7 Business Intelligence Terms Everyone Should Know | BI For Beginners
4 Video Views·Oct 5, 2024
Architecture & Engineering
Like any industry, business intelligence has its own jargon and terms that can be confusing if
you're unfamiliar with them. In this video, I'm going to break down 7 of the most common terms
in BI that everyone should know. Hello and welcome to Vitamin BI,
bringing you business intelligence for beginners and beyond. My name's Adam and on this channel,
I help you do more with data. So if you're interested in finding out more,
then don't forget to subscribe and hit that bell to get notified when I post new content.
Right, onto the video. Term number one.
A data silo is a standalone source of data that's not connected to any others or to a
centralized data repository. Basically, most businesses generate tons of data in different
places through all of their different activities, from simple Excel files and on-premise databases
to social media profiles and web SaaS services. You can even have different data silos within
different departments of the same company. So sales, marketing, finance, etc. could all be
using different internal systems that collect data and store it separately from the rest.
One of the basic premises of business intelligence is to harness all of the data from these different
data silos and either to use tools to query the data where it is and join or blend it with other
data from other sources or transfer that data from there to a centralized repository, which
brings us nicely onto the next term. A data warehouse is a kind of database that, as the
name suggests, warehouses data that's been extracted from different places or silos.
Its purpose is to bring all of this data together to serve the analysis processes of business
intelligence. Data warehouses contain a copy of current and historical data from different sources,
which means that they can be built and configured to make analysis faster and easier while not
getting in the way of the original data source, which is busy collecting data because the data
is being extracted from its original source and loaded into the warehouse. This means that in
between the two, the data can be cleaned and transformed to make it more compatible with the
data model that best suits the business it serves. Next up we have ETL, which is a tool that's named
after the three tasks it performs. Those three tasks being the ones that we just talked about
with data warehouses. ETL stands for extract, transform and load. So basically its job is to
take data from one place, then apply transformations to it before loading it into another place.
It essentially solves two of the main issues surrounding the data part of the business
intelligence process. Firstly, that data is everywhere in silos that aren't necessarily
connected to one another. ETLs allow you to bring all of these different data sources together into
one place, like a data warehouse. Secondly, data can be in many different formats like a table in
a database, or JSON, XML, CSV, among others. The transform functionalities of an ETL allow you to
standardize and normalize all of these different data formats,
making them far easier to blend together and analyze together. Another important function
of ETLs is to clean data by fixing any errors that may exist in the original data source,
which in turn ensures that the data and the results of data analysis are more reliable.
RDBMS stands for relational database management system. It's software that's used to manage and
administer relational databases. The most widely known of these are MySQL, SQL Server,
and PostgreSQL, or if you're old school, Microsoft Access. Data warehouses are relational databases,
so you need an RDBMS to manage and query it. If you're not familiar with what a relational
database is, it's a database made up of different data sets that are contained in what are called
tables, because the data is in a tabular rows and columns tabular format. The reason it's called
relational is that every one of these tables has a relationship to one or more other tables.
This relationship takes the form of one or more common data fields that contain the same
information. They're called joining keys. Structuring a database like this is an efficient
way of storing data and makes data analysis faster. Just as a side note, if you're getting
value out of this video, please do show your appreciation by hitting that like button.
It helps feed the algorithm and gets the channel out to more people. So thank you in advance.
SQL, often pronounced sequel, stands for structured query language. It's a language
that communicates with relational databases or RDBMS. So in terms of business intelligence,
the SQL language is used to query databases and ask them to aggregate and filter data contained
in its tables, also to join different tables together. Most BI tools have some kind of
SQL query engine behind them to carry out these tasks. Each RDBMS, MySQL, SQL Server, etc.,
has its own version of SQL that differs slightly from others in terms of its syntax,
but on the whole, they're very, very similar.
KPI stands for key performance indicator. They're metrics or values that help businesses
measure their performance or success in achieving their objectives. KPIs will vary depending on the
nature of the business activity, and they can be high level, measuring the overall performance of
a company, or focus on specific aspects of a specific activity like marketing, finance,
sales, or website traffic. KPIs are a very important part of business intelligence.
And if you'd like to know more about what they are and how to define them,
please check out this video. Link also in the description.
Next up, we have dashboard. Now, these are the end product, if you like, of the
business intelligence process. Dashboards are reports that offer an at-a-glance snapshot
of a business's KPIs and aggregated data. It's this at-a-glance characteristic that
differentiates dashboards from traditional reports, which tend to be more narrative in nature.
Dashboards are single screen reports that are often, though not always, connected to live data
and are being constantly updated. Whereas a car dashboard displays things like how fast you're
going, how much is left in the tank, or the outside temperature, a business intelligence
dashboard displays things like revenue, stock levels, and social media engagement.
If you're interested to know more about business intelligence, please check out
this playlist here. Thank you very much for watching, and please do like and subscribe.
I've been Adam Finer, and I'll be back soon with another video. Until then, stay BI curious.

