Alteryx to Dataiku: Feature Differences

Dataiku Product Chris Helmus, Dmitri Ryssev

Hello and welcome to our final installment of the Alteryx to Dataiku series! If you missed the previous articles, they can be found here. For today’s episode, let’s have a look at some other, less obvious but still important differences between the two platforms, and what the effects of these differences are. 

Charts on Charts

Alteryx has quite a few options when it comes to batch reporting and there’s always the trusty Browse tool to get exploration started. Dataiku has similar functionality in its Explore pane but really starts to shine when creating drag and drop visualizations in-line.

On any dataset in a project, it’s easy to start rapidly creating visualizations to explore relationships, verify results, or create a view for a colleague to inspect. With dozens of chart types, you can chart things over time, break things down, or even make some quick pivot tables and KPIs. If the data doesn’t quite tell the story, you can even make custom aggregations that you forgot to create upstream in your process. If you’re working with large data sets, several chart types can dynamically create SQL in the background to make sure you get fast and accurate results.

pivot table in Dataiku

 Who doesn’t love a good pivot table?

Once you have some visuals you like, it’s possible to publish them right to a dashboard. You can also add in some filters that apply across multiple visuals at once and keep them up to date with Dataiku scenarios we covered in the last article in this series. And don’t worry, you can absolutely give permission for someone to see your dashboard without being able to edit your project!

Collaboration & Documentation 

A significant difference between Alteryx and Dataiku lies in their approach to collaboration. In Alteryx, you typically work on a workflow individually using their Designer product, whether desktop or cloud-based. To share your work, you would use their Server product or distribute the workflow file directly via email, file share, etc. Another product called Alteryx Connect also provides cataloging features for data assets.

In contrast, Dataiku facilitates real-time collaboration by default. Multiple users can access and work on the same project simultaneously, provided they have the necessary permissions. This collaborative environment allows power users to develop data flows, business domain experts to provide context, and data scientists to apply advanced techniques, all within a single, shared workflow. Dataiku ensures that every user, technical or not, can contribute meaningfully to the project.

customer lifetime value in Dataiku

With Dataiku, I can pick up where my colleague left off.

Seamless documentation is crucial for any collaborative data project. Dataiku’s native features for commenting, project wikis, to-do lists, and tagging (for users, projects, and other assets) make it significantly easier for others to understand the context and objectives. These features ensure that anyone new to the project can quickly get up to speed, promoting collaboration among all users. Don’t feel like writing the documentation yourself? Leverage Dataiku’s AI Explain feature to have a Generative AI model write the documentation for you based on your project’s content.

Built-in Wiki for documentation of your projects

Built-in Wiki for documentation of your projects

Working off of one copy of the project where multiple users can contribute reduces silos in an organization. And because there is only one project file in this case, you don’t run into the issue of “Are we sure we’re working off of the latest version of this project?”

Version Control

Speaking of versioning, let’s talk for a moment about version control. If you come from working in Alteryx, you will be relieved to find out that Dataiku provides a version history of every change made in a project, by default. You can always look at the ‘diffs’ (changes made since previous versions) and can choose to revert back to a previous version, if you feel so inclined. This version history can be viewed at the holistic project level or for individual recipes.

version history Dataiku

A full version history, at your fingertips

When publishing your project to more of a ‘production’ environment, you can take a snapshot of your project (referred to as a Bundle) which acts as a distinct ‘deployable’ version of that project. Each project carries a history of its bundles and if something were to break in the latest production version of your project, reverting back to the previous bundle is a couple of clicks away.

Something broke in prod? Rollback to the previous working version.

Something broke in prod? Rollback to the previous working version.

Coding — If You Want!

Though both platforms pride themselves on their ease-of-use for the non-coder population, giving users the option to write and execute code provides that additional level of flexibility when needed. In Alteryx, SQL code can be executed using input tools or in-database tools, though there is no dedicated interface for code editing, and visual logic cannot be converted to SQL. Python and R are supported through specific tools, though these offer limited functionality compared to full coding environments.

When working with SQL datasets in Dataiku, you are able to view the SQL code that gets executed behind the scenes and can even convert the visual recipe into a code recipe if you wish to modify it further. Regardless of which recipe type you use, both visual and SQL recipes execute as SQL statements under-the-hood (when working with SQL datasets).

Build a Join recipe visually – then convert to SQL!

Build a Join recipe visually – then convert to SQL!

Now, perhaps you are already familiar with SQL and would like to port your code over or write your own queries directly in the platform. Dataiku’s SQL recipe gives you the ability to do just that — and with SQL Notebooks, you also have an interactive interface to prototype and adjust your code before committing it to the flow.

Experimenting to find the right SQL statement

Dataiku supports Python, R, PySpark, and other code recipes similarly. Users can test their functions in notebooks or their preferred IDEs before committing them to the recipe. Additional benefits of coding in Dataiku include code environments, containerized execution, project libraries, and web apps (but we will save these topics for another time).

Making the Switch

Just like Alteryx, Dataiku acts as a single access point to all the data you’ll ever need, coupled with the ability to store your results just about anywhere. But once you try Dataiku, I know you’ll come to love how easy it is to find trusted data, explore it and share it with your colleagues.

You May Also Like

Moving Beyond Guesswork: How to Evaluate LLM Quality

Read More

Navigating Regulations With Dataiku’s Governance Capabilities

Read More

Custom Labeling and Quality Control With Free-Text Annotation

Read More

Get to Know NYC and Paris From the Point of View of an Algorithm

Read More