Datacoves blog

You now know what dbt (data build tool) is all about.  You are being productive, but you forgot what `dbt build` does or you forgot what the @ dbt graph operator does. This handy dbt cheat sheet has it all in one place.

dbt cheat sheet - Updated for dbt 1.8

With the advent of dbt 1.6, we updated the awesome dbt cheat sheet created originally by Bruno de Lima

We have also moved the dbt jinja sheet sheet to a dedicated post.

This reference summarizes all the dbt commands you may need as you run your dbt jobs or study for your dbt certification.

If you ever wanted to know what the difference between +model and @model is in your dbt run, you will find the answer. Whether you are trying to understand dbt graph operators or what the dbt retry command does, but this cheat sheet has you covered. Check it out below.

Primary dbt commands

These are the principal commands you will use most frequently with dbt. Not all of these will be available on dbt Cloud

dbt Command arguments

The dbt commands above have options that allow you to select and exclude models as well as deferring to another environment like production instead of building dependent models for a given run. This table shows which options are available for each dbt command

dbt selectors

By combining the arguments above like "-s" with the options below, you can tell dbt which items you want to select or exclude. This can be a specific dbt model, everything in a specific folder, or now with the latest versions of dbt, the specific version of a model you are interested in.

dbt graph operators

dbt Graph Operator provide a powerful syntax that allow you to hone in on the specific items you want dbt to process.

Project level dbt commands

The following commands are used less frequently and perform actions like initializing a dbt project, installing dependencies, or validating that you can connect to your database.

dbt command line (CLI) flags

The flags below immediately follow the dbt command and go before the subcommand e.g. dbt <FLAG> run

Read the official dbt documentation

‍

‍

As a managed dbt Core solution, the Datacoves platform simplifies the dbt Core experience and retains its inherent flexibility. It effectively bridges the gap, capturing many benefits of dbt Cloud while mitigating the challenges tied to a pure dbt Core setup. See if Datacoves dbt pricing is right for your organization or visit our product page.

Please contact us with any errors or suggestions.

I read an article on Anchor Data Modeling, more specifically, Atomic modeling where the author proposes a different way of Data Modeling. The rationale for this change is that there is a lack of skills to model data well. We are giving powerful tools to novices, and that is bound to lead to problems.

From the article:

"we are in a distressful situation both concerning the art as a whole but also its place in modern architectures"

Is this the case? Do we have a big problem on the horizon that requires us to make this big shift?

I'd say I am open-minded and expose myself to different ways of thinking so I can broaden my views. A few years ago, I learned a bit about COBOL, not because I had any real use for it but because I was curious. I found it very interesting and even saw its similarities with SQL. I approached the topic with no preconceived ideas; this is the first time I heard of Atomic Modeling.

Is there something wrong with atomic and anchor data modeling in general?

The issues I see with ideas like Atomic data modeling are not in their goal. I am 100% aligned with the goal; the problem is the technology, process, and people needed to get there.

What we see in the market is a direct result of a backlash against doing things "perfectly." But why is this the case?  I believe it is because we haven't communicated how we will achieve this vision of ideas like atomic data. The author even says a key phrase in the first paragraph:

"practitioners shying away from its complexity"

If doing anchor data modeling is "complex" how are we going to up-skill people? Is this feasible? I am happy if I can get more people to use SQL vs a GUI tool like Alterix 😁

I am by no means an expert. Yet, I am fairly technical, and if I am not convinced, how will we convince decision-makers?

As I read this article, here's what I am thinking:

1. First, I will need to deconstruct the data I get from some source like material master data form SAP. That will be a bunch of tables, and who is going to do all this data modeling? It sounds expensive and time-consuming.

2. I am going to need some tooling for this, and I am either going to build it or use something a few others are using. Will my company want to take a chance on something this early? This sounds risky.

3. After I deconstruct all this data, I need to catalog all these atoms. I now have a lot of metadata, and that's good, but is the metadata business-friendly?  We can't get people to add table descriptions how is this going to happen with this explosion of objects? Who will maintain it? How will we expose this? Is there a catalog for it already? Does that catalog have the other features people need? It sounds like I need to figure out a bunch of things, the biggest one being the change management aspect.

4. What sort of database will I use to store this? This is a great use case for a graph database. But graph databases are not widely adopted, and I have seen graph databases choke at scale. We can use a relational database, but these joins are going to be complex.  Someone may have figured all this out, but there's more tech and learning needed. It sounds like this will also add time and expense.

5. When I have managed to do all the above, I will need to construct what people can actually use. We need objects that work with tools that are available. I need to make relational tables I can query with SQL and viz tools, which are more user-friendly. This sounds like more work, more time, and more money.

I may have missed some steps and oversimplified what's needed for this type of change. I am also aware that I may not know what exists to solve all the above. However, if I don't know it, then there are a ton of other people who also don't know it and this is where we need to start. We need to understand how we will tactically achieve this "better" world.

What are we fixing?

I've had conversations on metadata-driven automation, and like atomic modeling, I am not clear on who we are helping and how. What are we improving and in what timeframe? In the end, it feels like we have optimized for something only a few companies can do. To do anchor modeling well would be a huge expense, and when things go wrong, there are several points of failure. When we look at business problems, we need to be sure to optimize the end-to-end system. We can't locally optimize one area because we are likely moving the problem somewhere else. This can be in terms of time, money, or usability.

Decision-makers are not interested in data modeling. They are expecting results and a faster time to market. It's hard enough getting people to do things "better." This is why I find it hard to imagine that we can get to this level of maturity any time soon.

What can we do instead?

There are incremental steps we can take to incorporate best practices into the modern data stack. We need to help people mature their data practice faster, and we should not let perfection get in the way of good. Most companies are not large enterprises with millions of dollars to spend on initiatives like atomic modeling. That being said, I have yet to see anchor modeling in practice, so I welcome the opportunity to learn. I remember years ago the debates about how Ruby on Rials was teaching people "bad practices."  The other side of that argument is that Rails helped companies like Twitter and Github launch businesses faster. Rails was also better than the alternative at the time, which included messy PHP code. Others advocated for well-crafted "scalable" and expensive Java applications. Rails may not be the powerhouse it once was, but it has had a huge influence on how we build software. I even see its influence in dbt even if it might not have been intentional or direct.

Conclusion

Tools like Snowflake and dbt allow us to build processes that are much better than what most people have. Should we focus on all the "bad" things that may come with the modern data stack? Should we focus on how practitioners are not well educated, and so we need to throw all they are doing out?

I don't think so; I believe that we can help companies mature their data practices faster. Will we have the best data models? Maybe not. Will users do things perfectly? Nope. But can we help them move faster and guide them along their journey to avoid big pitfalls? I think we can. Getting people to use git, automating testing, and creating DataOps processes is a huge step forward for many organizations. Let's start there.

There's a reason Data Mesh and the Modern Data Stack resonate with so many people. There's a desire to do things faster with more autonomy at many companies, not just the ones with multi-million-dollar budgets. Let's focus on what is achievable, do the best we can, and help people mature along the way. We don't need more complexity; we need less.

Using dbt with Snowflake is one of the most popular and powerful data stacks today. Are you facing one of these situations:

- You just joined a data team using dbt Core with Snowflake, and you want to set up your local dbt development environment

- Your company is already using Snowflake and wants to try out dbt.

- You are using dbt with another data warehouse and you want to try dbt with Snowflake before migrating.

If you are facing any of these, this guide will help you set up your dbt environment for Snowflake using VS Code as your dbt development environment. We will also cover useful python libraries and VS Code extensions that will make you more productive. Want to become a dbt on Snowflake ninja, keep reading.

Pre-requisites for dbt development on Snowflake

To get started, you need to set up your dbt development environment. This includes Python, dbt, and VS Code. You will also need to set up your Snowflake account properly so dbt can work its magic.

While dbt supports versions of Python greater than 3.7, some other tools like Snowpark require Python 3.8, therefore we recommend you stick with that version. You can find the installer for your particular Operating System on this page. Finding the right Windows Installer can be confusing, so look for a link titled Windows installer (64-bit). If you are on a Mac, you can use the Universal Installer macOS 64-bit universal2 installer.

When using dbt, you will also need Git. To setup git, follow this handy guide.

The preferred IDE for dbt is VS Code, you can find it on the official Microsoft site. click the big Download button and install like any other app.

Installing dbt is done using pip. You can find more information on this page.

For dbt Snowflake, simply run:

This will install the latest version of dbt core along with the dbt adapter for Snowflake. If you need to use an older version of dbt, you will need to specify it when you run pip. However, the version of the dbt adapter may not match the version of dbt core.

For example, as of this writing, the last version of dbt core 1.3.x is 1.3.4. However, dbt-snowflake for 1.3.x is 1.3.2.  So, if you want to install the latest dbt 1.3.x for snowflake, you would run

This will install dbt-snowflake 1.3.2 along with dbt-core 1.3.4

Configure Snowflake permissions

The final pre-requisite you will need to do is set up a Snowflake user that has been granted a role with the right access to a database where dbt will create views and tables. The role should also have access to a Snowflake Warehouse for compute. Here is a handy guide that gives you the basics. We would recommend a more comprehensive setup for a production deployment, but this will get you going for now.

The key items that are not covered in that guide is that you should create a role ANALYST and a database ANALYTICS_DEV and grant OWNERSHIP of that database to the ANALYST role. The ANALYST role should also be granted USAGE on the TRANSFORMING warehouse.  You also need to grant the ANALYST role to your specific user. Don’t run as ACCOUNTADMIN.  

This is all needed because when dbt runs it will create a schema for your user in the ANALYTICS_DEV database and you will use the TRANSFORMING warehouse when compute is needed like when dbt creates tables.

If all of this seems confusing or tedious, you should consider a managed dbt solution like dbt Cloud or Datacoves. For more information, checkout our in-depth article where we compare dbt cloud vs dbt core as well as managed dbt core.

Now with all the pre-requisites out of the way, let’s configure dbt to connect with Snowflake.

Configure your dbt Snowflake profile

To initialize your dbt project in Snowflake, dbt has a handy command dbt init. You can configure your dbt Snowflake profile using the dbt init command both for a new and a existing dbt project. First you will need to clone your repo using git. Then, simply run the dbt-init command and go through the prompts.  

Once you get your project set up, consider adding a profile_template.yml to your project. As stated on that page, using a profiles template will simplify the dbt init process for users on your team.

To make sure dbt can connect to Snowflake, run dbt debug. If dbt can connect to your Snowflake account, you should see “All checks passed!” If you have problems, then join the dbt Community search the forums or ask a question in the #db-snowflake channel.

Even though dbt performed the setup of your profile.yml to connect to Snowflake with your credentials, it only provides the basic setup. This page provides additional parameters. that you can configure for the Snowflake connection in your profiles.yml file.  

If you want to configure those parameters, you will need to open and edit the profiles.yml file. The profiles.yml file created by dbt init will be in your user’s home directory in a subdirectory called .dbt.

One handy configuration parameter to change is reuse_connections. Also, if you use SSO authentication with external browser, you should consider setting up connection caching on Snowflake, otherwise you will be prompted to authenticate for every connection dbt opens to the database.

Now that you have set up your dbt connection to Snowflake, there are some other options you can configure when dbt runs against your Snowflake account. This includes overriding the default warehouse for a specific model, adding query tags, copying grants, etc. This handy page has a lot more information on these dbt snowflake advanced configurations.

Improve your dbt Snowflake experience with SQLFluff and other Python libraries

Now that you have dbt connecting to your database, let’s talk about some python libraries you should set up to improve how you work with dbt.

dbt-coves

dbt-coves is an open source library created by Datacoves to complement dbt development by simplifying tedious tasks like generating staging models. It is a must-have tool for any dbt practitioner who wants to improve their efficiency. dbt-coves will automatically create your source.yml and staging models as well as their corresponding yml(property) files. It also has utilities for backing up Airbyte and Fivetran configurations.

SQLFluff

SQLFluff is a Python library for linting SQL code. SQLFluff seamlessly integrates dbt using a templater and it is the only linter compatible with dbt. If you have not heard of code linting it helps you enforce rules on how your SQL code is formatted for example, should everyone use leading or trailing commas, should SQL keywords be upper or lower case. We recommend everyone use a linter as this will improve code readability and long term maintainability.  

pre-commit with dbt-checkpoint

dbt-checkpoint is a tool that allows you to make sure your dbt project complies with certain governance rules. For example, you can have a rule that validates whether every dbt model has a description. You can also ensure that every column is documented among many other rules. We also recommend the use of dbt-checkpoint as it will assure developers don’t add technical debt from the start of a project.

In addition to these Python libraries, at Datacoves we set up the development environment with other handy libraries like Snowpark and Streamlit. We believe that flexibility is important especially in enterprise environments. If you want to learn what to consider when selecting a managed dbt core platform, check out our guide.

Improve your dbt Snowflake experience with dbt power user and other VS Code extensions

In addition to Python libraries, you can improve your dbt workflow with Snowflake by installing these VS Code extensions.

Snowflake VS Code Extension

The official Snowflake dbt extension keeps you in the flow by bringing the Snowflake interface to VS Code. With it you can explore your database, run queries, and even upload and download files to Snowflake stages. It is a must-have for any Snowflake user.

dbt power user

dbt power user is a VS Code extension that improves the dbt development experience by adding handy shortcuts to run dbt models, tests, and even let’s you preview the result of a dbt model or a CTE within that model.  

SQLFluff VS Code extension

The SQLFluff VS Code extension is the companion to the SQLFluff python library. It improves the development experience by highlighting linting errors right in line with your SQL code. It even has a handy hover which describes the linting error and links to the SQLFluff documentation.

There are many other great VS Code extensions and at Datacoves we are always improving the dbt developer’s experience by pre-installing them. One recent addition demonstrated on the video below is a ChatGPT extension that allows you to improve the dbt experience by writing documentation in addition to other functions.

Conclusion

Getting started with dbt and Snowflake is not hard and knowing how to streamline the development experience when working with dbt and Snowflake will maximize developer happiness.

Some users may run into issues configuring their development environment. If that happens, check out the #sqlfluff, #tools-dbt-libraries, and #tools-vscode channels on the dbt Slack community. There are many helpful people there always ready to help.

As you can see there are a lot of steps and potential gotchas to get a dbt environment properly configured. This gets more complicated as the number of dbt users increases. Upgrading everyone can also pose a challenge. These reasons and more are why we created the most flexible managed dbt-core environment. If you want your developers to be up and running in minutes with no installation required, reach out and we can show you how we can streamline your teams’ dbt experience with best practices from the start.

Since its inception, dbt has served the needs of small to mid-sized companies to transform data into actionable insights. Now as larger enterprises look to adopt dbt, they must consider the unique complexities of large enterprises. These include integrating with existing systems behind a corporate firewall, orchestrating end-to-end data pipelines, and implementing DataOps processes to transform how teams collaborate across the organization.

This article compares three popular dbt development and deployment options: leveraging dbt Cloud, using dbt Core alone, or using a dbt Core managed platform other than dbt Cloud. To learn more about specific dbt Cloud features and how to implement them using dbt Core using other open-source tools checkout dbt Core vs dbt Cloud - Key Differences.

dbt Cloud for your enterprise data management platform

dbt Cloud is a hosted environment to develop and deploy dbt Core projects. dbt Cloud leverages dbt Core, the open-source data transformation framework we all know as dbt. dbt Cloud pricing starts at $100 per developer for teams or if you are a single developer, they offer a free single seat with  limited features.

dbt Cloud IDE for dbt-core development

When companies are ramping up with dbt, one of the hardest parts is setting up and managing dbt IDE environments. Analytic Engineers coming to dbt may not be familiar with concepts like version control with git or using the command line. The dbt Cloud IDE simplifies developer onboarding by providing a web-based SQL IDE to team members so they can easily write, test, and refine data transformation code without having to install anything on their computers. Complexities like starting a git branch are tucked behind a big colorful button so users know that is the first step in their new feature journey.

Developers who are accustomed to more versatile local IDEs, such as VS Code, may find the dbt Cloud experience limiting as they cannot leverage extensions such as those from the VS Code Marketplace nor can they extend dbt Core using the vast array of Python libraries. An example of a common library employed in the dbt community is SQLFluff. With SQLFluff, teams can ensure a minimum level of code quality is achieved across developers. 

dbt Cloud for DataOps

DataOps is a set of practices that combines data management and software engineering to improve the quality, speed, and collaboration of data analytics teams. In an enterprise environment, implementing DataOps processes early on in a project's lifecycle is crucial.

dbt Cloud can be paired with Continuous Integration (CI) tools like GitHub Actions to validate data transformations before they are added to a production environment.

If CI integration checks include things like SQLFluff and dbt-checkpoint, there is currently no way for users to run these validations in dbt Cloud prior to executing the automated CI checks in GitHub Actions. This may lead to a lot of back and forth between the GitHub Actions run failures and the dbt Cloud IDE to fix and commit the changes. Aspects such as code reviews and approvals will occur in the CI/CD tool of choice such as GitHub and dbt Cloud can report job execution status back to GitHub Actions. This allows teams to know when it is safe to merge changes to their code.

Companies that have tools like Bitbucket, Jira, and Jenkins within their corporate firewall may find it challenging to integrate with dbt Cloud.

dbt Cloud for deployment 

dbt Cloud makes deploying a dbt Core project simple. It allows you to define custom environment variables and the specific dbt commands (seed, run, test) that you want to run during production runs. The dbt Cloud scheduler can be configured to trigger at specific intervals using an intuitive UI.

dbt Cloud is primarily focused on running dbt projects. Therefore if a data pipeline has more dependencies, an external orchestration tool may be required. Cases where this is necessary include when you need to connect the Extract and Load step to the Transformation happening in dbt. Without connecting these steps, you have no assurance that the load is complete before the transformation starts. This can be mitigated with freshness checks, but without an explicit connection between the load step and the transformation, you may be delaying transformation unnecessarily.

Fortunately, if you do use an external orchestrator, dbt Cloud offers an API to trigger dbt Cloud jobs from your orchestrator. Another reason an external orchestrator may be needed in an enterprise is when there are additional steps to run after the dbt transformation. These may include sending data to downstream BI tools, marketing campaign tools, MDM systems, or refreshing machine learning models. dbt Cloud recently announced a webhooks API that would allow for triggering steps after transformation. However, if these endpoints reside within a corporate firewall, this may not be an option and an orchestrator like Airflow may be necessary.

dbt Cloud extensibility

The dbt Cloud IDE does not allow for the installation of Python libraries and dbt Cloud cannot create and orchestrate Python scripts that leverage tools like SQLFluff (this is coming to dbt Cloud later this year), dbt-coves, dbt-fal, Snowpark, Permifrost, or Streamlit. Additionally, the dbt Cloud IDE cannot be extended with extensions such as the official Snowflake or Databricks VS Code extensions.

While dbt Cloud can send notifications via email and Slack, it does not support other popular enterprise systems such as MS Teams.

dbt Cloud security 

dbt Cloud offers industry-standard SaaS security appropriate for any enterprise. At the enterprise level, it offers Single Sign-On and Role Based Access Controls. Enterprises can also choose to have dbt Cloud provisioned on a dbt Labs managed private cloud account. dbt Cloud is only offered as a managed SaaS solution on AWS and thus cannot be deployed to a company's existing private cloud account.

Who should consider using dbt Cloud

dbt Cloud is especially attractive for organizations with limited in-house technical resources or those seeking to minimize the time and effort spent on initial setup, maintenance, and upgrades. dbt Cloud runs in the browser and thus removes the need for manual installation and setup of dbt components, allowing your team to focus on data transformation tasks. dbt Cloud handles updates and upgrades for you, so it is simple to stay up-to-date and secure. dbt Cloud includes built-in scheduling capabilities, allowing you to automate dbt runs and testing without relying on external schedulers or orchestration tools.

dbt Cloud may introduce additional complexities in large enterprises with more complex orchestration needs or those that need to host the environment within their private cloud. Finally, the dbt Cloud IDE may feel constraining for users who prefer a VS Code environment. 

All these features can in turn reduce the time to value in setting up an end-to-end enterprise data management platform. It is possible to start out with a simple setup, but inevitably not accounting for the breadth of the entire platform may create technical debt and cause rework in the future.

Pros and cons of using dbt Cloud

Pros:

- Simplified setup and management of dbt environment

- Enhanced collaboration capabilities for data teams

- Integrated dbt job scheduling and monitoring

- Managed upgrades and updates

- Web-based IDE for streamlined development

- Metadata API

Cons:

- Reliance on a third-party platform

- May be more expensive for larger teams or organizations with complex requirements

- Not able to deploy within the corporate's private cloud

- Unable to leverage VS Code extensions and other Python libraries

- May require additional components or additional SaaS solutions for complex orchestrations

dbt Core for your enterprise data management platform

dbt Core is the heart of dbt and since it is open source, anyone can use it freely. That being said, an enterprise data management platform must account for additional requirements such as the development experience, deployment and scheduling dbt runs, and creating a DataOps process. This will impact the total cost of ownership and the platform's time to market.

IDE for dbt Core development

Leveraging dbt Core allows for ultimate flexibility, but it involves setting up a development environment on each team member's local machine or on a virtual environment such as an AWS workspace. Installation includes installing a popular dbt IDE like VS Code, installing dbt Core, configuring a connection to the data warehouse, and managing additional dependencies such asa compatible version of Python.

While the overall installation process for dbt is similar for both Windows and Mac users, there are some differences and potential gotchas to be aware of. Additionally, most organizations provide users with pre-configured computers and these systems may be delivered with software that conflicts with dbt Core. This complexity can lead to frustration and delays and may require assistance from senior members of the team, pulling them away from value-added work.

Once everything is installed, configuring dbt to connect to a data warehouse can also introduce complexity before a new team member can begin familiarizing themselves with the organization's dbt project structure and coding practices.

An enterprise setup of dbt should also include components that improve developer code quality and productivity. Popular IDEs like VS Code include a vast array of extensions that also improve the development experience. These include dbt Power User, VS Code extension for SQLFluff, and the dbt Snowflake extension.

Configuring a development environment can take anywhere from a few days to as long as a week. Depending on the person’s technical aptitude and experience with dbt. To improve this process, companies spend weeks or months dialing in the installation process and documenting it in knowledge base tools like Confluence or Notion. 

A development environment is not evergreen, it must be upgraded as new versions of dbt Core are released for added functionality or to fix bugs. Upgrading can be as error-prone as new installations and some companies opt not to keep their enterprise data platform current. This, however, is ill-advised because this type of technical debt only gets harder to remediate over time.

dbt Core for DataOps

When using dbt Core for your enterprise data platform, you will need to not only define and configure the automation scripts, but you will also need to ensure that all the components, such as a git server, CI server, CI runners, etc. are all working harmoniously together.

Since dbt Core runs within the corporate firewall, it can be integrated with any CI tool and internal components such as Jira, Bitbucket, and Jenkins. This flexibility comes at a price. To do this well, all the project dependencies must be packed into reusable Docker containers. Notifications will also need to be defined across the various components and all of this will take time and money.

dbt Core for deployment

When setting up a deployment environment, companies using dbt Core can leverage any orchestration tool, such as Airflow. They can connect steps prior to or after the dbt transformations and they can trigger any tool that exists within or outside the corporate network.

However, scaling tools like Airflow is not straightforward and may require knowledge of additional technologies like Kubernetes. By default, Airflow uses Python files and therefore, the team working with these orchestration scripts will need to also know Python on top of SQL.

A mature enterprise platform will also require multiple Airflow environments for development, testing, and production, adding complexity and increasing the total cost of ownership.

dbt Core extensibility 

Using dbt Core is inevitably the most flexible option, which is why many organizations choose to go this route. Companies can run dbt Core and leverage leading code editors like VS Code. They can install any additional Python dependencies, including running any proprietary code that is only available within the corporate firewall. They can also take advantage of the ever-growing VS Code extensions such as those for Snowflake or Databricks.

Teams using dbt Core can send notifications to Slack, MSTeams, or any other platform as long as they develop and maintain these solutions.

However, this extensibility will only increase the platform's total cost of ownership and the time it takes to deploy the end-to-end platform.

dbt Core security

One reason some enterprises choose to use dbt Core is that they are not able to leverage SaaS solutions due to compliance or data privacy requirements such as GDPR. By leveraging dbt Core, they can achieve ultimate security as dbt Core would run in and comply with corporate network policies.

Running dbt Core within the corporate firewall also eliminates the need for lengthy risk assessments. However, ensuring that security patches are applied is left to the organization. Managing Single Sign-on with tools like Airflow and setting up Role Based Access Controls are also things that will have to be solved when using dbt Core as a dbt enterprise data management platform.

Who should consider using dbt Core

A lot of the power and flexibility with using dbt Core comes from being able to customize the entire dbt experience. However, all of this comes with the expense of having to build a data management platform. It is not uncommon for organizations to take six months or up to a year to set up and test their custom dbt enterprise data management platform. They will also likely need to hire external consultants who have worked with a myriad of technologies and understand how to integrate them.

Building a platform is not for the faint of heart and since it is something custom to the organization, it will also increase the total cost of ownership. Some organizations get the basics in place just to make their teams productive, but eventually, they find that without proper design and implementation, they cannot take full advantage of the value of dbt and the rest of the modern data stack.

Using dbt Core as a dbt enterprise data platform can be a cost-effective option for smaller teams with limited requirements and budgets. It can also be good for those trying out dbt. Large organizations with stringent security requirements of sensitive data and strong technical platform teams may also leverage this option as it allows them to deploy in a way that is compliant with corporate policies.

As platform complexity grows, so does the cost and time needed to implement the end-to-end platform. Larger organizations should consider whether developing something custom is better than using a managed solution or taking shortcuts that can reduce the value of dbt and the modern data stack. 

Pros and cons of using dbt Core

Pros:

- Full control over the dbt environment and configurations

- No reliance on third-party services

- No subscription costs

- Able to meet corporate security requirements by installing within the corporate firewall

Cons:

- Requires technical expertise in defining the installation steps and documenting them

- May require hands-on involvement during setup and upgrades

- It can be time-consuming and challenging to maintain consistency across team members

- Upgrades and dependency management may be more complex compared to managed solutions

- Resolving installation issues can consume senior team members' time

- Total cost of ownership of the platform may be higher than using a managed solution

- Additional expertise will be needed to set up all the required technologies

- It will take longer to implement than a managed solution

Managed dbt Core for your enterprise data management platform

We have seen that dbt Cloud is a great dbt Core managed platform, but it has some limitations for large or complex deployments. On the other hand, using dbt Core alone gives organizations ultimate flexibility, but it comes at an added expense in terms of the time and cost that it takes to stand up and maintain a custom, one-of, enterprise data management platform.

What if you could combine the simplicity of dbt Cloud with the flexibility of dbt Core? A managed dbt Core platform, like Datacoves, can combine some of the best of both worlds. However, not all managed platforms are the same, below we discuss what to look for when considering a managed dbt Core enterprise data platform. Checkout our dbt pricing page for more pricing information.

Managed VS Code as a dbt IDE for development

By far, the tool most of the dbt enterprise users will interact with is the IDE. When considering a managed dbt Core platform, consider the developer experience. Will they have VS Code in the browser? Will you get a curated, best in class dbt IDE like VS Code with integrated libraries like SQLFluff and extensions that simplify the developer workflow? When required, will you be able to customize the developer environment with company specific Python libraries and extensions?

In a best-in-class developer setup, new users are onboarded in minutes with simple configuration screens that remove the need to edit text files such as profiles.yml and remove the complexity of creating and managing SSH keys. Version upgrades of dbt or any dependent library should be transparent to users.

Some platforms will also have added flexibility and let you leverage VS Code for non-dbt development tasks, such as opening a Jupyter notebook environment or a Python development environment with libraries such as Python Black.

Spinning up a pristine environment should be a matter of clicks. To test new dbt versions, you should be able to run your project in an isolated environment before switching all developers to the new version of dbt.

Managed dbt Core for DataOps

The advantage of a managed dbt Core platform when it comes to DataOps is that you do not have to worry about creating Docker images with all the dependencies you need as you perform your Continuous integration tests. Working with a partner that has experience with enterprise CI/CD platforms like Jenkins will also accelerate your time to value. A knowledgeable dbt Core managed platform partner will know best practices and guide you in setting up a world-class workflow that will scale with your enterprise. Developers should also be able to execute CI validations locally before pushing their changes to GitHub, thus reducing the back and forth to fix issues during the CI checks.

Some managed dbt Core platforms can run within your corporate firewall, allowing you to leverage existing tools like Jira, Bitbucket, Jenkins, or Gitlab. All this simplifies and accelerates the rollout of a dbt data management platform.

Managed dbt Core for deployment

Enterprises are complex, and as such, they must consider a flexible orchestration tool that has wide adoption, documentation and can scale to the needs of the company.

A mature enterprise dbt Core platform will include a full-fledged orchestrator such as Airflow. With Airflow, organizations can orchestrate pipelines that include Extract, Load, Transform, and Activation steps. Other aspects to consider is whether Airflow jobs can scale by leveraging technologies such as Kubernetes and whether it will be simple to send notifications to enterprise tools such as MS Teams.

Some dbt Core platforms that leverage dbt have also adopted mechanisms that simplify Airflow DAG creation by letting users create them using yml files vs creating traditional Python-based DAGs.

A complete enterprise data management platform will allow you to easily create multiple Airflow environments for development, testing, and production. It will also eliminate the burden of having to create Docker images with your dbt dependencies.

Managed dbt Core extensibility

An enterprise dbt Core platform will streamline the process of ramping up with dbt and it will offer the organization the flexibility it requires. This may include triggering AWS Glue jobs as well as Fivetran jobs. It may include triggering internal and external APIs or it may require a pre-configured development environment with Python libraries like Snowpark or Streamsets.

Managed dbt Core security

When it comes to security, different types of industries have different requirements. Regulated industries in Life Sciences or Finance may not want to or and unable to use SaaS solutions, so being able to deploy a dbt enterprise data management platform within the corporate firewall is a must.

Integrating with SSO solutions like Okta or PingID is also a must, and managing user's authorization using existing mechanisms like Active Directory simplifies compliance with corporate policies. 

Another area that may be overlooked when using open-source solutions is validating that libraries are up to date and there are no vulnerabilities introduced into the network. A robust dbt Core managed platform will be able to leverage internal registries like Artifactory, which are set up to scan for these issues. 

Finally, at enterprise scale, the dbt Core managed platform must have robust Role Based Access Controls to assure users have the appropriate level of access.

Who should consider using a managed dbt Core platform

Just like dbt Cloud, a reliable enterprise dbt Core data management platform will reduce the time to value. Teams will not have to figure out how to host and configure the multitude of tools and libraries and they will have a starting point for implementing best practices at the start of the project. 

By eliminating the need to stand up a platform, teams can focus on change management and new ways of working that will have the biggest impact on the organization. The total cost of ownership will be lower for the company since they do not have to build and constantly maintain their one-of platform. In contrast, they will benefit by working with a partner that works with other large organizations and looks for ways to improve and enhance the platform for everyone.

A managed dbt Core enterprise data platform is necessary for organizations that are not in the business of creating and managing complex technologies and those who need the flexibility that comes with complex enterprise tools and processes. When dbt Core must be deployed within the corporate firewall, an enterprise managed dbt data platform can also be a good option for companies with stringent security requirements to protect sensitive data.

Smaller organizations that want to leverage a managed VS Code environment or who want to implement DataOps from the outset can also realize the benefits of a SaaS managed dbt Core data platform. 

A managed dbt Core enterprise data platform gives companies the simplicity of dbt Cloud with the peace of mind of dbt Core. Best of all, if the platform leverages open-source technologies, there is no vendor lock-in.

Pros and Cons of using a managed dbt Core platform

Pros:

- Simplified developer onboarding

- Robust IDE such as VS Code

- Ability to configure a dbt environment

- Managed orchestrator such as Airflow

- Able to meet corporate security requirements by installing within the corporate firewall

- Requires a lot less technical expertise to set up and run

- Simplified dependency management across the development environment, CI/CD worker, and Airflow

- Lower total cost of ownership

Cons:

- Unable to have full control over the platform

- Reliance on third-party

- Licensing costs

Conclusion

Selecting between dbt Cloud, dbt Core, or a managed dbt Core enterprise platform comes down to the internal experience within an organization and their desire to focus on data vs developing and maintaining a custom dbt platform. The beauty of the dbt ecosystem is that the components are freely available and with enough time and money, anyone can do what they need. However, enterprises that want to move fast and leverage the full potential of dbt, should consider a managed dbt Core enterprise platform such as Datacoves to reduce the total cost of ownership and accelerate time to market. It is important to be aware of the dbt alternatives to make an informed decision.

Genesis of Datacoves and beyond

A few years ago, we identified a need to simplify how companies implement the Modern Data Stack, both from an infrastructure and process perspective. This vision led to the creation of Datacoves. Our efforts were recognized by Snowflake as we became a top-10 finalist in the 2022 Snowflake Startup Challenge. We have become trusted partners at Fortune 100 companies, and now we are ready for our next chapter.

Datacoves is thrilled to announce that we have been selected to be part of the Spring 2023 cohort at TinySeed. Their philosophy and culture align flawlessly with our values and aspirations. Our goal has always been to help our customers revolutionize the way their teams work with data, and this investment will enable us to continue delivering on that vision.

The TinySeed investment will also strengthen our commitment to supporting open-source projects within the dbt community, such as SQLFluff, dbt-checkpoint, dbt-expectations, and dbt-coves. Our contributions will not only benefit our customers but also the broader dbt community, as we believe open-source innovation enhances the tools and resources available to everyone.

We want to express our deep gratitude to Tristan Handy, CEO & Founder at dbt Labs and the entire dbt Labs team for partnering with us on the 2023 State of Analytics Engineering Community survey and for creating the amazing dbt framework. Without dbt, Datacoves would not be where it is today, and our customers would not enjoy the benefits that come with our solutions.

As we continue to grow and evolve, we look forward to finding new ways to collaborate with dbt Labs and the dbt community to further enhance how people use dbt and provide the best possible experience for everyone. Together, we can empower data teams around the world and revolutionize the way they work.

‍

‍

In continuation of our previous blog discussing the importance of implementing DataOps, we now turn our attention to the tools that can efficiently streamline your processes. Additionally, we will explore real-life examples of successful implementations, illustrating the tangible benefits of adopting DataOps practices.

Which DataOps tools can help streamline your processes?

There are a lot of DataOps tools that can help you automate data processes, manage data pipelines, and ensure the quality of your data. These tools can help data teams work faster, make fewer mistakes, and deliver data products more quickly.

Here are some recommended tools needed for a robust DataOps process:

1. dbt (data build tool): dbt is an open-source data transformation tool that lets teams use SQL to change the data in their warehouse. dbt has a lightweight modeling layer and features like dependency management, testing, and the ability to create documentation. Since dbt uses version-controlled code, it is easy to see changes to data transformations (code reviews) before they are put into production.  dbt can dynamically change the target of a query's "FROM" statement on the fly and this allows us to run the same code against development, test, and production databases by just changing a configuration. During the CI process, dbt also lets us run only the changed transformations and their downstream dependencies.
2. Fivetran: Fivetran is an Extract and Load(EL) tool that has been gaining in popularity in recent years since it removes the complexity of writing and maintaining custom scripts to extract data from SaaS tools like Salesforce and Google Analytics.  By automating extracting data from hundred's of sources Fivetran removes complexity freeing data engineers to work on projects with a bigger impact. Finally, Fivetran has a robust API which allows you to save configurations done vie their UI for disaster recovery or to promote configurations form a development to a production environment.
3. Airbyte: Airbyte is another data-ingestion EL tool that is appealing because it is open source, requires little or no code, and is run by the community. It also makes it easier to extract and load data without having to do custom coding.  Airbyte also offers a connector development kit to help companies build custom connectors that may not be available. This allows companies to leverage most of the Airbyte functionality without too much work. There's also an API that can be used to retrieve configurations for disaster recovery.
4. SQLFluff: SQLFluff is an open-source SQL linter that helps teams make sure their SQL code is consistent and follows best practices. It gives you a set of rules that you can change to find and fix syntax errors, inconsistencies in style, and other common mistakes. Sqlfluff can be added to the CI/CD pipeline so that problems are automatically found before they are added to the codebase. By using a tool like SQLFluff, you can make sure your team follows a consistant coding style and this will help with long term project maintainability.
5. dbt-checkpoint: dbt-checkpoint provides validators to make sure your dbt projects are good. dbt is great, but when a project has a lot of models, sources, and macros, it gets hard for all the data analysts and analytics engineers to maintain the same level of quality. Users forget to update the columns in property (yml) files or add descriptions for the table and columns. Without automation, reviewers have to do more work and may miss mistakes that weren't made on purpose. Organizations can add automated validations with dbt-checkpoint, which makes the code review and release process better.
6. Hashboard: Hashboard is a business intelligence (BI) product built for data engineers to do their best work and easily spread the data love to their entire organizations. Hashboard has an interactive data exploration tool that enables anyone in an organization to discover actionable insights.
7. GitHub: GitHub offers a cloud-based Git repository hosting service. It makes it easier for people and teams to use Git for version control and to work together. GitHub can also run the workflows needed for CI/CD and it provides a simple UI for teams to perform code reviews and allows for approvals before code is moved to production.
8. Docker: Docker makes it easy for data teams to manage dependencies such as the versions of libraries such as dbt, dbt-checkpoint, SQLFluff, etc.. Docker makes development workflows more robust by integrating the development pipeline and combining dependencies simplifying reproducibility.

Examples of companies who have successfully implemented DataOps

DataOps has been successfully used in the real world by companies of all sizes, from small startups to large corporations. The DataOps methodology is based on collaboration, automation, and monitoring throughout the entire data lifecycle, from collecting data to using it. Organizations can get insights faster, be more productive, and improve the quality of their data. DataOps has been used successfully in many industries, including finance, healthcare, retail, and technology.

Here are a few examples of real-world organizations that have used DataOps well:

1. Optum: Optum is part of UnitedHealthcare. Optum prioritizes healthcare data management and analytics and when they wanted to implement new features and apps quickly, they turned to DataOps. DataOps helped Optum break down silos, saving millions of dollars annually by reducing compute usage. Optum managed data from dozens of sources via thousands of APIs, which was its biggest challenge. A massive standardization and modernization effort created a scalable, centralized data platform that seamlessly shared information across multiple consumers.
2. JetBlue: DataOps helped JetBlue make data-driven decisions. After struggling with an on-premises data warehouse, the airline migrated to the cloud to enable self-service reporting and machine learning. They've cleaned, organized, and standardized their data and leveraged DataOps to create robust processes. Their agility in data curation has enabled them to increase data science initiatives.
3. HubSpot: HubSpot is a leading company that makes software for inbound marketing and sales. It used DataOps to improve the use of its data. By using a DataOps approach, HubSpot was empowered to do data modeling the right way, to define model dependencies, and to update and troubleshoot models, which resulted in a highly scalable database and opened up new data application possibilities.
4. Nasdaq: Nasdaq, a global technology company that provides trading, clearing, and exchange technology, adopted DataOps to improve its data processing and analysis capabilities. They launched a data warehouse, products, and marketplaces quickly. After scalability issues, they moved to a data lake and optimized their data infrastructure 6x. The migration reduced maintenance costs and enabled analytics, ETL, reporting, and data visualization. This enabled better and faster business opportunity analysis.  
5. Monzo: Monzo is a UK-based digital bank that used DataOps to create a data-driven culture and improve its customer experience. By letting everyone make and look at the different data maps, they are helping teams figure out how their changes affect the different levels of their data warehouse. This gave the Monzo data team confidence that the data they give to end users is correct.

What is the future of DataOps adoption?

DataOps has a bright future because more and more businesses are realizing how important data is to their success. With the exponential growth of data, it is becoming more and more important for organizations to manage it well. DataOps will likely be used by more and more companies as they try to streamline their data management processes and cut costs. Cloud-based data management platforms have made it easier for organizations to manage their data well. Some of the main benefits of these platforms are that they are scalable, flexible, and cost-effective. With DataOps teams can improve collaboration, agility, and build trust in data by creating processes that test changes before they are rolled out to production.

With the development of modern data tools, companies can now adopt software development best practices in analytics. In today’s fast-paced world, it's important to give teams the tools they need to respond quickly to changes in the market by using high-quality data. Companies should use DataOps if they want to manage data better and reduct the technical debt created from uncontrolled processes. Putting DataOps processes in place for the first time can be hard, and it's easier said than done. DataOps requires a change in attitude, a willingness to try out new technologies and ways of doing things, and a commitment to continuous improvement. If an organization is serious about using DataOps, it must invest in the training, infrastructure, and cultural changes that are needed to make it work. With the right approach, companies can get the most out of DataOps and help their businesses deliver better outcomes.

At Datacoves, we offer a suite of DataOps tools to help organizations implement DataOps quickly and efficiently. We enable organizations to start automating simple processes and gradually build out more complex ones as their needs evolve. Our team has extensive experience guiding organizations through the DataOps implementation process.

Schedule a call with us, and we'll explain how dbt and DataOps can help you mature your data processes.

The amount of data produced today is mind-boggling and is expanding at an astonishing rate. This explosion of data has led to the emergence of big data challenges that organizations must overcome to remain competitive. Organizations must effectively manage data, guarantee its accuracy and quality, in order to derive actionable insights from it. This is where DataOps comes in. In this first post we go through an overview of DataOps and in our next post we will discuss tooling that support DataOps as well as discuss companies that have seen the benefit of implementing DataOps Processes.

What is DataOps?

DataOps is a methodology that merges DevOps principles from software development with data management practices. This helps to improve the process of developing, deploying, and maintaining data-driven applications. The goal is to increase the delivery of new insights and the approach emphasizes collaboration between data scientists, data engineers, data analysts, and analytics engineers.

Automation and continuous delivery are leveraged for faster deployment cycles. Organizations can enhance their data management procedures, decrease errors, and increase the accuracy and timeliness of their data by implementing DataOps.

The overarching goal is to improve an organization's capacity for decision-making by facilitating quicker, more precise data product deliveries. DataOps can speed up task completion with fewer errors by encouraging effective team collaboration, ultimately enhancing the organization's success.

DataOps is now feasible thanks to modern data stack tools. Modern data tools like Snowflake and dbt can enable delivery of new features more quickly while upholding IT governance standards. This allows for employee empowerment and improves their agility all while helping to create a culture that values data-driven decision-making. Effective DataOps implementation enables businesses to maximize the value of their data and build a data-driven culture for greater success.

The importance of DataOps in solving big data challenges

It is impossible to overstate the role that DataOps plays in overcoming big data challenges. Organizations must learn how to manage data effectively as it continues to increase if they want to stay competitive.

DataOps addresses these challenges by providing a set of best practices and tools that enable data teams to work more efficiently and effectively. Some of the key benefits of DataOps include:

1. Improved agility and flexibility: DataOps empowers data teams to work in an agile manner, enabling them to quickly adapt to shifting requirements and business needs. By basing decisions on up-to-the-minute information, businesses can stay one step ahead of the competition.
2. Increased collaboration: DataOps encourages collaboration between different teams involved in the data management process, including data engineers, data scientists, and business analysts. This leads to better communication, faster feedback loops, and more efficient workflows.
3. Enhanced quality and accuracy: DataOps lowers the risk of errors and inconsistencies in the data by automating data deliverable processing and analysis pipelines. As a result, businesses are better able to make decisions based on accurate and trustworthy data
4. Reduced costs: DataOps helps businesses save money by reducing the time and resources needed to manage and perform data analytics. As a result, they can invest in other aspects of their business to maintain their competitiveness.

In a world without DataOps, data-related processes can be slow, error-prone, and documentation is often outdated. DataOps provides the opportunity to rethink how data products are delivered. This promotes a culture of federated delivery. Everyone in the organization with the necessary skills is empowered to participate in insight generation.

We can achieve comparable results while preserving governance over the delivery process by learning from large, distributed open-source software development projects like the Linux project. To ensure high-quality output, a good DataOps process places an emphasis on rules and procedures that are monitored and controlled by a core team.

What are the components for successful DataOps implementation?

To implement DataOps well, one needs to take a holistic approach that brings together people, processes, and technology. Here are the five most important parts of DataOps that organizations need to think about in order to implement it successfully:

1. Collaborative Culture: Organizations must foster a collaborative culture and they must encourage cross-functional teamwork and make it easy for data scientists, data engineers, data analysts, and analytics engineers to talk to each other and have visibility into each other's work. Collaboration breaks down silos and makes it easier for everyone to understand how data-driven applications work. A collaborative culture also helps team members feel like they own their work and are responsible for it, which can help improve the quality of the data and the applications that use it.
2. Automation: Automation is a key part of DataOps because it helps organizations streamline their feature deliveries and cut down on mistakes. Data ingestion, data integration, and data validation are all examples of tasks that are done over and over again. By automating these tasks, organizations can free up their resources to focus on more complex and valuable tasks, like analyzing data and deriving insights from it.
3. Continuous Integration and Delivery: Continuous integration and delivery (CI/CD) are important parts of DataOps. Together they enable the deployment of working code to production in a continuous and iterative way. CI/CD can help speed up the time it takes to get data-driven apps to market so companies can respond quickly to changing business needs and provide more value to their customers.
4. Monitoring and Feedback: Monitoring and feedback are important parts of DataOps because they help make sure that data-driven applications work as they should. By keeping an eye on how well data-driven applications work, organizations can catch problems early and fix them as needed. By asking users for feedback, the code can be made more reliable and maintainable leading to better better outcomes.
5. Data Governance: Data governance is a set of policies, procedures, and controls that make sure data is managed correctly throughout its lifecycle. Data governance includes data ownership, data quality, data security, data privacy, and data lineage. By putting in place strong data governance practices, organizations can make sure that their data is correct, consistent, and in line with company standards. Data governance can help make the data more trustworthy and lower the chance of a data breach.

Having identified what needs to be done to achieve a good DataOps implementation, you may be wondering how we turn these ideas into actionable steps.

What are the best ways to implement DataOps in organizations?

1. Start Small and Grow: When putting DataOps in place, it's important to start small and grow gradually. This method lets companies try out and improve their DataOps processes before putting them to use on bigger projects. By starting out small, organisations can find potential problems and solve them before rolling them out to a larger audience.
2. Align DataOps with Business Goals: For an organization's DataOps implementation to work, its DataOps processes need to be in line with its business goals. Organizations should figure out what their most important business goals are for their data-driven applications, and then design their DataOps processes to help them reach these goals. This approach makes sure that DataOps is focused on adding value to the business and not just on implementing new technology.
3. Foster collaboration between teams: One of the most important parts of DataOps is getting teams to work together. Organizations need to help data scientists, data engineers, data analysts, and analytics engineers work together. This method breaks down silos and makes it easier for everyone to understand how data-driven applications work. Organizations can make it easier for people to work together by putting them on cross-functional teams, encouraging open communication, fostering transparency, and where possible encouraging the same tools, e.g. using dbt for data pipelines.
4. Invest in tools for automation. Automation is a key part of DataOps. Organizations should invest in automation tools that can help them speed up their data management processes, such as data ingestion, data integration, and data validation. Automation can help cut down on mistakes, increase productivity, and free up resources to be used on more complicated tasks, like analyzing data.
5. Implement CI/CD processes: Creating a robust feature release process with automated validations and code review will encourage collaboration and build transparency in the process. Users will be encouraged to move faster by removing the fear that a new feature will break production.
6. Monitor Performance and Get Feedback: Two important parts of DataOps are monitoring performance and getting feedback. Organizations should use monitoring tools that can keep track of how well data-driven applications are working and get feedback from users. This method helps find problems early on and makes the data more accurate and reliable. It also gives organizations useful information about how users leverage tools and this can help them improve the user experience.

Looking to implement DataOps quickly and efficiently?  

At Datacoves, we offer a comprehensive suite of DataOps tools to help organizations implement robust processes quickly and efficiently. One of the key benefits of using Datacoves is the ability to implement DataOps incrementally using our dynamic Infrastructure. Organizations can start by automating simple processes, such as data collection and cleaning, and gradually build out more complex processes as their needs evolve.

Our agile methodology allows for iterative development and continuous deployment, ensuring that our tools and processes are tailored to meet the evolving needs of our clients. This approach allows organizations to see the benefits of DataOps quickly, without committing to a full-scale implementation upfront. We have helped teams implement mature DataOps processes in as little as six weeks.

The Datacoves platform is being used by top global organizations and our team has extensive experience in guiding organizations through the DataOps implementation process. We ensure that you benefit from best practices and avoid common pitfalls when implementing DataOps.

Set aside some time to speak with us to discuss how we can support you in maturing your data processes with dbt and DataOps.

Don't miss out on our latest blog, where we showcase the best tools and provide real-life examples of successful implementations.

Get ready to take your data operations to the next level!

Big Data Analytics - Making information tasty and accessible

No one bakes for the sake of baking alone; cakes are meant to be shared.  If no one bought, ate, or gifted someone with their delicious chocolate cake, then there would be no bakers. The same is true for data analytics. Our goal as data practitioners is to feed our organization with the information needed to make decisions.  

If our cake doesn’t taste good or isn’t available when people want dessert, then it doesn’t matter that we made it from scratch. When it comes to big data, your goal should be to have the equivalent of a delicious cake - usable data - available when someone needs it.

Big Data Analytics - The ever-changing tastes of data consumers

Life would be simple if everyone were happy with a single flavor of cake. Metrics play a crucial role in our organization, and two of the most fundamental ones are ARR (Annual Recurring Revenue) and NRR (Net Recurring Revenue). These metrics are like chocolate and vanilla - they remain popular and relevant. Yet, these flavors alone are not enough. Just as with ice cream flavors, we eventually need to try something new, the same goes for insights. It's important to experiment and explore different perspectives.

When something is novel, we love it. When we first start baking, we are not consistent. But, over time, quality improves. We go from ok, to good, and to great. Even with our newfound expertise, the chocolate cake will become boring. We all want something new.  

With data, you often start with a simple metric. Having something is better than nothing, but that only lasts so long. We discover something new about our business and we translate that information into action. This is good for a while, but we will eventually see diminishing returns. While a LTV (Life Time Value) analysis may have a significant impact today, its usefulness is likely to diminish within a short period of time. Your stakeholders will crave something new, something more innovative and updated.

Someone is going to ask for a deep dive into how CAC (Customer Acquisition Cost) impacts LTV, or they might ask for a new kind of icing on that cake you just made. Either way, the point remains – as those around you start making use (or eating) of what you’re providing, they will inevitably ask for something more.  

‍

Big Data Analytics - How much expertise do you need to bake successfully?

It depends. Some organizations can do more than enough with the reports and dashboards built into their CRM, web analytics, or e-commerce system. This works for many companies. There is no need to spend extra time and money on more complex data systems when something simple will suffice. There is a reason that chocolate and vanilla are popular flavors; many people like their taste and know what they are getting with their order. The same can be said for your data infrastructure.

However, flexibility is the challenge. You are limited by your reporting options and the data analytics you can access. You can have a cake you can eat, but it’s going to look a certain way, having only one type of icing, and it certainly will not have any premium fillings.  

If you want those things, you need to look for something a bit more nuanced.  

‍

Big Data Analytics - Start with an Easy Bake Oven

How do you start to accommodate your organization’s new demand for analysis or your family’s newly refined palate? By using new tools and techniques.  

The Easy Bake Oven is the simplest first step – you get a pouch of ingredients, mix them, and within a few minutes, you have a cake. You might think it’s a children’s toy, but there are very creative recipes out there for the adventurous types. Unfortunately, you can only go so far; you’re limited by the size of the oven and the speed you can bake each item.  

The corollary in the data world is Excel – a fabulous tool, but something that has its limitations. While it is possible to extract data from your tools and manipulate it in a spreadsheet to make it more manageable, you are still limited by the pre-designed extracts offered by the vendor tools. Excel is flexible enough for many, but it is not a perfect end-state reporting solution for everyone.  

Eventually, you’ll need to address inconsistencies between systems and automate the process of data prep before it gets into your spreadsheet.

‍

Big Data Analytics - Using more advanced techniques to elevating your baking

We have a feel for the basics, but now we want to improve our process. There are certain things we do repeatedly, regardless of the recipes we’re making. We need to create an assembly line. Whether we’re calculating metrics or mixing the wet ingredients and the dry ingredients, there are steps we need to complete in a specific order and to a certain level of quality.  

We’ve moved on from the Easy Bake Oven and are now baking using a full-size oven. We need to be more careful about our measurements, ensure the oven is at the correct temperature, and write down notes about the various steps in our more complicated recipes. We need to make sure that different cakes, fillings, and decorations are ready at the same time, even if they have wildly different prep times. We need consistent results every time we make the cake, and we need others to be able to make the same recipe and achieve the same results by following our instructions.

Documenting and transferring this knowledge to others is difficult. Sure, you can write things down, but it is easy to skip a step that you take for granted. Perhaps your handwriting is hard to read, or someone is having trouble with the oven. You may be aware that using a hair dryer while baking can cause the circuit breaker to trip, but unless you document this information, others will not know this quirk. If your recipe has changed, such as using individual ingredients instead of a pre-made cake mix, it is important to clearly specify these adjustments, otherwise your friends and family may struggle to replicate your delicious cake.

There are plenty of data analytics tools out there that assist at this stage – Alteryx, Tableau Prep, and Datameer are just a few of many. In large enterprise organizations, you might find Informatica Power Center, Talend, or Matillion. These types of tools have graphical user interfaces (GUI); they give you the flexibility to extract data, load data, inspect data and transform it. Many enable you to define and calculate metrics. But they require you to work within each tool’s set of rules and constraints. This works well if you are starting and need something less complex.

The process that was once simple now is not; there are hidden assumptions, configurations, and requirements. You’re not using a pen-and-paper recipe anymore; now you’re working within a new system.  

GUI-based tools are great for companies whose workflows fit the way the systems work. But between the way some tools are licensed, and the skill needed to use them, they are only available to the IT organization. This leaves users to find shortcuts, develop workarounds, and become dependent on the business’ shadow IT. Inevitably, you’re going to run into maintainability issues.  

Recipes often have plenty of steps and ingredients; “data recipes” are no different. There are dependencies between operations, different run times for different transformations, different release cadences, and data availability SLAs. Your team might be able to manage your entire workflow quite well with one of these tools, but once your team starts introducing custom SQL logic or additional overlapping tools into the ecosystem, you introduce another layer of complexity. The result is an increase in the total cost of ownership.  

Often, this complexity is opaque, too. It is not obvious what that GUI component is doing, but they are strung together to build something usable. Over time, the complexity continues to grow; custom SQL logic is introduced, and more steps in the chain. Eventually, abstractions begin to form. The data engineers decide that these processing pipelines look quite similar and can be customized based on some basic configurations. Less overhead, more output.  

You’re now on the path of building custom ELT (Extract, Transform, and Load) pipelines, stitched together within the constraints of a GUI-based system. For some companies, this is okay, and it works. But there is a hidden cost – it is harder to maintain high quality inputs and outputs. The layers are tightly coupled and a mistake in one step is not caught until the whole pipeline is complete.  

IT may not be aware of downstream issues because they occur in other tools outside of their domain; one change here breaks something else there. This is like buying a ready-made cake mix and “enhancing it” with your custom ingredients. It works until it does not. One day, Duncan Hines changes their ingredients, and without you realizing, there is a bad reaction between your “enhancements” and the new mix.  Your once great recipe is not so great anymore, but there was no way for Duncan Hines to know. They expected you to follow their instructions; everything has been going according to plan until now.  

Even if your tool has strong version control built in, it’s often difficult to reverse the changes before it is too late. If your recipe calls for 1 cup of sugar but you accidentally add in 11, we don’t want to wait until the cake is baked to discover the error. We want to catch that mistake as soon as it happens.  

‍

Big Data Analytics - Building consistence into the baking process

Everything up to this point serves a specific profile of a business, but what happens when the business matures beyond what these tools offer? What happens when you know how to bake a cake, but struggle to consistently produce hundreds or thousands of the same quality cakes?

When we aim to expand our baking operations, it's crucial to maintain consistency among bakers, minimize accidental mistakes, and have the ability to swiftly recover from any errors that occur. We need enough mixers and ovens to support the demand for our cakes, and we need an organized pantry with the correct measuring cups and spoons. We need to know which ingredients are running low, which are delayed, and which have common allergens.  

In data analytics, we have our own supply chain, often called ELT. You will find tools like Airbyte and Fivetran are common choices for bringing in our data “ingredient delivery”. They manage data extraction and ingestion so you can skip the manual CSV downloads that once served you so well.

We want to ensure quality, have traceability, document our process, and successfully produce and deliver our cakes. To do this, we need a repeatable process, with clear sequential steps. In the world of baking, we use recipes to achieve this.  

All baking recipes have a series of steps, some of which are common across different recipes. For example, creaming eggs and sugar, combining the wet and dry ingredients, and whipping the icing are repeatable steps that, when performed in the correct order, result in a delicious dessert. Recipes also follow a standard format: preheat the oven to the specified temperature, list the ingredients in the order in which they are used, and provide the preparation steps last. The sequence is intentional and provides a clear understanding of what to expect during the baking process.

We can apply this model to our data infrastructure by using a tool called dbt (data build tool). Instead of repeating miscellaneous transformation steps in various places, we can centralize our transformation logic into reusable components. Then we can reference those components throughout our project. We can also identify which data is stale, review the chain of dependencies between transformations, and capture the documentation alongside that logic.

We no longer need a GUI-driven tool to review our data; instead, we use the process as defined in the code to inform our logic and documentation. Our new teammate can now confidently create her cake to the same standard as everyone else; she can be confident that she is avoiding common allergens, too.

Better yet, we have a history of changes to our process and “data recipe”. Version control and code reviews are an expectation, so we know when modifications to our ingredient list will cause a complete change in the final product. Our recipes are no longer scattered, but part of a structured system of reusable, composable steps.  

Big Data Analytics - Are you ready to bake well, consistently?

Maturing the way we build data processes comes down to our readiness. When we started with our Easy Bake Oven, little could go wrong, but little could be tweaked. As we build a more robust system, we can take advantage of its increased flexibility, but we also need to maintain more pieces and ensure quality throughout a more complex process.

We need to know the difference between baking soda and baking powder. Which utensils are best, and which oven cooks most evenly? How do we best organize our new suite of recipes? How do you set up the kitchen and install the appliances? There are many decisions to make, and not every organization is ready to make them. All this can be daunting for even large organizations.

But you don’t have to do things all at once. Many organizations choose to make gradual improvements, transforming their big data process from disorganized to consistent.

You can subscribe to multiple SaaS services like Fivetran or Airbyte for data loading and use providers like dbt Cloud for dbt development. If your work grows increasingly complex, you can make use of another set of tools (such as Astronomer or Dagster) to orchestrate your end-to-end process.  You will still need to develop the end-to-end flow, so what you gained in flexibility you have lost in simplicity.

‍

Big Data Analytics - Becoming a master baker quickly

This is what we focus on at Datacoves. We aim to help organizations create mature processes, even when they have neither the time nor resources to figure everything out.

We give you “a fully stocked kitchen” - all the appliances, recipes, and best practices to make them work cohesively. You can take the guess work out of your data infrastructure, and instead, use a suite of tools designed to help your team perform timely and efficient analytics.

Whether your company is early in its data analytics journey or ready to take your processes to the next level, we are here to help. If your organization has strong info-sec or compliance requirements, we can also deploy within your private cloud. Datacoves is designed to get you “baking delicious cakes” as soon as possible.

Set aside some time to speak with me and learn how Datacoves has helped both small and large companies deploy mature analytics processes from the start. Also check out our case studies to see some of our customer's journeys.

Imagine yourself baking the next great cake at your organization. You can do it quickly with our help.

‍