Tableau vs KNIME

Tableau Desktop is a BI solution for data visualization, dashboarding and location analysis. In online reviews, users said they found its drag-and-drop charting a boon for creating charts and maps. Regarding customization, many users praised the platform for its various labeling and design options.I recently tried the Tableau Desktop 2024.1.3 version. The trial is only for 14 days and is enough for a sneak peek into Tableau’s dashboarding and data storytelling capabilities. For more straightforward use cases, Tableau is incredibly user-friendly and fast. Creating a new sheet gives you a canvas to create a visualization. Once you have the required sheets, combining them into a dashboard view is straightforward — select and add.My dataset included healthcare data, including details of patients, their hospital visits and insurance payer details. One use case was to find the total claim settlement amount. I dragged the Total Claims Cost and Payer fields to the column and row shelves, and Tableau gave me a bar graph. The toolbar had single-click options for sorting data from increasing to decreasing values or the other way around.To view the number of encounters by payer, I dragged the Payer field to the row shelf and used the SUM(ROW_COUNT()) function on the column shelf. The chart popped up with more visualization and layout options.I wanted an interactive filter to view the average claim cost by birthdate. I dragged the Birthdate field to the Filters shelf and right-clicked on it to set the end date as October 22, 1961. Selecting Show Filter added a slider conveniently to the right of my visualization. I could see the data for people born before October 22, 1961, and if required, I could change the end date.Another use case would be viewing the data by the type of hospital visits — how many people were inpatients, outpatients or those who needed emergency care. I dragged and dropped the Total Claims Cost and Payer fields into columns and rows, respectively. Similarly, I dropped Encounterclass into the Filters shelf and clicked on Show Filter to enable a checkbox on the screen. It had all the categories of visits, giving users the option to select the desired views.One-fourth of the users discussing adoption said there was a steep learning curve. Tableau relies on Python and R scripts for statistics in its visualizations. It's where the named licenses can prove to be a blessing, as you can opt to train upcoming Creators and Explorers. We recommend factoring in training if you want to hit the ground running.Some reviewers felt discounted packages for business editions should be available, similar to the free student licenses. At $70 per user, the Creator license can seem costly when compared to Power BI ($9.99 per user) and Qlik Sense ($30 per user).Here's the good news, though. Its built-in user management acts as a permissions layer for your organization - users can only access the relevant content. Plus, an organization will have very few Creators and a greater number of Viewers and Explorers, and the license fee reduces from Creator to Explorer to Viewer.We recommend opting for a wise license combination to get the most out of the product.On the upside, the vendor constantly releases new features, the latest one being Einstein CoPilot in beta.Overall, Tableau is a competitive BI solution, but if the pricing seems inflexible, quite a few other solutions offer live insights and advanced analytics out of the box.

KNIME is a robust open-source solution with cross-platform interoperability. It integrates with a range of software, such as JS, R, Python and Spark. With a variety of nodes and functions, it can process large datasets with a decent level of control in each step. Workflows are displayed as connected nodes, making it easy to isolate and fix specific steps. It also contains built-in tools to create and test supervised and unsupervised machine learning models. Users found the UI very intuitive and flexible. On the flip side, they found the tool visually lacking and primitive. The system also has performance and stability issues. Processing big data is very time consuming since the platform isn’t cloud-based. Users reported excessive memory usage as well. It also lacks reporting or monitoring features. Decent technical knowledge is required to fully leverage its capabilities.