If you're working with Salesforce Data Cloud, you've probably heard about Data Lake Objects (DLOs). But what exactly are they, and why should you care? Let's break it down in simple terms with practical examples and demos you can try in your Salesforce org.
What Are Data Lake Objects?
Think of Data Lake Objects as special storage containers in Salesforce Data Cloud. They hold large amounts of raw data that doesn't necessarily fit into the standard Data Cloud model. It's like having a warehouse where you can store items of different shapes and sizes, even if they don't fit neatly on your regular shelves.
In technical terms, DLOs allow you to bring external data into Data Cloud and store it in its original format without forcing it into a predefined structure. This data lives in what we call a "data lake" – a centralized repository that can handle massive volumes of information.
Architecture Overview
Here's how Data Lake Objects fit into the Salesforce Data Cloud architecture:
Key Components:
- Data Sources: Your external systems (S3 buckets, SFTP, APIs, databases)
- Data Streams: The pipelines that bring data into Data Cloud
- Data Lake Objects: Storage for raw, unprocessed data
- Data Mappers: Connect DLO fields to Data Model Objects
- Data Model Objects: Standardized, structured data ready for use
Understanding the Data Lake: Where All Your Data Comes Together
One of the most compelling aspects of Data Lake Objects is their ability to handle every type of data your business generates. Let me show you what I mean.
The Three Types of Data
Your organization deals with three distinct categories of data, and here's the beautiful part—DLOs can store them all:
Structured Data
This is your traditional, organized data. Think databases, ERP systems, and CRM platforms. Everything lives in neat rows and columns with predefined relationships. It's the data you're probably most comfortable with.
Semi-Structured Data
Here's where things get interesting. Logs, XML files, JSON documents, and sensor data fall into this category. They have some organizational elements, but they're more flexible than traditional databases. Semi-structured data is becoming increasingly important as IoT devices and APIs generate more information.
This is the wild card—and it's everywhere. Documents, social media posts, images, videos, emails. This data doesn't fit into traditional database structures, but it often contains some of your most valuable insights about customer sentiment, brand perception, and emerging trends.
The Data Ingestion Process
All three types of data flow through a streamlined ingestion process into your Data Lake. This is where the magic happens. Instead of forcing your unstructured social media data into the same rigid format as your CRM data, the Data Lake accepts each type as it is.
Your Data Lake: The Central Hub
Picture your Data Lake as a vast reservoir of raw data storage. It sits at the center of your data ecosystem, collecting everything from structured database exports to unstructured customer feedback. Nothing gets transformed or lost—it's all preserved in its original form, ready for whatever analysis or use case comes next.
Notes:
DLOs Are Not a Replacement for DMOs
Don't make this mistake: Storing everything in DLOs and never structuring it. DLOs are for raw data; DMOs are for actionable insights. You need both.
Think of it this way: DLOs are your ingredients, DMOs are your prepared meals. You can't serve ingredients to customers.
Data Lake Objects vs Data Model Objects
Remember: DLOs and DMOs work together. DLOs store the raw ingredients; DMOs serve the finished product.
Key Characteristics of Data Lake Objects
-
Flexible Storage: Your Data, Your Way
Here's the thing about DLOs—they're refreshingly simple. You don't need to spend hours restructuring your data to fit some rigid database format. Got JSON files? Great. CSV files? Perfect. Something else entirely? No problem. DLOs accept your data exactly as it is. It's like having a storage space that says, "Come as you are," instead of making you jump through hoops just to get in the door.
-
Scalability: Room to Grow
Remember the last time you ran out of storage space? Yeah, not fun.
With DLOs, that's one less thing to worry about. Whether you're working with a few gigabytes today or planning for terabytes tomorrow, they've got you covered. Your data grows, your storage grows with it—simple as that.
-
Raw Data Retention: Keeping Your Options Open
This one's a game-changer. DLOs keep your original data exactly as you received it, untouched and intact.
Why does this matter? Because six months from now, you might want to analyze that data in a completely different way. Maybe you'll discover a new trend, or maybe you'll need to reprocess everything with fresh insights. With your raw data preserved, you always have that option. It's like insurance for your analytics.
-
Integration-Friendly: Connects Where It Counts
DLOs don't exist in a vacuum. They work hand-in-hand with the rest of Data Cloud's ecosystem, particularly with Data Model Objects (DMOs).
Think of it this way: your raw data lives in DLOs, and when you're ready to structure it for specific business needs, you can easily map it to DMOs. It's the best of both worlds—flexibility when you need it, structure when you want it.
-
Schema Evolution: Built for the Real World
Let's be honest—your data won't look the same a year from now. New fields pop up, requirements change, and business needs evolve.
The beauty of DLOs? They roll with the punches. When your data structure changes, DLOs adapt without breaking everything you've already built. No need to panic, no need to rebuild from scratch. Just smooth, continuous evolution.
-
The Real Deal: Why DLOs Matter?
Data Lake Objects give you something rare in the data world: true flexibility without sacrificing power. They let you store diverse data at scale, preserve your raw information for future possibilities, and integrate seamlessly when you're ready to take the next step.
Whether you're just getting started with Data Cloud or looking to optimize your existing setup, understanding DLOs is key to unlocking their full potential.
Creating Your First Data Lake Object: A Journey Worth Taking
There's something exciting about building your first Data Lake Object. It's like opening the door to a world where your data finally has room to breathe. Let me guide you through this journey—trust me, it's easier and more rewarding than you might imagine.
Step 1: Choose Your Path
When you're ready to create a new Data Lake Object in Data Cloud, Salesforce welcomes you with four beautiful pathways. Each one is designed with a specific purpose in mind, and picking the right one is like choosing the perfect starting point for your adventure.
Go to Data Cloud → Data Lake Objects
-
From External Files: Embrace the Raw and Real
This is where the magic begins for unstructured data. Have CSV files scattered across your systems? JSON documents filled with customer insights? This path welcomes them all with open arms. No preprocessing, no forced transformations—just pure, authentic data finding its home.
-
From Existing: Build on What You Know
Sometimes the best foundation is the one you've already laid. This option lets you breathe new life into existing objects, transforming them into flexible DLOs. It's like taking something good and making it even better, saving you time while honoring the work you've already done.
-
New: Your Blank Canvas
This is for the visionaries, the architects, the ones who know exactly what they want to create. Starting from scratch gives you complete creative control—every field, every data type, every detail crafted to your exact specifications. It's your masterpiece waiting to happen.
-
Create from a Data Kit: Stand on the Shoulders of Giants
Why reinvent the wheel when you can start with proven templates? Data Kits offer you carefully designed foundations, whether you need structured precision or unstructured flexibility. It's the perfect blend of guidance and freedom.
Click New Data Lake Object
Step 2: Bring Your Vision to Life
Once you've chosen your path (let's imagine you've selected "New"), you'll arrive at your creation space. This is where your vision starts taking shape.
Configure:
- Name: CustomerFeedback_DLO
- Label: Customer Feedback Data Lake
- Description: Raw customer survey feedback data
- Define the schema (fields)
The Foundation: Name Your Creation
Every great creation deserves a meaningful name. Your Data Lake Object Name should tell a story—make it descriptive, make it memorable. "Account_Home_copy" might not sound poetic, but it tells you exactly what it holds.
The API Name flows naturally from this, giving your DLO its technical identity. And here's something important: the Category you choose isn't just administrative detail—it shapes how your DLO interacts with the broader Data Cloud ecosystem, affecting everything from billing to data model mappings.
Step 3: Craft Your Fields with Care
This is where your DLO truly comes alive. Think of each field as a carefully chosen ingredient in a recipe—each one serving a purpose, each one contributing to the whole.
As you build your field structure, you'll define:
- Field Labels that speak in human language—clear, intuitive, welcoming
- Field API Names that whisper to the machines—precise, consistent, reliable
- Data Types that honor the nature of your information—Text for stories, DateTime for moments, Numbers for quantities
- Primary Keys that give each record its unique identity
- Record Modified Fields that remember when things change
Fields That Tell Your Data's Story
The example before us reveals fields that many DLOs share:
- Data Source: Where did this information originate?
- Data Source Object: What was its original form?
- Internal Organization: Which part of your business does it serve?
- Source Version: How has it evolved over time?
- Site or Location: Where does it belong in your world?
- Last Viewed or Modified: When was it last touched by human hands?
Need more? The "+ Add Field" button sits patiently at the bottom, ready to help you expand your vision whenever inspiration strikes.
Step 4: The Moment of Creation
With everything in place, you're ready for that satisfying moment—hitting save. Your Data Lake Object springs to life, ready and waiting to fulfill its purpose.
Save the Data Lake Object.
What Happens Next?
After creating your DLO, you can:
- Set up data streams to populate it
- Map it to Data Model Objects when you need structured access
- Run queries and analytics on the raw data
- Integrate it with other Data Cloud features
Architecture Best Practices
1. Data Organization
Best Practice: Use clear naming conventions for your DLOs.
Format: [Source]_[DataType]_DLO
Examples:
- Ecommerce_Orders_DLO
- ServiceCloud_Cases_DLO
- MarketingCloud_EmailClicks_DLO
- IoT_DeviceTelemetry_DLO
2. Partitioning Strategy
Best Practice: Partition large DLOs by date for better performance
Example Structure:
CustomerFeedback_DLO/
Benefits:
- Faster queries when filtering by date
- Easier data retention management
- Reduced processing costs
3. Data Retention Policy
Best Practice: Define how long to keep data in DLOs
Policy Example:
- Hot Data (0-90 days): Keep in active DLO, frequent access
- Warm Data (91-365 days): Archive to separate partition
- Cold Data (365+ days): Move to external archive or delete
Conclusion
Data Lake Objects are more than just storage—they're your gateway to smarter, more flexible data management in Salesforce Data Cloud. Start with one focused use case, build it thoughtfully, and let your success guide the way forward. With the right care and attention, your data lake becomes a powerful foundation for insights, innovation, and experiences that truly resonate with your customers. The journey begins with a single step, and that step starts with you.
Have questions? Learn more about our services at support@astreait.com or visit astreait.com to schedule a consultation.