identify which metrics were most important for the client’s business needs, the USDM
team needed to get input from a variety of different people from across the organization. For
example, they needed to know what charts the sales and marketing teams who used this data
for their reports needed, what their existing processes were, and how to address these needs
in the new system. But, they also needed to know what data the product development team
used in order to measure efficacy.
USDM worked closely with different teams to determine what charts they needed for reports,
how they were accessing and using the database system currently, and what they were
hoping to achieve with the new system. As a result, the team was able to determine a
selection of key metrics that represented their client’s business needs. These metrics
included:

 Sales performance
 Product performance
 Insurance claims
 Physician information
 Facility data

To enact a business intelligence solution there must be both the business interaction with
stakeholders and the technical interaction with the architects of other team’s systems. Once
these metrics were identified by the client, the USDM team collaborated with other members
of the client’s team to begin building a new solution that could capture these measurements.

But, almost every project comes with unexpected challenges; the database tool the team was
using to develop the new system didn’t have all of the features the team needed to capture
their must-have metrics. In this case, the USDM team collaborated with leadership to develop
a list of requests from the tool vendor, who was able to address their team’s unique needs.

The results

By the end of the project, the USDM BI team architected a data storage system that
consolidated all of the data their team needed from across a variety of sources. The system
captured the key metrics the client needed to understand their product’s effectiveness,
forecast sales demand, and evaluate marketing strategies. The reporting dashboards created
with this data storage system included everything the stakeholders needed. By consolidating
all of the KPIs in one place, the system could provide faster insights and save the client time
and improve efficiency without having to run reports from every individual system. The
solution was more automated and efficient—and importantly, designed specifically with their
team’s most useful metrics in mind.

Conclusion

Collaborating with users and stakeholders to select metrics early on can help determine the
long-term direction of a project, the specific needs stakeholders have, and how to design BI
tools to best address unique business needs. As a BI professional, a key part of your role will
be considering key metrics and how to tailor the tools and systems you create to capture
those measurements efficiently for reporting use.