Dataiku vs. SAS Enterprise Guide

Dataiku DSS is very well suited to handle large datasets and projects which requires a huge team to deliver results. This allows users to collaborate with each other while working on individual tasks. The workflow is easily streamlined and every action is backed up, allowing users to revert to specific tasks whenever required. While Dataiku DSS works seamlessly with all types of projects dealing with structured datasets, I haven't come across projects using Dataiku dealing with images/audio signals. But a workaround would be to store the images as vectors and perform the necessary tasks.

Incentivized

For writing out longer code creation for shaping data on complicated reports, the clean UI is helpful. If exploring data though, SAS Studio would be better suited given its easier interface for GUI graph building.

Incentivized

• Low-code platform.
• Open source version includes most valuable modules.
• User friendly documentation.

Incentivized

• It can load a huge amount of data as compared to R Studio and Excel.
• Data processing speed is very fast, millions of records are loaded into this software very easily and data manipulation is also very easy.
• Inbuilt Statistical functions and procedures make it very comfortable to use for non analytics professionals as well.

• The visualization feature of flow still has a lot room to improve, when the flow is complex.
• The "non-coding" template/building block for deep learning lack of many important configurable parameters.
• Lack of the unified way to allow applying the "design pattern" on the Python codes (if we want to develop our own module or building blocks.

Incentivized

• I would like to see advance interactions with external databases to be able to kill ongoing queries from SAS. As of now, you can stop pretty much any ongoing process besides the one running on a remote database (killing SAS/EG doesn't stop the remote process)
• When creating prompts for programs, it would be nice to be able to have conditional prompts (based on the selection of other prompts). The prompts are clearly a recent feature and constantly under development but I wish it would be more powerful.
• More of a SAS metadata issue but when loading SAS/EG (first connection to the server), it takes a few seconds which feels like a long time. I really don't understand why the initialization of the session can take so long. Don't get me wrong, this has no real impact on productivity but that 10s delay just feels really like eternity when you want to run some code in a new session.

On account of current user experience and the organization-wide acceptance.

The user experience is very good. Everything feels intuitive and "flows" (sorry excuse the pun) so nicely, and the customization level is also appropriate to the tool. Even as a newer data scientist, it felt easy to use and the explanations/tutorials were very good. The documentation is also at a good level

Incentivized

It's not all bad, but I don't believe that an enterprise purchase of SAS is worth the expense considering the widely available set of tools in the data analytics space at the moment. In my company, it's a good tool because others use it. Otherwise, I wouldn't purchase a new set of it because it doesn't have some of the better analytical functions in it.

Incentivized

The amazing part of Dataiku DSS is their customer service. Based on urgency and technical level, you get a reply from the Dataiku engineer when you raise a query. So far, my queries have been pretty complex to solve, so I have received solutions even from the CTO of the company as well, which is why I would describe their customer support as very good.

Incentivized

Although I use SAS support for information on functions, these are SAS related and haven't really come across anything that is specifically for SAS EG.

Incentivized

I've not worked hands-on with the implementation team, but there were no escalations barring a few hiccups in the deployment due to change in requirement & adoption to our company's remote servers.

Dataiku was selected for me, but I am happy about that. I like Dataiku for the user experience, it feels less code-y and I like to demo things to non technical stakeholders because they can still follow along. When you open some other notebooks, you can see that peoples eyes start to glaze over

Incentivized

Python-based platforms like Pandas or Spark are very good too at displaying data and do exploratory analysis. I definitely prefer them to SAS EG. It's just too slow, and doesn't let you peek into the data very easily. Lots of clicking, and I'd rather just write some code, rather do clicking.

Incentivized

• So far it has had a positive impact. Multiple departments are coming to us with their business problems.
• I can't specifically say about ROI as I'm a developer, though I have heard this solution is economical compared to other AI/ML enterprise tools.
• By using this tool, my client has let go of software that was used earlier, and we have created a simpler framework to replace that software.

Incentivized

• Faster decision making, through powerful big data handling functionalities.
• Faster operations on daily basis, once the project tree is built, unskilled personnel can use it in their daily operation.
• Don’t need to choose SAS EG if you are not going to be handling big data. (such as over 1 million rows and 50 columns)
• You need skilled personnel to build the initial project tree.

Incentivized