Guide to Zero Copy

Remember the last time you moved? You probably had to pack up too much stuff, transport it in a truck and unpack it in the new location — hoping it survived the trip. Imagine if your furniture and belongings could just teleport to your new place in perfect condition. It’s not possible (yet) in the physical world, but with zero copy integration you can integrate your data without actually moving it.

Zero copy lets you share data between two or more data stores without duplicating or moving it. This is great news to companies that store data in a cloud data warehouse such as Snowflake or Google BigQuery. Some organizations are reluctant to adopt a customer data platform (CDP) because they don’t want to duplicate data. With zero copy integration you can get the benefits of a CDP — such as data harmonization, identity management, built-in analytics, and activation — without the downside of physical data movement.

Read on to learn more about zero copy integration, how it differs from other transfer methods, and how to get started.

Zero copy is an integration technology that lets you combine data from multiple sources into one centralized location without duplicating or physically moving the data. With zero copy, you can access data from multiple databases without having to move, copy, or reformat anything. In addition to making access faster and easier, zero copy cuts down on the expense and risk of errors that can happen when data is moved or changed.

Using zero copy, a centralized platform can establish links with other data repositories (data sources). Here’s how you can establish pathways to share data.

Data federation is the process of bringing your data from external data platforms to your central repository. Once you federate data, you can analyze it all in one place. To begin, you will create connections to your external data platforms and define your data streams.

1. Create connections. Initiating zero copy requires setting up external integrations. To establish an external integration, you’ll need access to the external system or systems.
2. Create data streams. Data streams are continuous flows of near real-time data from your external data platforms to your integration platform. When you select a data source you can make a connection to your specific data source to identify the desired databases and schemas. During this process you can select which specific data fields you want to bring to your internal system. For example, a hotel conglomerate can import guest data and use the information to make recommendations that increase guest loyalty.
3. Define data streams. After selecting your external objects, you determine whether to perform a live query with your data stream or if you’d rather cache your data. With live queries you pull the most up-to-date information each time you perform a query, which can decrease your storage costs. Using cache, you only import in the new and changed records. This is useful for datasets that are frequently accessed but rarely change.

Data Sharing (data out) lets you share data from your central data repository to external data platforms. This way you can view your internal data in the external systems, where it can be natively presented as tables or views. You can share not just data from your internal central repository but also any data insights produced in it. Here’s how to share data with zero copy.

1. Create a data “share” and add objects to it. Before you share your data, it is important to identify what data (or data objects) you want to send out of your central data repository. For example, you may decide to share your hotel booking and reservation data with Snowflake, where external partners can have access to it.
2. Create a data share target. This is the destination where you want to share data from your data repository. To share with multiple sources, you usually have to create multiple share targets.
3. Link your data share to your share target. Once you have created both the data share and share target you want to link the two. The link allows you to create a virtual database or table where all the information will be shared between your central repository and, in our example, Snowflake.
4. Access data in the external system. Once your data is connected to the external system, you’ll be able to access it there.

Zero copy can offer your organization important advantages over traditional methods.

Data governance with zero copy lets you simplify the process and reduces the risk of failing to synchronize your data across multiple sources — which helps preserve the integrity of the source data. The simplified process helps you maintain data security and reduce unauthorized access.

Zero copy reduces storage costs because it avoids duplication of data. You get near real-time access to data from where the data resides, without additional investments in storage space.

Zero copy gives you access to data from a wide variety of data sources and pools without transforming, duplicating, or copying data. Traditional methods of data transfer usually require ETL pipelines — which can be both fragile and expensive to maintain. Zero-copy integration lets your teams access data where it lives in near real-time.

Since you’re able to access data in near real-time, you can see changes reflected right away when the data in the original location gets updated. Copying or duplicating data results in multiple versions of it, and when someone changes one copy, all the rest have to be updated separately. Zero-copy integration removes this obstacle.

Zero-copy integration lets you share insights found on your data lake or lakehouse with partner systems. For example, if you use your centralized platform to analyze customer behavior, you can share the analysis with other connected platforms. The centralized repository would allow your teams to see your findings instantly — without any need for data duplication or copying.

Agentic AI is becoming a vital part of digital work and data is its fuel. With zero copy you can integrate all types of data — structured (addresses, sales quotas), unstructured (PDFs, call transcripts, videos, emails etc.) or semi-structured (JSON, XML). The unified data creates the grounding for agentic or generative AI to take action and provide important insights that can drive improvements in your organization.

Transferring data from one storage location to another is a common practice. Often, this process entails some form of data transformation such as extract-transform-load (ETL) or extract-load-transform (ELT). But transferring data has its challenges, which is why some organizations may avoid the practice. Zero copy mitigates some of the challenges associated with transferring data, making it easier to use, for example, centralized systems such as customer data platforms.

Zero-copy integration can be used to help a wide variety of projects and industries. Below are a few examples.

If you work for a manufacturing, e-commerce, or retail company, you can use zero-copy integration to merge supply chain and inventory data into one central location. This will help you track changes in inventory levels or production processes, and help with decision-making regarding delivery times and operating costs.

Putting customer data from web, mobile, in-store, and any other locations into one central location helps you compile strong insights and create unified customer profiles. Centralized data in a CDP can then help your marketing teams compile effective marketing campaigns to reach your target customers.

Zero-copy integration can help you detect potential suspicious activity. For example, if you work for a financial institution, you can compile and analyze financial transactions from various systems in one system. This will help you identify patterns and potential fraudulent activity.

Healthcare organizations can compile more complete patient profiles by allowing clinical teams to access patient data from electronic health records (EHRs), lab systems, and wearable devices.