Salesforce Data 360

AI & Agentforce Integration: It makes building context-aware agents in Agentforce much more effective without complex custom code. DataCloud triggered Flow: I appreciate how easily we can trigger standard Salesforce flows based on data model changes or calculated insights. Zero Copy Architecture: The ability to access data from external lakes (like Snowflake or AWS) without investing large amounts of integration effort.

Calculated insights are especially helpful in augmenting segmentation to personalize interactions with consumers. Zero-copy ETL minimizes the amount of data movement; would love to see this feature handle a more robust amount of data. The increasingly robust number of integration options to bring data in has minimized IT work and increased integration speed.

The most significant benefits we've found with Salesforce Data Cloud stem from a few key capabilities that drive a more intelligent and unified approach to customer engagement. 1. Identity Resolution: The platform's powerful identity resolution is a standout feature, as it intelligently matches and reconciles data about the same individual from various sources. This is essential for breaking down data silos and creating a single, unified customer profile, which serves as a foundational layer for all personalization efforts. 2. Real-time Data Processing: The ability to process data from ingestion to activation in less than a second is a game-changer. This real-time capability allows us to make faster, more data-driven decisions and act on a customer's live behavior, moving beyond static, batch-based campaigns to more dynamic and timely interactions. 3. Cross-Cloud Activation: Data Cloud's strength is its deep integration across the entire Salesforce Customer 360 platform. Unified data is not siloed; it can be leveraged to drive a wide range of actions, insights, and automations across Sales, Service, and Marketing, enabling a consistent and personalized customer experience at every touchpoint.

Complex Deployment: moving data cloud configuration between environments (like from sandbox to production) feels less mature than the core platform. Limited debugging visibility: it can feel like a black box when things go wrong. Troubleshooting why a specific record didn't unify or why a segment failed to publish is difficult due to the lack of granular error logs. Credit Consumption Management: Predicting and monitoring credit usage during the development phase is tough.

Need detailed insight into errors, the ability to stop processes which are stuck/long running, and the ability to receive error notifications of failed processes. Removing data from the system is challenging (to address changes in business needs, archival of data, etc. ). Need increased visibility into consumption and process dependencies to control costs.

The most significant challenges we've encountered are centered in three key areas: 1. Implementation Complexity and Time-to-Value: The initial setup is not a simple plug-and-play process and requires deep planning across data modeling, ingestion pipelines, and governance. The learning curve for data ingestion and transformation can be steep, which often leads to long implementation times before the platform's full value can be realized. 2. Usage-Based Pricing: The credit-based pricing model, unlike traditional per-user licenses, is usage-based, which can be difficult to predict and manage. Ingesting large volumes of data and performing identity resolution are the biggest drivers of cost, and miscalculating these can lead to unexpected budget overruns. The pricing structure can also be confusing to new users, and it requires continuous monitoring to avoid overspending. 3. Data Quality and Governance: The platform's effectiveness is entirely dependent on the quality of the data it receives. A major pitfall is ingesting large volumes of data without first ensuring it is clean and well-structured, which can lead to a lack of accurate insights and a poor outcome. Without a strong focus on data governance and an established data hygiene culture, the foundation for AI is weak, making it difficult for the platform to function effectively.