Introduction

A data basket allows a report viewer to build up a collection of interesting records they've found while exploring a report. There are many different scenarios where this would be useful, for example:

• In a stocks and shares report, picking a set of stocks to invest in.
• In a property report, creating a shortlist of properties to rent or buy.
• In a recruitment report, identifying and selecting a list of candidates to interview.

In this article we are going to create a simple dashboard to allow a viewer to create a 'watchlist' basket of Netflix movies and TV shows. Hopefully by the end of this article you should have a good understanding of how useful data baskets can be, and will will able to incorporate them into your own Omniscope reports.

For a more general overview of data baskets and custom queries, see:

https://help.visokio.com/support/solutions/articles/42000104910.

You can try out the dashboard created in this article by clicking here.

Setup

The first step, as with any report, is to obtain and clean-up the data.

In this example we are going use a dataset from Kaggle:

https://www.kaggle.com/datasets/shivamb/netflix-shows

If you don't already have a Kaggle account, you will need to sign up before you can download the data.

The dataset is a CSV file containing about 8000 movies and TV shows from the Netflix platform. 

After downloading, create a new Omniscope file and drag the CSV file onto the workflow. Open the File block and switch to the data tab to examine the data:

The dataset consists of a listing of Netflix movies and tv shows, with fields such as 'Director', 'Cast' and 'Country' providing details of each title.

We can now build a dashboard by connecting a Report block to the File block, then clicking on the Report block to begin editing.

Building the dashboard

The purpose of this dashboard it to allow viewers to add movies or tv shows to a basket. We need to create a report where a viewer can use filtering or selection to identify titles of interest, which can then be added to the basket.  I've used the following views, but feel free to add your own (I'm sure you can come up with some better ideas!).  

• A Pie view, split by 'Type'. This will allow a viewer to select either TV shows or Movies.
• A Bar view, split by 'Category'. This will allow a viewer to select the genre of the TV show or Movie (for example 'Comedies' or 'Kids' TV'.
• A Bar view, split by 'Rating'. A viewer can select choose which rating they are interested in (for example 'TV-PG' or 'R').
• A Choropleth Map view, joined on the 'Country' field to allow a viewer to select a country of interest.
• An Area view, split by 'Year' and stacked by 'Type', showing the breakdown of titles released by each year.
• a Table view showing a list of titles.
• A series of Content views, each showing a Measure value ('Number of titles', 'Number of directors' etc.).

Creating a data basket

The report we've created so far allows us to select and drill-down the data to identify movies or TV shows of interest.

For example a viewer may do the following:

1. Select 'TV Shows' in the Pie view, then 'Keep' to only show TV shows.
2. Select 'Kids TV' in the the Bar view, then 'Keep'.
3. Select 'America' in the Map and then 'Keep'.

By performing 3 simple actions we can reduce the number of titles from 8,800 to 42. It's then easy for a user to pick a show of interest.

This is where a data basket comes in handy. It allows a viewer to add TV shows of interest to a basket. They can then reset the filters and repeat the process, adding more and more titles to the basket.

To add a data basket to the report click the 3-dots menu in the top-right corner, then select 'Report settings'. We can choose to create a global basket, available in all the tabs, or a basket for the current tab only. In this report lets create a global basket. Ensure the 'Global' tab is selected, then in the Multi-tab queries section click 'Add basket'.

Now enter the name of the basket, for example 'Watchlist'.

Close the report settings dialog, and switch to Viewing mode. You should see the Baskets menu button on the main toolbar:

Click on the Baskets button to see a menu showing our 'Watchlist' basket and the number of titles in the basket, currently zero.

Adding data to the basket

Now let's show how a viewer can add movies and TV shows to the Watchlist basket.

Filter the data by making selections inside the views in the report and clicking 'keep' to filter the data, as outlined previously. Select the title in the table view you want to add to the basket. In this example I'm going to choose 'Saved by the bell' - a classic show every 90's kid will remember.

Now open the baskets menu and click 'Add 1 selected'.

The watchlist basket should update to show the new title added.

Let's repeat this process, adding a few more titles of interest.

At any time a viewer can look at the contents of the basket by opening the baskets menu, then clicking on the 'Show data' button.

Here they can see a list of titles currently added to the watchlist backet. The viewer can copy the list to the clipboard, or to download the data as a CSV file.

Linking a view to a basket

Data baskets are a form of named query, which means that any view in a report can be configured to dynamically show the content of the basket.

In this example we are going to add a new Tab with a Table view linked to the watchlist basket.

Add a new tab to the report, then add a Table view. Open the view options and expand the 'Query' selection. In the dropdown select the 'Watchlist' basket.

Now switch to Viewer mode, and add titles to the basket. If you switch to the Watchlist tab you should see the Table view updating as the basket is filled.

Further enhancements

In this example we built a simple dashboard that allowed users to add Netflix titles to a data basket. Some ideas for enhancing this dashboard include:

• Joining the data with other data sources. For example IMDB scores, movie/tv show posters/photos, box office information. This would provide a viewer with more criteria to identify interesting titles.
• Using different views and visualisations. We could use a Scatter view to compare popularity with release year, or a Tag view to show a word cloud based on the title description.
• Using charts inside tooltips to provide a method for users to drill-down the data inside a single view. For example, in the Bar view we could configure a tile view to show the breakdown by director.
• Adding an Editable table to allow a viewer to enter their own comments or score.

We'd love to see any ideas you come up with, so feel free to let us know!