Top 5 TrailblazerDX takeaways

Salesforce’s annual developer conference, TrailblazerDX, took place in San Francisco last month. This year, my primary interests were what is new and upcoming in the Salesforce ecosystem, including the latest in AI, Data Cloud, MuleSoft, Salesforce’s Well-Architected Framework, and enhancements to core functionality like setup. Here are my top five takeaways from the conference!

---

1. Einstein, Prompt Builder, and AI

Unsurprisingly given the hype, the new features leveraging the Einstein platform’s generative artificial intelligence capabilities were a central focus. As the technology matures, Salesforce continues to position AI tools to solve true business problems. The Einstein Prompt Builder enables declarative Salesforce administrators and developers to build methods for Salesforce to interact with a generative model like ChatGPT, using your specific data. For example, it can summarize a group of recent product reviews or case data and provide an “executive summary” while ensuring human validation of responses before storing them for reporting and future utilization.

A technologist using Einstein Prompt Builder will need to understand the best way to ground these requests to generate the best response for the business need.

---

2. Data Cloud

Salesforce’s Data Cloud is one of the key growth areas for the ecosystem. It is a product that has been evolving over time, since its original go-to-market as Salesforce CDP and Salesforce Genie. This powerful product serves as the foundation for many AI use cases, as it collects customer data from external systems and harmonizes it with your Salesforce data. Previous versions of Data Cloud may have presented this as a technology primarily for marketing, but now the focus extends to almost any use case centered around the three key types of their data object model.

The three types of data objects to consider with Salesforce’s Data Cloud are the Profile (which represents who – like a customer), an Engagement (which represents when something happened – a time-based activity like an order), or Other (which represents what was involved with a Profile or Engagement, like a product).

Some recent enhancements to Data Cloud include the capability to have zero-ETL integrations with Snowflake, and improved data collection to make better use of either predictive or generative AI. Unlike many recent Salesforce additions (think MuleSoft and Tableau), Data Cloud was built internally by Salesforce, so it is inherently more tightly connected to core Salesforce functionality. This interconnectedness creates advantages like being able to present data from Data Cloud as a Related List on a core Salesforce object like an Account or Contact, much like is possible with other objects.

---

3. MuleSoft and other integration technology

MuleSoft continues to be the leader in the marketplace for seamless system integration. Much like how changing a record in Flow can trigger an update to another Salesforce record, MuleSoft can be used to prompt cross-platform actions, such as automating a Slack message when a Salesforce record is updated.

MuleSoft is the foundation for integrating data from disparate systems and products, both within a Salesforce ecosystem and beyond it. There are also tools built on MuleSoft, like dataloader.io, that may be impactful for smaller organizations that lack the requirements for MuleSoft, but still need to load or extract data from Salesforce on a regular basis. Dataloader.io plans to add support for some database access in the future, which is anticipated to include support for more basic data loading use cases.

Even for those not using MuleSoft, Salesforce has announced key enhancements around Platform Events and Apex that allow users to focus on no-code or low-code declarative tools first, and then add minimal code in the places where it is truly needed. Invocable Actions in Apex allows the ability to fill in key bits of functionality using code. That code then becomes a building block for declarative developers and admins to use and reuse, including calling out to external systems beyond Salesforce. With Platform Events and the enhancements to the Pub/Sub AI, users can scale to a more robust, decoupled integration model.

---

4. Salesforce Well-Architected patterns and anti-patterns

The Salesforce Well-Architected Pattern and Anti-Pattern explorer uses the key concepts of Salesforce implementation – trusted, easy, and adaptable – as guardrails for the architecture decision-making process. Salesforce’s Well-Architected Framework provides the tools and sample diagrams to communicate and document Salesforce architecture to both business and technical teams.

---

5. Salesforce setup and core user enhancements

The Salesforce setup user experience for administrators has become more and more complex as Salesforce has continued to add features over the years; there is a clear focus on making this experience easier over the next several releases.

While Salesforce is no longer planning to sunset permissions on the Profile next year, the best practice of using a permission set-led permissions model remains. Throughout setup, using permissions sets will make administration more secure and straightforward. (See my colleague Kendra’s helpful blog post for more tips about permission sets.)

There are also new features coming to help admins set up User Access Policies and summaries of permissions. A common ask for new user setup is to “give them the same permissions as someone already in the system.” Ways to clone users, as well as other ways to make user administration more straightforward, are on the roadmap for upcoming releases.

Additionally, if troubleshooting user access to objects or fields is a pain point of yours, you can look forward to new tools and report types that will help diagnose access issues.

---

Summary and next year

TrailblazerDX provides an incredible opportunity for superusers like me to get hands-on experience with Salesforce’s new technology and latest releases, and connect with technical members of the Salesforce team, as well as the broader Salesforce user community.

This year’s DX event focused on how AI moves from hype to actual business use cases and solutions. Data Cloud helps deliver on the promise of collecting all your customer information in one central location to feed insights generated by AI or other analytics tools like Tableau. MuleSoft and other tools useful for integration, such as Platform Events, will help make your organization’s systems work together cohesively, creating a 360-degree solution. The Salesforce Well-Architected Framework will assist in making your Salesforce implementation trusted, easy, and adaptable. Finally, Salesforce’s enhancements to setup will help make administration more streamlined and able to address both new and upcoming features more effectively.

Much of the content of TrailblazerDX is now available on demand at Salesforce+. Next year’s TrailblazerDX event will take place in San Francisco and online on Salesforce+ on March 5 and 6, 2025!