Subscribe to Folding Data

What it means to be at the frontier of velocity and quality in data work

The next frontier for dbt, launching our open source data-diff & dbt integration, and Data Quality Meetup is back!

dbt + python = ?, ChatGPT & are we out of a job yet, and our virtual hands-on lab!

Contracts vs. handshakes, liberating data, Data Quality Meetup is back

The Airflow problem, learning data through beautiful design, ESports as in Excel

Never get alerts, Never break Hightouch, Never wait for queries

Open sourcing data diff, why SQL can’t tell you if the chicken or the egg came first, and how Hex continues to become more magical

We data people need better tools, and a better quality of life. But as mighty as dbt Labs is, we can't expect them to solve all of our needs.

Data Quality vendors are typically focused on monitoring for anomalies. Datafold is focused on preventing anomalies.

Software engineering teams and data teams need to collaborate better to ensure high quality data

Manually creating assertion tests for data has diminishing returns. It is a great first step to ensuring higher data quality but then you need automated tooling to do regression testing to deliver high quality data at scale.

Data activation represents a huge shift in where data is being stored that needs to be accessed by operational tools.

Data is complicated and often takes significant analysis to become confident enough to report an accurate metric.

Analytics Engineering is a new discipline in data originated by dbt. It is growing quickly and there are many new courses to help teach the skills now.

Data team and data quality have been evolving from reactive to proactive.

Data moves between many systems and is transformed through various pipelines which makes it hard for a single tool to guarantee the data quality.

These times are challenging for our colleagues and their family members in both countries

With many new data quality tools in the market, we should focus on productivity rather than just observability.

Every week, we’ve shared an interesting read. Here are the ones got the most attention from our readers in 2021

Companies should treat data as products to prevent analytics from becoming difficult to use.

As the Modern data stack evolves its important to keep iterating on how your data team operates

Data science promises to help us scales decision making, but there are still some concerns to be aware of.

Price efficiency and performance of major data warehousing platforms are likely to converge over time, and the true differentiating factor will be the end-user experience and data ops functionality.

A realistic peek at what it's like to experience multiple stages of a hyper-growth environment at one of the foundational companies in the data space.

The modern data stack has had a meteoric rise in the data community, but it's good to check in with Gartner to see where the hype is at from an outside perspective.

Beyond learning SQL, there are many great data communities to join and many professionals to learn from.

Continuous Deployment for fast incremental updates and Continuous Delivery for releasing multiple changes at once is critical to building quality dbt models

While the data ecosystem lags software's evolution it is rapidly accelerating.

Column level lineage is a great first step to understanding what data needs to be protected or anonymized.

Cognitive biases are easily accidentally baked into models and pressure to hit certain metrics can cause data scientists to make poor feature decisions.

As data grows in size and the more data is being utilized for operational use cases finding issues quickly is key.

Data Engineering is one of the fastest growing jobs in tech.

BI has seen a bunch of new exciting startups, but sometimes all we want is SQL, markdown, and open source.

AB Testing allows companies to optimize their products in a statistical way at scale.

Here’s a guide if you’re planning to build your ideal data stack using only open-source technologies.

Hera ate the best data quality tools and how they facilitate quality data at every step.

Every data analyst has to communicate their findings which requires data story telling

When data isn't maintained it becomes less useful which causes people to lose trust which causes them to use data less.

While many companies are adopting the modern data stack many still need column level lineage to be data driven.

There are now many data science notebooks that are far more powerful and collaborative than your plain old Jupyter notebook.