Data integration without copying data from other systems? Zero-ETL technology makes it possible.

Today, businesses have access to more and more data — it is collected through social media, e-commerce platforms, and digital transactions, for example. However, many companies face the challenge of utilizing this data efficiently. Often, data is stored in isolated silos and synchronization between systems occurs with delays.

Let's consider a specific problem: Your marketing team wants to launch a campaign targeting a specific individual. However, the data available in the marketing tool is outdated or incomplete because e-commerce store data has not yet been synchronized with the CRM or marketing automation tool. Perhaps the email address was recently updated or the individual changed their preferred communication channel from email to SMS—but this information hasn't been updated across all systems.

In this article, I'll explore how Zero-ETL can help minimize this problem, the concept behind it, and its implications for data integration in Salesforce Data Cloud and Marketing Cloud Growth—and ultimately, your marketing efforts.

1) What does the traditional ETL process look like?

If you're already familiar with the ETL process, you can skip this section and jump to the next. But to better understand ETL—Extract, Transform, Load—let's look at a simplified example:

In a traditional ETL data processing pipeline, data is extracted from various sources (1), such as databases, APIs, JSON files, XML files, or data warehouses. For simplicity, let's assume a table containing customer information like 'First Name,' 'Last Name,' 'Email,' 'Purchased Product,' and 'Price Paid.' Once the data is extracted, the next step is data transformation (2). This might involve combining column values, performing calculations, merging tables, or removing unnecessary information. In our example, we create a new column that combines 'First Name' and 'Last Name' into a full name ('Full_Name') and stores the value "True" for customers who spent more than 20 currency units. After transformation, the data is loaded (3) into a platform for further analysis - for example, we can apply AI methods or visualize the data in dashboards and reports. In our example, we create a simple histogram to show how many customers spent over 20 CHF.

This example is super simplified, but demonstrates the key steps. Of course, in real-world projects, much larger datasets are extracted, transformations are far more complex and data is loaded into systems like databases, data warehouses, data lakes, or marketing automation tools.

You may now be asking yourself, what is the challenge of this approach?

In traditional ETL processes, data is not available in real-time: It is usually processed or copied in batches. Additionally, this approach requires significant resources and incurs higher costs. This is where Zero-ETL comes into play.

2) What is Zero-ETL?

We live in an age of immediacy. Every message, every film, every song must be available immediately and at any time - a change that we can thank for successes such as WhatsApp, Netflix and Spotify, among others.

This is exactly what cloud providers such as Amazon Web Services, Google Cloud, Microsoft Azure and Salesforce have now taken to heart: Data should be able to be processed and analysed almost in real time and without major delays.

Zero-ETL, or Zero-Copy, is a concept in data integration. Unlike the traditional ETL approach, Zero-ETL enables seamless data flow between systems. The term was introduced in 2022 by AWS in relation to integrating Amazon Aurora with Amazon Redshift.

What's new about this concept?

It allows data to be analyzed directly in its original format in near real-time, without physical movement. This minimizes latency, enabling data transformation and analysis within a single platform.

3) Benefits of Zero-ETL technology for your marketing team

1. Real-Time Data Integration
   Zero-ETL enables data to be processed and analysed almost in real time without having to physically move it between systems.
   
   For example, a customer visits your website and shows interest in a product by adding it to their shopping basket. If the systems are connected correctly, the Marketing Cloud can immediately send the customer a personalised email with relevant recommendations or a discount on the product. This allows your marketing team to react immediately to customer actions and trigger targeted marketing campaigns.
   
   
2. Improved Customer Segmentation
   If your team has more up-to-date data and data from different systems at its fingertips, you also have other options for segmentation.
   
   For example, a customer clicks on a landing page for a specific product. By linking the landing page, the Data Cloud and the Marketing Cloud Growth, the contact is added directly to a segment for potential buyers. Your team can use this segment immediately in an email campaign.
   
   
3. Dynamic Customer Journeys
   It is easier to adapt customer journeys to current data.
   For example, a customer has just completed a purchase in the web shop. This automatically moves them to a journey for after-sales campaigns. The advantage for your team? The customer experience is significantly improved as customers receive more relevant content.
   
   
4. Scalability
   Large amounts of data from various sources can be collected in the Data Cloud via Zero-ETL. This gives your marketing team the opportunity to carry out analyses or target group formation based on insights and use these findings in the Marketing Cloud Growth.

4) How Does Zero-ETL Work in the Data Cloud with Marketing Cloud Growth?

5) Zero-ETL and now everything is done? Challenges & Considerations

1. Without the data being physically copied, it can be difficult to keep track of data sources and access rights. It must be precisely defined which users have access to which data and from which sources.
2. As the data is virtually integrated, network failures can block access to the data. Complex queries across multiple data sources can also lead to delays.
3. Zero-ETL and the Data Clod require a certain technical understanding of databases, data integration and the underlying technologies. However, as soon as the data in the Marketing Cloud Growth is usable, it can be used without extensive technical knowledge.
4. The efficiency of data integration depends on the quality of the data.

Conclusion

References

• Salesforce Blog - Zero-Copy
• Salesforce Help - Enabling Data Cloud
• DataCamp - What is Zero-ETL?