A regional manager once asked me for a dashboard, and what she actually wanted was a single screen she could glance at with her morning coffee and know whether the week was going well. She did not want a report she had to read, she wanted a picture she could trust in about five seconds. That gap, between a document you study and a screen you glance at, is the whole reason dashboards exist.

In Part 13 you learned to pick a chart that answers a question honestly. A dashboard is where several of those charts live together on one canvas, wired so that clicking one filters the rest. This is the moment your analysis stops being a file on your laptop and becomes something a stakeholder opens on their own. We will build the same small dashboard twice, once in Looker Studio and once in Power BI, so you can feel the difference and choose the tool that fits how you work.

What a dashboard is, and how it differs from a report

A dashboard is a single screen that shows the current state of a few important numbers at a glance, usually refreshed on its own so the picture stays live. A report, by contrast, is a longer document meant to be read in order, with commentary and context around each chart. The two words get used loosely, and Power BI even uses the term report for the multi page files you build, but the useful distinction is about the reader. A dashboard reader glances and moves on. A report reader sits and studies. When someone asks for a dashboard, they are almost always asking for the glance.

That difference decides what you put on the canvas. A dashboard earns its place by answering the two or three questions a stakeholder asks every single day, and nothing more. How are sales tracking this week against last. Which channel is bringing people in. Where did signups dip. If a chart on the screen does not answer a question the reader actually asks, it is stealing attention from the ones that do. The discipline of a good dashboard is subtraction, deciding what to leave off, far more than deciding what to add.

Which tool should you start with?

The two tools a beginning analyst meets first are Google Looker Studio, formerly called Data Studio, and Microsoft Power BI. Both let you drag charts onto a canvas without writing code, and both are used in real jobs, so learning either is time well spent. The honest way to choose is to follow your data and your budget rather than a feature list. Looker Studio runs entirely in the browser, is free, and connects to Google Sheets, BigQuery and hundreds of other sources through connectors. Power BI Desktop is a free Windows application that connects to Excel, databases and files, and shines when you need to shape and model the data before charting it.

Cost is where the two part ways once you want to share, so it pays to understand it before you commit a team. Building is free in both. Sharing is where Power BI starts charging: to hand a live report to a colleague through the Power BI service, both of you generally need a Power BI Pro license, which is 14 US dollars per user per month as of 2026, having risen from 10 dollars in the April 2025 price change. Looker Studio lets you share a report with a link at no cost, and only asks for money if you want the Pro tier, which is about 9 US dollars per user per project each month for team governance features. The table below lays the real numbers side by side.

Both are free to learn on. The number that surprises teams later is the Power BI viewer license, since every person opening a shared report usually needs Pro too.

Whichever you pick, the shape of a dashboard is the same, and it is worth having that shape in your head before you open either tool. A title and a date range sit at the top so the reader knows what and when. A row of single number tiles, called scorecards or cards, carries the headline metrics. Below them, a couple of charts show the trend and the breakdown, and a filter control lets the reader narrow the whole screen at once. The diagram below is the skeleton nearly every good dashboard hangs on.

How the build actually goes

Every dashboard, in every tool, follows the same six steps, and once you have done them once the buttons in a new tool stop being scary. You connect to a data source. You shape or clean the data so the columns are the right types, the same work you met in Part 9. You drop visuals onto the canvas and bind each one to fields. You arrange them into the skeleton above. You wire filters so the screen responds to the reader. Then you publish and share. The flow below is the loop I run through every time, and I still catch myself skipping step two when I am in a hurry, which is exactly when a chart comes out wrong.

flowchart LR
  A[Connect data source] --> B[Shape and clean columns]
  B --> C[Add visuals to canvas]
  C --> D[Arrange into layout]
  D --> E[Wire filters and dates]
  E --> F[Publish and share]
  F --> G{Reader questions answered?}
  G -->|No| C
  G -->|Yes| H[Done for now]

Build your first dashboard in Looker Studio

Open Looker Studio in your browser and sign in with a Google account. Click Create, then Report, and the editor opens with an Add data to report panel already showing. That panel has two tabs, Connect to data and My data sources. Pick a connector, and for your first build the Google Sheets connector is the friendliest, choose the sheet holding your signup data, and click Add in the bottom right. Looker Studio reads your columns and drops a starter table onto the canvas, which you can delete once your own charts are in place.

Now you build the skeleton. From the Add a chart menu, add a scorecard and set its metric to the count field, and Looker Studio sums it for you into a single headline number. Duplicate it for a second card. Add a time series chart and set its dimension to date and its metric to count, and you have the trend line. Add a bar chart with channel as the dimension and count as the metric for the breakdown. Drag each onto the grid so the cards sit in a row above the two charts. Everything here is drag, drop and pick a field from a menu, with no formula to write.

Sharing is the part that makes Looker Studio pleasant for beginners. Click Share in the top right, set the access to anyone with the link, and send it. The reader opens a live dashboard in their browser with nothing to install and no license to buy. That single free link is the reason I still reach for Looker Studio when a stakeholder just needs to see the numbers and does not live inside the Microsoft world.

Build the same dashboard in Power BI

Download Power BI Desktop, which is free, and open it on Windows. Click Get data on the Home ribbon and choose your source, Excel or a text file to start, then load your signup table. Power BI opens in the Report view, the canvas where you build. On the right you will see two panes that matter: the Data pane, listing your fields, and the Visualizations pane, holding the chart types. Power BI is field driven, so selecting a field often creates a visual for you, and it guesses the chart type from whether the field is a category or a number.

To build the skeleton, click the card visual in the Visualizations pane and drag the count field into it for a headline number, then repeat for a second card. Click the line chart icon, put date on the axis and count on the values, and the trend appears. Click the clustered bar chart, put channel on the axis and count on the values, and you have the breakdown. Arrange the four visuals into the cards over charts layout. Power BI gives you more control over shaping along the way, through Power Query, which is where its extra depth over Looker Studio lives.

Saving locally with File then Save gives you a pbix file. Sharing a live version means clicking Publish on the Home ribbon, signing in to the Power BI service, and choosing a workspace. Remember the license math from earlier: to hand that live report to a colleague, both of you generally need a Power BI Pro license. For a personal portfolio piece you can skip that entirely, publish to your own account, and export a screenshot or a PDF to show off the result.

The row of cards at the top carries the numbers a reader wants before any chart. A card shows one figure large, often with a comparison to the prior period, so the reader reads the state of the business in a single line. Here is the scorecard row from our example, with this week set against last week.

Four cards, four headline numbers. The change column is what turns a raw figure into a signal the reader can act on.

Make it interactive with filters

What lifts a dashboard above a set of static images is that the visuals talk to each other. In both tools, clicking a bar filters the rest of the screen to that selection, so clicking the Search bar in our breakdown narrows the trend line and the cards to Search only. Power BI calls this cross filtering and it is on by default. Looker Studio offers the same through a setting called cross filtering that you switch on per chart. Add one more control, a date range picker in Looker Studio or a slicer in Power BI, and the reader can move the whole dashboard to last month or last quarter without you touching a thing.

Interactivity is where beginners tend to overreach, so a light hand wins. One date filter and a single slicer for the dimension people care about, channel in our case, covers the vast majority of what a reader needs. Every extra control is another thing to explain and another way to leave the dashboard in a confusing filtered state. Give the reader the two knobs they will actually turn, and leave the rest off.

Lay out a dashboard a stakeholder reads every morning

People read a screen the way they read a page, starting at the top left and moving right and down, so put the single most important number in the top left corner and let importance fall away as the eye travels. The headline cards belong across the top. The trend that people ask about most goes directly under them. Secondary breakdowns sit lower or to the right. This is not decoration, it is respecting the order in which attention arrives, and it is the difference between a dashboard someone actually uses and one they open once and forget.

Two more habits keep a dashboard readable. Keep the tile count low, because six well chosen visuals beat twelve that compete for the same glance, and every tile you cut makes the rest louder. And hold your colours steady, tying one colour to one meaning across the whole screen, so a reader learns the code once rather than relearning it on every chart, exactly the consistency you practised in Part 13. A calm, consistent screen reads as trustworthy, and trust is most of what a dashboard is selling.

Start in Looker Studio, move to Power BI when you must

Here is the plain recommendation. If you are learning, build your first dashboard in Looker Studio this week, because it is free, runs in the browser, and lets you share a working link with a friend in one click, so you get the whole loop from data to shared screen with nothing to install and no license to buy. Reach for Power BI when your data lives in Excel or a database, when you need to shape it heavily before charting, or when the job you want lists Power BI by name, which many do. Learning one transfers almost entirely to the other, so the goal is to ship a real dashboard, not to agonise over the choice.

You can now take the charts you learned to choose in Part 13 and arrange them into a live screen someone checks every morning. Part 15 steps off the drag and drop canvas and into pandas, the Python library analysts use to shape data with code when a spreadsheet or a connector runs out of room. Keep the Data Analyst guide open as your map through the series.

This week, take one small table you already have and build the four tile dashboard from the worked example, two cards, a line and a bar, in whichever tool you can open right now. Share the link or export the screen, then run the five second test on it. Do that once and a dashboard stops being a mystery and becomes a thing you can make on any Tuesday.

References

• Get started with Power BI Desktop, Microsoft Learn
• Tutorial: Create a new report, Looker Studio Help
• Power BI pricing, Microsoft Power Platform

About The Author

Dr. Pranay Jha

Dr. Pranay Jha is a Cloud and AI Consultant with 18+ years of experience in hybrid cloud, virtualization, and enterprise infrastructure transformation. He specializes in VMware technologies, multi-cloud strategy, and Generative AI solutions. He holds a PhD in Computer Applications with research focused on Cloud and AI, has published multiple research papers, and has been a VMware vExpert since 2016 and a VMUG Community Leader.

See author's posts