This is the Grafana dashboard I am going to
create in this video. I will start with a brief description of all
the dashboard elements and then proceed to the step-by-step instructions. Obviously, before you can follow along, Grafana
needs to be installed, or you need to have a Grafana Cloud account. I have a video, this one, about the Grafana
installation. Please, watch it first if you are still wondering
where to begin. In this video, I assume you know how to login
into Grafana and cannot wait to start experimenting. Let's go. Hello and welcome to Volkov Labs Youtube channel. My dashboard is created for a company whose
business spins across the globe with sales in the United States. On the top, I have time zones and a company
logo. They are followed by sales per month in a
bar chart and three kinds of sales plans. I use color-coding to indicate how close the
actual sales are to the company's goals. Next, I have single stat elements where
color also has an indicative function. On the bottom of my dashboard, I placed a
couple of rows. A Grafana row is a group of panels that could
be shown or hidden using this little arrow. That row contains 2 PDF documents. And the other row displays two detailed visualizations. The first one utilizes a geomap showing sales
by region where I can zoom in and zoom out. On this one, on the bottom, you can see sales
per month per product. I can filter by product with multiple selections
by clicking on a product name and holding the Shift button. The last elements on my dashboard are filters
by product name and by product category. All right, now it is time to go over the step-by-step
part for each of these components. You can find timestamps in the description
box below if you need to switch to a particular element. Right after you install Grafana and before
you can create your first dashboard, you need to add a data source. A data source is a translator between your
data storage and Grafana visualizations. There is no other way; there must always be
a data source. To add one, go to the configuration and then
hit the add data source button. Look at all these options; there are time-series
databases, logging, and document databases, tracing, SQL, and more. If you do not see the data source for your
data storage, well, with pleasure, Volkov Labs can build it for you. For this demonstration, my data is stored
in the PostgreSQL database; therefore, it is my choice. After selecting the data source, I need to
fill out its configuration. For the PostgreSQL, here is what the form
looks like. The name could be anything. I keep the default value. Since I use PostgreSQL in a docker container,
my host is host.docker.internal, and I keep the default port 5432. Next, my database is postgres, user is postgres,
and the password is mysecretpassword. I disable the TLS/SSL Mode and now can save
and test my connection. This successful message tells me that I am
all set with the data source! To start my dashboard, I hover over this cross
and select dashboard. Next, I save my dashboard by using this save
icon and typing a good descriptive name. In my case, it is Sales Analytics. I click save. Here is it, a brand new, empty dashboard. I want to add a star so that later on, I can
make it my default dashboard. Now I need to add a panel. Every visualization belongs to a panel. Click this icon and select the Add a new panel
option. Now I am "inside" the panel; almost all panel
configurations are here for me to adjust. Here is a drop-down with all installed visualizations. I select a bar chart. After that, I switch the data view to a table
view. In a table view, all fetched data is displayed
in a columnar format, and this is what I always double-check before working with any graphical
visualizations. Grafana was initially created to visualize
and monitor servers and network activities. That made time series its default data format. However, the table format is well-supported
as well. That brings me to my next stop, where I change
from Time Series format to a Table format. This area is designed to specify your query. It can be done via query builder, but I am
native to SQL, so I prefer to switch to the raw SQL format view. How exactly you write and test your SQL is totally
up to you. I use the pgAdmin4 tool, which could easily
be installed in a docker container. I explain docker containers in my other video. Check it out if it is still confusing. All right, back to the SQL query creation. This is my SQL to fetch sales per month. I run it, confirm everything is excepted,
then copy it into the Grafana query area. After that, I check the table view. Everything looks good. Now I can switch to the graphical view and
work on the visual representation. Note that when I am hovering over the bars,
Grafana displays the data point details. The captions come from the SQL aliases. All right, the query is in place, and the
visual is working. Next, I want to adjust my style a little bit. Here I work with the title, bar width, line
width, set units to US dollars, and color to a fixed blue. Looking good enough to save and apply. Here you go. The first visualization with sales per month
is ready. Now let's add a bar gauge. For that, click on Add panel icon, then use
Add a panel button. Select a bar gauge from the drop-down menu. Make sure the data source is correct, meaning
the one you added for this visualization. The format is Table, and switch to raw SQL. By my design, I need two queries for every
bar gauge. One is for as of today's sales, and that query
goes into the query A section. Then I click plus query button and enter my
second query, for the plan value, into the query B section. Next, I need to specify how queries A and
B are related. That could be done by using Grafana Transformations. For that, I can go to the 'Transform' tab. Scroll down and select 'Config from query
results' option. That transformation adjusts the results of
one query based on the other query. My config query is B, I want to apply it to
all numeric fields and use it as Max. Those settings make the results of query B
to be used by the bar gauge to set the max length of its bar. That's all for query prep. Let's work on visual appearance. On the right side, I am going to change the title. I like a Horizontal orientation and Retro
LCD style. I work with sales in Us Dollars. And I want explicitly specify the minimum
as 0. Otherwise, by default, it takes the value
from the existing query. The last configuration I would like to change
would be colors. I want them to start from red with a gradient
into green. Uh, nice! Click save and apply. Adjust the size and location. I am going to repeat the same steps for two other bar
gauges. But to make the process a little faster, I
will use the duplicate function. Looking great so far. Let's move on. The following three visualizations are a single
value stat with different flavors. The first one, units sold, is pretty straightforward. I add a new panel, then select a Stat visualization
from a drop-down, specify the query, and change the panel title and style. For the next one, I would like to add a color
coding and use red if the metric requires immediate attention. For color coding configuration, I need to
add a second query which I use in a transformation like this, and then I will add a threshold
on the right side. Here you go. Now the background color depends on the % of
the returned units. For the third single stat, I use SQL that
fetches the series of data, not a single value. Here I have sales per day. The line in the background would indicate
the trend. Sometimes, you want to add a document to your
dashboard. Maybe a file with some instructions or a list
of specific to your business abbreviations. That could be done by employing the Base64/PDF
plugin. It is not part of the core Grafana. It needs to be installed from the Marketplace. Grafana Marketplace is in the Configuration->Plugin
Menu. Search for Base64 and click Install. After that, I can select the Base64/PDF plugin
from the visualizations drop-down list. In the database, I already have the data prepared. Meaning I already converted my PDF files into
Base64 format and inserted them into the database table. What is left is to write a straightforward
query and adjust the panel on the dashboard. Repeat if needed for more documents. PDF documents are informative but could take
up too much space and are not needed all the time. That's why I want to hide them in a row. For that, I click add a panel, then add a
new row button. Specify the header and place this raw anywhere
I like. I can add panels by dragging and dropping
them into this raw. Geo Map visualization looks stunning and is
incredibly easy to work with. As you already guessed, it requires two main
parameters: latitude and longitude. I have that prepared in my database and is
ready to specify them in my query, Next, I adjust the initial view to start from
North America. I make the size of data points dependable
on Sales in that location, maybe a couple of other things, and that's it. Moving on to the clock panels. I need to select Clock from the visualization
drop-down. The configuration is straightforward. I specify a time zone and a time format for
display. Hit the apply and save. Resize and rearrange the dashboard if needed. I repeat this exercise two more times for
two other time zones. And here is the look of my dashboard so far. Neat! In order to display a logo, I will use the
plugin that I installed earlier for PDF documents. Select Base64/PDF, and specify the query,
title, and size. Resize the panel on the dashboard. It is looking better with every new panel,
right? The last elements to cover are the across
the dashboard filters. They are located on the top of my dashboard. The Grafana filters are created as variables. Those variables are predefined in the dashboard
settings and can be referenced in any panel. To add a variable, I click on this icon and
then select variables. Click new. Here, the name is what I will use for reference
in the queries. The label is what is displayed on the dashboard. One of my variables contains a list of all
product names. And here is where I set up the query for that. Yes, I need to have the All option and be
able to multi-select. Repeat for the second variable. That one contains a list of all product categories. Here you go. Now I have two drop-downs on my dashboard. Let me show you how to use variables. Here in the query section, I reference them
with a dollar sign. After I am done, the panel refreshes accordingly
when I change the filter. This was my quick guide to your first Grafana
dashboard. If you have any questions or suggestions,
do not hesitate to reach out. And before I wrap it up, I know when you first
launch Grafana, it looks like this. It looks dark. If you like me and want to change that, on
the bottom left side, find an icon that is designated for your avatar, hover over it,
and select the Preferences menu. Then switch IU Theme to light and click Save. Much better! Please support this channel by giving me a
thumbs up if you find my content helpful! Let me know in the comments below if there
is any plugin you would like me to review. Thanks for watching, and see you next time!
