2020 Gartner Critical Capabilities for Data Science and Machine Learning Platforms: Key Takeaways

2020 Gartner Critical Capabilities for Data Science and Machine Learning Platforms: Key Takeaways

Analyst house Gartner, Inc. has released its 2020 Critical Capabilities for Data Science and Machine Learning Platforms, a companion research to the popular Magic Quadrant report. Used in conjunction with the Magic Quadrant, Critical Capabilities is an additional resource which can assist buyers of data and analytics solutions in finding the products that best fit their organizations.

Gartner defines Critical capabilities as “attributes that differentiate products/services in a class in terms of their quality and performance.” Gartner rates each vendor’s product or service on a five-point (five points being best) scale in terms of how well it delivers each capability. Critical Capabilities shows you which products are best for each use case and includes a comparison graph for each, along with in-depth descriptions on the various points of comparison.

The study highlights 16 vendors Gartner considers most significant in this software sector and evaluates them against 15 critical capabilities and now four use cases prevalent in the space, including:

  • Business exploration
  • Advanced prototyping
  • Production refinement
  • Augmented data science and machine learning

The editors at Solutions Review have read the report, available here, and pulled out three key takeaways.

Augmented data science is here to stay

In covering last year’s Critical Capabilities report, we wrote: “Data and analytics leaders are recommended to incorporate data science and machine learning into their BI strategies. However, given the popularity of open source data science software, Gartner tells buyers to avoid long-term lock-in with commercial product vendors.” The 2020 edition tells us that augmented data science and machine learning is “getting better by the day” at enabling citizen data science and improving the productivity of expert software users.

In this light, Gartner recommends evaluating data science and machine learning platforms by working with data science teams and business units to identify challenges, needs and the impact of software selection. Further, the best-fit platform will almost always balance a desired mix of data science use cases and the deployment environment.

Dataiku leads the way in two-of-four use cases

Dataiku offers an advanced analytics solution that allows organizations to create their own data tools. The company’s flagship product features a team-based user interface for both data analysts and data scientists. Dataiku’s unified framework for development and deployment provides immediate access to all the features needed to design data tools from scratch. Dataiku released Data Science Studio 7 in March, adding new prediction explanations and data science features to the platform.

Dataiku was awarded the top scores for business exploration and augmented data science and machine learning. What’s more is that the provider earned above-average scores for advanced prototyping and production refinement as well. Though Dataiku customers also considered RapidMiner, DataRobot, Alteryx and Databricks, Gartner sample users speak highly about Data Science Studio’s ease of use, speed of model development, and the ability to manage a large number of models.

SAS nets a top-three score in three-of-four use cases

The top scoring vendor for advanced prototyping, SAS also finished runner-up to Databricks for production refinement, and earned the third-best score for business exploration. The data science major player also earned above-average scores for augmented data science and machine learning. Gartner reference customers scored SAS Visual Data Mining and Machine Learning (VDMML) as excellent for user interface and data access. The tool also earned the top scores for model management and machine learning.

SAS Visual Data Mining and Machine Learning automatically generates insights for common variables across models. It also features natural language generation for creating project summaries. SAS Model Manager enables users to register SAS and open source models within projects or as standalone models.

Read Gartner’s Critical Capabilities for Data Science and Machine Learning Platforms.

Timothy King
Follow Tim