Introducing Soda: Proactive Data Monitoring

Soda is a data monitoring platform designed to identify potential problems within your data processing systems. This allows for swift responses and ensures a complete and accurate data landscape.

The Rise of Data-Driven Businesses

Companies operating with a digital-first approach, and their customers, typically generate substantial volumes of data. This data is frequently utilized to refine and improve products – consider dynamic hotel pricing, restaurant recommendations on delivery platforms, or loan applications processed by fintech firms. These are all examples of data-intensive applications.

The Modern Data Platform

As Maarten Masschelein, co-founder and CEO of Soda, explained, “Companies construct what they refer to as a data platform, typically within one of the major cloud providers – Amazon Web Services, Google Cloud, or Microsoft Azure.” Data is then deposited into these platforms and made accessible for analytics and other purposes.

Utilizing Data Lakes and Warehouses

Data lakes and data warehouses are then leveraged to power analytics displays, data visualizations, and service monitoring. However, what occurs when disruptions arise within your data workflows?

The Impact of Data Issues

Detecting missing data or inaccuracies can often be a delayed process. A notable example is Facebook’s miscalculation of average video view times over several years. Such issues can significantly impact critical business functions when they are discovered.

Soda’s Approach to Early Detection

Soda aims to identify data issues promptly through automated and scalable monitoring. Masschelein notes, “We position ourselves further upstream, closer to the data’s origin.”

Real-Time Alerts and Anomaly Detection

Upon initial setup, Soda provides immediate alerts when anomalies are detected. For instance, a sudden drop in record generation – from a typical 24,000 records per day to only 6,000 – signals a potential problem. Similarly, a pause in data entry, such as a 15-minute gap when entries are usually received every minute, indicates data staleness.

Beyond Basic Anomaly Detection

“However, this represents only a fraction of the potential data issues,” Masschelein points out. “There’s a greater degree of logic that requires testing and validation.”

Data Validation Rules and Test Suites

Soda enables the creation of custom rules to test and validate data. This concept is analogous to test suites in software development, where code undergoes rigorous testing before a new version is released to prevent critical failures.

Immediate Feedback and Automated Responses

Soda provides immediate test results. If a test fails, automated responses can be triggered, such as halting a process or isolating affected data.

Introducing Soda Cloud: Collaborative Visibility

The startup is also launching Soda Cloud, a web application designed to provide organizational-wide visibility into data flows. This allows individuals without technical expertise to easily examine metadata and confirm the proper functioning of data pipelines.

Soda SQL and Soda Cloud: A Combined Solution

Soda customers utilize Soda SQL, a command-line tool for data scanning, in conjunction with Soda Cloud, a web application for viewing the results generated by Soda SQL.

The Future of Data Tools

Soda envisions data as an emerging category within software products. Development teams benefit from a wealth of tools for automated testing, integration, deployment, and version control. There is significant potential for tools specifically tailored to the needs of data teams.

Recent Funding

Soda has recently secured $13.5 million in Series A funding (€11.5 million), led by Singular, a Paris-based venture capital firm. Existing investors, including Point Nine Capital, Hummingbird Ventures, DCF, and various angel investors, also participated in the round.