TechnologyDec 18, 2018

Reflections on Dreamforce 2018: Innovations, Favorites, & Einstein

Brian Hafferkamp

It’s been a few weeks since I had the privilege of attending and presenting at the annual Salesforce conference, Dreamforce, and I’ve finally had the chance to sit down and process all of the information presented there. It was an overwhelming, busy, and exciting week, filled with session after session of upcoming features and innovative Salesforce solutions.

custom development

I spent most of my time attending sessions on custom development in the Salesforce ecosystem, focusing mainly on Developer Experience (DX). There are some major innovations coming to DX that will make developing and devops in Salesforce a breeze. Here are three major takeaways I’m excited about:

1. Snapshots

One of the major pain points of using DX has been the configuration of the Scratch Org. While the tool is a vast improvement to previous solutions, there is no way to automate certain org settings that may be needed for more complex applications. This results in a developer having to manually configure the Scratch Org after creation depending on the needs of the environment. With the introduction of Scratch Org Snapshots, developers will be able to create a template for their Scratch Orgs so this manual configuration will no longer be necessary. This will be a huge timesaver and will open up the door for more automation possibilities in complex scenarios.

It is important to note that Snapshots will still require a small amount of manual setup. Users will have to configure the first Scratch Org, which will be used to generate the template.

2. Packaging and Dependency Management

While this has been possible since the release of DX, packaging will see some improvements in the near future. This isn’t necessarily due to an upcoming feature, but rather through the conception of a Salesforce development framework known as Force-DI and a corresponding set of libraries. The work done by Andy Fawcett, an active Salesforce contributor and vice president of product management at Salesforce, and his colleagues will allow Salesforce developers to introduce dependency injection to their development lifecycle, something that has been sorely-needed. With the introduction of this framework, we will begin to see decoupled Salesforce services, modularized organizations, and mocking for unit tests.

3. Enterprise Message Platform (EMP)

EMP is not directly related to DX, but it is still a feature that is bringing Salesforce up to speed in the development world. EMP introduces an event-driven architecture to Lightning (Salesforce’s React-esque front-end framework). With an EMP component, a developer is able to generate a platform event, which can be received by multiple listeners to take further action. This will result in simplified code for more complex action. More information on EMP events can be found here.

favorite sessions

In addition to the innovations I’m excited about, there were a couple sessions that made a big impression.

Top Custom Development Session

Advanced Techniques to Adopt Salesforce DX Unlocked Packages was not only my favorite session for custom development, but my favorite for all of Dreamforce. It was so jam-packed with information that I actually attended it twice, and felt like I could’ve gone another time. John Daniel and John Storey went super in-depth on the best practices for implementing the Force-DI framework and the various related libraries in order to write the best possible code in the Salesforce ecosystem. If you are interested in introducing well-written code into your Salesforce organization, this video is a must-watch.

Top Marketing Cloud Session

My favorite of the marketing cloud sessions was Advanced Audience Targeting Using Salesforce DMP, a hands-on workshop that gave me insight into the power and simplicity of DMP. This tool gives you the ability to use first-, second-, and third-party data to send consumer-specific content. Additionally, you are easily able to capture and act off of platform and site events. Within minutes, we set up an email to send a special offer to users who had placed a certain category of item in their cart but hadn’t yet closed the deal. In the moment, it felt straightforward and easy; but as I reflected, it amazed me how easily I had accessed and used this data to generate and send a custom-tailored email, all in a software I hadn’t even seen 30 minutes before. DMP also has a built-in tool called Segment Overlap, an easy-to-use app that shows overlaps in your audience groups to better drive marketing decisions. All-in-all, DMP is an easily accessible and user-friendly tool that allows marketers to get the most out of their data.

Einstein Analytics

Finally, I was overwhelmed with how many good sessions there were on Einstein—the artificial intelligence app for Salesforce—and I tried to attend as many as I could. The more I learned, the more I realized that Einstein truly does have applications in any part of a Salesforce org. Here are two key takeaways that I’m excited about:

1. Einstein Platform

While Einstein Analytics was previously powerful, it was limited in that it could mostly act on standard Salesforce objects, even though 80% of customer data resides in custom objects. Einstein Platform allows us to use Einstein Analytics to build predictions and discover trends on data stored in custom objects. Prediction Builder, an app in the Einstein Platform, allows you to easily create and train a model that can begin to drive business decisions. It provides a score so you know how accurate your model is, and shows how changes in key metrics will affect outcomes. With Einstein Discovery, you have the ability to analyze data across multiple dimensions, and paired with Einstein Stories, you can run data analysis over every permutation across every dimension. Again, the name of the game is simplicity, as the platform provides easy-to-use declarative tools to create, train, and use these models.

2. Einstein Voice

During the main keynote, Marc Benioff revealed Salesforce’s response to Siri and Alexa: Einstein Voice. This smart assistant is more than just a speech-to-text helper: it can update data, drive Salesforce functionality, and even provide voice technology to Einstein Bots. Using natural language processing, Einstein Voice is able to analyze a voice memo—say, after a meeting—and create actionable items or update records. The following video shows Einstein Voice in action:

Dreamforce 2018 did not disappoint. Between learning about the latest innovations, presenting in front of and networking with other Field Service Lightning users, and seeing Metallica perform live, it was a time I won’t soon forget. I’m excited to see what these new developments mean for the future of Salesforce. It’s clear they’re not slowing down after their recent success, and I look forward to what new features they reveal next year.

If you’re looking for help with Salesforce development, reach out to us at We are practitioners, not salespeople, so you’ll speak with an actual Credera consultant. We look forward to chatting to see how we might be able to help.