Live TechTalk

Join experts from Google Cloud:  How to Scale Data Integration to and from Google BigQuery: Thursday, May 30th, 2PM EST/ 11AM PST


Data Engineering Best Practices

As Analytics, ML, and AI have become mainstream they have driven a tremendous increase in data, data systems, and data users. New tools, technologies, and companies are constantly emerging to meet the increased demand for faster, cheaper, and easier data storage, processing, and analysis. As data engineers navigating, designing, and implementing in this ever-growing field, it is important to follow industry best practices and not reinvent the wheel. This article will discuss the six most helpful data engineering best practices to stay current and ensure operational efficiency.

Best Practice Benefits
Design Efficient and Scalable Pipelines Lowers development costs and lays the groundwork for scaling up
Be Mindful of Where the Heavy Lifting Happens Avoids the repetition of costly tasks and allows the selection of the right ETL solution
Automate Data Pipelines and Monitoring Shortens debugging time and ensures data freshness and adherence to SLAs
Keep Data Pipelines Reliable Helps with making decisions based on trustworthy data
Embrace DataOps Increases development efficiency and provides faster time to insights
Focus on Business Value Improves the user experience and key business metrics, increasing return on data investment

Top Six Best Practices in Data Engineering

Design Efficient and Scalable Pipelines

Start with a simple pipeline design and keep it simple as long as possible. It can naturally become complex over time, but don’t let it get complicated. As beautifully put in “The Zen of Python”:

> import this
The Zen of Python, by Tim Peters
Simple is better than complex.
Complex is better than complicated.

Efficiency comes from using the right tools and techniques. Start by picking a batch, streaming, or hybrid solution appropriate for the business goals and systems involved. 

Don’t manage or develop connectors on your own for loading and unloading data. Use a managed data engineering platform that supports dozens of connectors for standard file formats, open-source tools, and third-party vendors. Professionally built connectors can be far more reliable in authentication management, capable of parallelized ingestion, and resilient to errors with their retry mechanics – all things that can take months of development and maintenance work. Moreover, the best of these are bi-directional giving complete flexibility in where to read or write data.

Identify bottlenecks when dealing with increasing data volume, velocity, or variety. Building atomic and decoupled tasks helps scale various parts of the pipeline independently. For example, instead of performing multiple transformations on data in the same pipeline, breaking it down into several smaller, simpler tasks enables orchestration tools to run tasks in parallel, reducing overall pipeline runtime and yielding faster time to analytics.

What is the impact of GenAI on Data


Pay Attention to Where the Heavy Lifting Happens

“Heavy lifting” refers to pipeline steps involving costly operations using significant computing and storage resources; an example is joining multiple large files or tables to generate aggregate analytics. Follow these best practices to reduce their impact:

  • Isolate resource-heavy operations from the rest of the pipeline, improve their resiliency, and persist their output, so when downstream jobs fail, you don’t have to repeat the costly operations. 
  • Don’t operate on rows one by one, especially when working with large data sets.
  • Pick the appropriate pipeline method: ETL (extract, transform, and load) or ELT, which puts the transform last. Use ETL to ensure that the data in the warehouse is in good shape, PII safe, and erroneous or unnecessary data is filtered out. Use ELT if you want to keep the raw data in the warehouse and meet unforeseen transformation needs quicker. Either way, build a single source of truth. 
  • When generating high-quality data in the data warehouse using significant resources, it is often a good idea to make this valuable data accessible to the broader organization. This can be done in the form of standardized data-products. Another way is to push data back to standard applications through API calls. This is called Reverse ETL and is implemented by referring the data in the warehouse back to operational systems such as CRM, marketing, and finance. 



A basic ETL diagram

Is your Data Integration ready to be Metadata-driven?

Download Free Guide

Automate Data Pipelines and Monitoring

Automation is sometimes confused with simply triggering a pipeline based on a schedule, but this is only part of the process. Triggering pipelines doesn’t have to be just based on time: There are also event-based triggers, such as HTTP requests, file drops, new table entries, or even a particular data record in an event stream. The best practice is to build event-based triggers whenever appropriate instead of setting up a schedule and hoping that everything the pipeline needs is ready on time. 

Parametrize pipelines to enable code reuse for different dates and other arguments. Sometimes temporary network and disk issues can disrupt a running pipeline, so adding automated retries—preferably with a backoff time of a few minutes—can automatically resolve such issues. 

Note that pipelines can increase in complexity naturally. To handle this, ensure that all the dependencies are checked and resolved when running a pipeline. Use orchestration tools with dependency-resolution features that help visualize the pipeline and update individual task statuses. 

Finally, monitor your pipelines continually. Capture and log all errors and warnings; never pass them silently. If feasible, extend the automation tool of choice with your error and warning messages. In case of failure, automatically create a monitoring ticket and assign it to team members who are responsible or on call.

Keep Data Pipelines Reliable

Maintaining data reliability is hard, and once a data pipeline is live, applications that consume the data quickly create new downstream dependencies. Schema changes, such as adding, removing, and renaming columns occur as business requirements evolve. The best practice is to build pipelines that are resilient to such schema changes, a.k.a. schema drift, but this can be complicated to implement. Look for options in your data pipeline tool that allow for and automatically handle schema drift. Advanced tools built with a data fabric architecture are able to handle schema changes dynamically and notify users of breaking changes avoiding erroneous processes that can be hard to unravel.

Another strategy to build resilient pipelines is incorporating the ability to handle and quarantine errors. You never want to send erroneous data downstream because cleaning it up afterward can be quite costly: Especially if it is a real-time or near-real-time pipeline, business decisions might be made based on erroneous data. To prevent such situations:

  • Perform data validation and data quality checks.
  • Automatically stop the pipeline or filter out erroneous records when an issue is detected.
  • Notify downstream users and applications about errors and potential delays. 
  • Use appropriate instruments to debug errors and determine the root cause.

Embrace DataOps

DataOps describes the process of managing data, which includes everything: collecting, storing, processing, and analyzing it. DataOps also encompasses the people and tools doing these activities. While the term is relatively new, the underlying concepts and best practices are established, particularly in software development. This is reflected in the similar term DevOps, and we will draw a few parallels between the terms to give you a point of reference. 

DataOps treats data the same way DevOps treats software code. It is often claimed that data is a business asset, so it makes sense to treat it like other precious physical assets that are tagged, tracked, and quality-assured, and used in well-defined processes. Following the “data as a product” principle enables domain-specific teams to build their own data pipelines and share data across multiple data silos, platforms, and applications in a secure and well-regulated manner. 

The best practice in this approach is ensuring interoperability, which means having teams follow common domain-agnostic data standards. Managing and guiding multiple teams toward the same standards can be quite challenging. Investing in a data platform that supports interoperable and domain-driven data products enables subject matter experts, data engineers, and AI/ML teams to work together to build data assets that are discoverable, secure, and trustworthy.

The second parallel is continuous improvement, which DevOps refers to as continuous integration and continuous delivery (CI/CD). This process involves creating a continuous pipeline from data ingestion to data consumption by an application. As mentioned earlier, it’s best to start small and build pipelines incrementally—don’t change too many parts all at once, which makes rollbacks difficult. Test all the changes and ensure that code quality is not sacrificed due to quick delivery requirements. Building dev pipelines and then promoting it to production is also part of the data engineering process. It helps to choose pipeline tools that are compatible with CI/CD tools such as Jenkins. That’s because everything you have learned in CI/CD of applications can now apply to CI/CD of data pipelines. 

Guide to Metadata-Driven Integration


Learn how to overcome constraints in the evolving data integration landscape

Shift data architecture fundamentals to a metadata-driven design

Implement metadata in your data flows to deliver data at time-of-use

The third parallel between data engineering and DevOps best practices has to do with security where the core tenets are:

  • Keep your secrets and data store credentials out of the code. 
  • Use secrets managers and vaults to store encrypted keys. 
  • Safely store credentials in credential stores, and use managed identities wherever supported. 

Focus on Business Value

The purpose of building data pipelines, data warehouses, or lakehouses is the means to the end of supporting business decision-making. The best practice is to survey or interview customers to learn what they want, how they behave, and what they value, so data teams can create products, services, and experiences that delight end-users. Understand not just how data is transformed in specific ways along the journey but why it’s being transformed and what business questions the transformed data is meant to answer. 

Failing to keep the strategic context in mind can lead to “shiny object syndrome,” where data professionals chase one trendy tool after another regardless of how valuable or helpful it is.

Empowering Data Engineering Teams

Free Strategy


Data Extraction

Data Warehousing

No-Code Automation

Auto-Generated Connectors


Multi-Speed Data Integration




Learn More About Data Engineering Best Practices

Data engineering is an evolving and fast-paced field with new tools and technologies regularly emerging. The best practices in this guide represent countless hours spent by data professionals designing and implementing data processing solutions. Check out the chapters below to delve deeper into the art and science of data engineering and avoid common mistakes.

Chapter 1: Data Pipeline Tools

Learn the key features to look for in a data pipeline tool like integration count, scalability, auditability, automatability, monitoring, and more

Chapter 2: Kafka for Data Integration

Learn the benefits of using Kafka for data integration such as extensive and easy data routing, flexible data ingestion, durability, fault tolerance, and more

Chapter 3: What is Data Ops?

Explore the key components of Data Ops in an enterprise and learn about the most common use cases. Implement the right solution using Infrastructure-as-Code.

Chapter 4: Redshift vs Snowflake

Redshift and Snowflake both solve the fundamental problem of storing and processing data at scale yet they take different approaches that are charged based on usage.

Chapter 5:AWS Glue vs. Apache Airflow

AWS Glue and Apache Airflow offer some overlapping functionalities. Yet both are designed to solve entirely different problems. Explore why organizations choose one over the other. 

Chapter 6:Data Mesh: Tutorial, Best Practices, and Examples

Learn the core principles of data mesh, follow an example applying those principles, and follow the best practices to start your own implementation

Chapter 7:Data Connectors: Common Types, Benefits, & Use Cases

Learn how to automate data ingestion into modern analytics systems to drive data-driven decisions.

Chapter 8:Data Management Best Practices: Challenges & Recommendations

In-depth overview of five top challenges of data management – integration, automation, quality, security, and analysis – and the best practices to tackle them and mitigate their associated risks.

Chapter 9:Automated Data Integration: Concepts & Strategies

Automation plays an essential part in every stage of a data integration pipeline. This article will cover five data integration concepts, discuss associated best practices, and present the benefits of automation.

Chapter 10:Data Products: Transforming Data Delivery

Learn how data products revolutionize the way data-driven assets and insights are delivered and consumed.

Like this article?

Subscribe to our LinkedIn Newsletter to receive more educational content

Subscribe now