How Salesforce Develop Their Platform (True to the Core)

True to the Core sessions are highly anticipated and consistently popular. These sessions could be described as an “ask me anything” to Salesforce executives and product managers. The audience can ask even the thorniest questions and get a transparent response, including why Salesforce are taking longer to deliver a feature (dependencies), and the tradeoffs they are making internally.

True to the Core (TTTC) is designed to connect Trailblazers with the product teams making decisions about the direction the product is going. This unfiltered view into Salesforce’s forward-looking roadmap makes TTTC unmissable sessions. Personally, I prioritize attending this session when it’s held, and you’ll find TTTC at every Dreamforce; it was transformed into a monthly live broadcast for a period of time on Trailhead Live. 

The “True to the Core” Journey

One of Salesforce’s greatest assets is the community of users, admins, consultants, and developers – the Trailblazers who evangelize the Salesforce platform on their own accord and, as a result, become active stakeholders in the product roadmap.

The IdeaExchange is an open feedback platform that uses a points-based system to rank ideas for Salesforce product enhancements. From 2007 (the IdeaExchange launch) to September 2021 (the most recently quoted figure), 3500 ideas were delivered from the IdeaExchange.

However, there was a growing discontent with how customer feedback was being actioned. Salesforce aimed to increase transparency with the community with multiple initiatives (shown in the image below). The two to focus on here are True to the Core sessions and the IdeaExchange prioritization cycles.

IdeaExchange Reimagined: Prioritization Cycles

The IdeaExchange was launched in 2006, and the cracks had been forming in how the IdeaExchange functioned: 

• Aging ideas: “This idea was created when Miley Cyrus was still Hannah Montana.” Aside from the comedy value, little movement on some ideas became apparent – for example, one quote from 2017: “Dependent Page Layouts have been on the IdeaExchange for a whopping ten years.. accumulating over 73,000 points, and 29 merged ideas.”
• Lack of feedback from Salesforce: Long periods of silence without updates from Salesforce gave some people the impression that ideas were being ‘swept under the carpet’. (Note that communication from product managers has significantly improved since this time.) 
• Ideas or bugs?: Some ideas posted were considered bugs (a limiting flaw in current features) versus an idea which is a future ‘wish list’. These belong in a different place (i.e. bugs should be logged on Known Issues). 

Come 2018, Salesforce knew that it needed a smart way to capture and rank customer-requested ideas for the future Salesforce that would remedy the issues listed above.

On the outside, we don’t know how complex an idea would be to implement. From our surface-level point of view, we don’t truly know the intricate dependencies between features and functionality.

The driving ethos was a platform that would democratize which items would join the product roadmap, using a points-based system to fairly rank ideas that could be feasible for the engineering teams to work on.

IdeaExchange prioritization list takes the user wish list and marries it up with reality – how Salesforce product development actually happens, taking complexity into account.

The IdeaExchange prioritization cycles have been running since 2019 – a feedback loop that means customers get heard but aren’t left in the dark on why popular ideas aren’t up for the roadmap development. 

To launch the reimagined IdeaExchange, Salesforce took a novel approach during the Dreamforce session – to have Salesforce executives read out scathing criticism from Twitter. While the video is no longer available, you will have to take our word that it was great entertainment!

How Salesforce Develop Their Platform

Salesforce has five innovation themes: The Trusted Enterprise, Salesforce Easy, Cross-Cloud Integration, Hyperforce, and Customer-Focused Innovation.

Let’s highlight Salesforce Easy – one that has come up in conversation since Dreamforce ‘21. 

Salesforce Easy

Did you wonder how Lionel Richie and Salesforce are connected? “Ooh, that’s why I’m easy, I’m easy like Salesforce Easy”

Salesforce Easy is the movement towards all setup, installation, and configuration on the Salesforce platform to be as easy as possible. The executives shared their experience walking through the setup process in admins’ shoes for various Salesforce products.

Their conclusions? Ease of setup was not what they expected, and there should only be one way to do something: “We just add, add, add, and we rarely simplify.”

That’s where initiatives come in, such as: 

• Retiring Workflow Rules and Process Builder, and instead, delivering Flow, migration tools, and Flow Orchestrator. 
• MuleSoft Composer: An interface that allows low-code professionals to integrate Salesforce with other systems using clicks, not code.
• Renaming products: In a move to make Salesforce products easily understandable to outsiders, major examples of renaming are a handful of the studios and builders in the Marketing Cloud suite and the long-standing brand Pardot, now renamed Account Engagement.

Aside from Salesforce Easy and the other four high-level themes, innovations that Salesforce professionals can relate to can be categorized as:

• The admin experience 
• The Lightning platform UI
• The end user experience
• Collaboration
• Automation

The Admin Experience

Salesforce Admins have always been core to Salesforce success, embedded within Salesforce customers and working to maximize their orgs. This section covers a number of requests, admin ‘bugbears’, and where Salesforce’s vision for platform configuration is heading:

• Salesforce Setup: The Setup Tree and ‘Admin Home’
• Flow as an Admin’s Automation Tool
• Process Builder and Workflow Rules Retirement
• What’s up with “Modify All Data”?
• Why is Dataloader.io not on the Platform?

Salesforce Setup

“Salesforce Setup (or what we call “setup”)” – said Jon Sigler, EVP Salesforce Lightning, when asked for his top priorities for the future. 

He went on to explain that the term could be considered misleading, as Salesforce Setup is not only used for the initial setup of a Salesforce org but instead, is where admins go to continually enhance Salesforce. In short, ‘setup’ does not justify the ‘life of the org’.

Similarly, the Salesforce Setup ‘tree’ (i.e. the way that items in the sidebar menu are collapsible) and the ‘admin home’ were areas that Salesforce focused on improving. Previously, the setup sidebar collapsible menu didn’t expose everything an admin required. This has now been changed to launch into specific areas – for example “Users”, “Company Settings”, etc., which align with the admin’s flow of work. 

Flow as an Admin’s Automation Tool

“How can Flow be supported as an admin tool?” asked Scott from the audience, then relayed previous experiences of raising Salesforce support cases, to only be turned around in circles. 

Wind back two years ago, Salesforce Flow was considered a developer tool in terms of support, which mismatched the messaging that Flow is the go-to automation tool for Salesforce Admins.

All executives on the panel agreed that support offerings do need to match the product messaging soon now that it’s on everyone’s radar.

Process Builder and Workflow Rules Retirement

Following the announcement that Process Builder and workflow rules were scheduled to be retired, word spread fast around the Salesforce community. 

At the time, the product manager responsible for the retirement, Patrick Stokes, delivered some peace of mind – that a formal end-of-life roadmap, governed by an end-of-life council, would be put in place. In other words, a gradual process over the course of years, not months. 

So far, we are no longer able to create new workflow rules, and we’ve been given the workflow rules → Flow migration and the Process Builder → Flow migration tools. In the Summer 2023 release, Salesforce will turn off the ability to create new Process Builders. Salesforce is working towards the end of life for both workflow rules and Process Builder in 2025.  

Read the full roadmap and what you need to know in the guide below.

What’s up with “Modify All Data”?

This conversation started years ago when admins pointed out the flaws in this all-or-nothing, very powerful user permission. Put it into the wrong hands and “Modify All Data” could be disastrous.

Cheryl Fieldman, now the product manager for authorization (including permissions), was one of the original Trailblazers leading the conversation. Cheryl said she’s passionate about splitting up the “Modify All Data” permission – on a ‘permission mission’ you could say! 

One aspect she’s looking for feedback on is how “Login As” should be split from “Modify All Data”. Although there are no timelines for any changes, she promised to revert back with a plan at the first opportunity.

Why is Dataloader.io not on the Platform?

Dataloader.io is a web-based application that plugs into Salesforce to import, upsert, and remove record data. Core to an admins’ toolbox, it offers more functionality than the other Salesforce data loader options.

Originally developed by MuleSoft, Dataloader.io has since been brought into the Salesforce family through the MuleSoft acquisition. Yet, it’s still a freemium product – and it’s still off-platform.

“Where Salesforce came from 20 years ago, there was (relatively) next to no data”, the panel reflected. Salesforce native data loaders didn’t need to be as sophisticated as they need to be in today’s world. “We should consider this”, the panel agreed.

“We had to show our product implementation end-to-end (at an executive offsite) and discovered too many ways to load data, and most of them aren’t great. Where we have 12 ways of doing something… (that’s) technical debt we’ve accrued over time… we’re in a spring clean mode.” 

The Lightning Platform UI

Page load times in Lightning get airtime during TTTC sessions. 

Performance has been the focus since organizations began transitioning from Salesforce Classic to Lightning – and will continue to be. You only need to read “Lightning Speed Please” on the IdeaExchange to see how hot this topic is.

There’s some positive news on the topic. Scott Yancey, SVP Software Engineering UI Platform, shared that over a period of four years, median page load times improved by 60%. Note that this is a median figure – some orgs have seen even greater improvements than 60%.

So, how have the improvements been achieved?

• Optimizing core Aura codebase, which powers Lightning.
• Adding a CDN (content delivery network).
• Verifying any new features perform well on Lightning to prevent any ‘backslide’.
• Sharing the responsibility of improving page load times with Salesforce Admins. With Page Analyzer, performance is based on the “Predicted Page Load Time” metric, which considers the specific configuration of the elements on the Lightning app page. Salesforce Optimizer also gives insight into underutilized or overburdened Salesforce features, with actionable tips to remedy performance issues. 

After all, it’s not just Salesforce’s responsibility. There are multiple factors that affect page load times – users’ browser speed, network latency, and number of cores are all contributing factors that Salesforce can’t control. 

For the page load time factors that Salesforce can control, Salesforce are striving for improvements by:

• Moving the core Lightning codebase from Aura → Lightning Web Components (LWCs): A challenging process that started with the ‘heavy-weight’ components e.g. record detail and the highlights panel. Within the next year, list components will move over, predicted to have another significant impact.
• Advance network routing and caching capabilities.
• Continue to advocate for org customization efficiency, to warn and inform admins about customizations that could negatively impact page performance, plus guardrails that will help prevent performance issues in the first place.

“The process of moving from Aura to LWC clearly takes time – but it’s an investment of time. An LWC codebase will be the foundation that allows for more agility to develop and enhance components.”

The End User Experience

End users, the people using Salesforce day in and day out to be more productive at work, are clearly key stakeholders. A couple of themes that fall into this section are: 

• Mobile and Offline Capabilities
• Classic vs. Lightning Feature Parity

Mobile and Offline Capabilities

The customization capabilities we have with desktop Lightning were extended to Lightning on mobile. As we know, the proportion of desktop users versus mobile users is a scale that’s continually tipping in favor of mobile.

LWC has been driving this because it can be cached and then ‘served up’ on a device even when the user is offline (known as ‘statically analyzable’).

The product manager for Salesforce mobile shared that Mobile Publisher had seen more adoption, resulting in more requests for extensions and offline capabilities – the focal themes have been:

• Mobile Home: Users can personalize their Salesforce home screen on the mobile app (now delivered).
• Nimbus: Opens up access for developers to use the same software that the Salesforce mobile app can tap into, such as address book, GPS, and QR code scanning.
• Dynamic Forms for mobile (now delivered).
• Offline support for the Salesforce mobile app, bringing offline capabilities while continuing to use the rich customization of Lightning, regardless of whether you have an internet connection (now delivered).

Classic vs. Lightning Feature Parity

Lightning was developed as the Salesforce UI that can support customers for at least 20 years. 

Lightning is not an extension of Classic: “Not a reskinning, updates of font and colors… moving to Lightning is a change management experience, it’s not a ‘lift and shift’”.

Lightning was brand new, which meant the team spent significant time considering what to bring forward and what to leave behind.

Salesforce stated they have no intention of full feature parity. Instead, they make suggestions for new features that replace Classic gaps. Wind forward to today, and the remaining gaps are minimal:

Collaboration

Slack Developer Experience

“Which platform do you want us to develop as a joint company?”, asked Brett Taylor in 2021. 

Would your answer be Slack?

Unifying the developer experiences between Slack and Salesforce has been a theme over the past few years. 

The power of Slack is that it can be integrated with tools beyond Salesforce to create collaboration experiences that wouldn’t ever be possible to create in Chatter. A benefit of Chatter is that it can be embedded in the Salesforce interface. 

So, will Slack ever be embedded in Salesforce?

Rather than asking: “Can it be embedded in Salesforce… focus on the workflows, focus on the use cases”, advised David Schmaier, Chief Product Officer.

Salesforce productized their own requirements – in other words, acting as the first customer for use cases that will eventually be launched into the product offering. 

Since then, we’ve seen multiple Slack-Salesforce releases, including (but not limited to): 

• Slack Apps for Sales Cloud, Service Cloud, and Marketing Cloud.
• Flow in Slack (clicks/low-code)
• Apex SDK for Slack (pro-code)
• A new set of developer tools and modular, reusable building blocks.

If you’re interested in learning more, don’t forget to check out the following resources:

• The New Salesforce Platform for Slack: Flow, Apex SDK & More
• Slack Announces New Developer Platform 

“We Face a Complicated Balance of Prioritization”

The ‘platform’ team at Salesforce provides the services for other ‘cloud’ products for customers and partners to build on top of.

When a specific feature initially comes to this team to enhance, they often find it has potential applicability across multiple product areas. For example, Service Cloud could develop a messaging feature, then the platform team could see it has the potential to be extended and made available as a shared service across multiple platform areas (e.g Sales Cloud).

Often, a requested feature means that the most sensible direction for the team is to take ‘two steps back’, and make new features both performant and extensible.

The process of moving from Aura to the Lightning Web Component (LWC) codebase clearly took time – but it’s an investment for the future that will serve as the foundation for more agility to develop and enhance components.

One idea that was floated within both Salesforce internally and the community is to have a release dedicated to solving performance issues (minus any new features). However, that’s not what Salesforce considers the best allocation of their engineering resources:

“We can make Lightning screaming fast, but if it lacks critical features your business needs, that’s not going to be helpful”.  

Plus, we in the community don’t always see the full picture. The product leaders used inline editing as an example of what’s going on behind the scenes:  

‘While particular items, such as inline editing on related lists, are not marked as “In Progress” right now, they actually are. We’re on a journey that requires that we first deliver the aura capabilities in LWC… which will unlock the rich experiences people are looking for us to deliver.’

Tough Cuts

Tough cuts are the requests that the product leaders may have wanted to get done but haven’t been able to with a full pipeline and, of course, always striving to meet the highest standards for releases. With an active IdeaExchange and regular prioritization cycles, there come tough decisions on what to keep and what to cut.

Summary

True to the Core’ sessions are highly anticipated and consistently popular. The audience can ask even the thorniest questions and get a transparent response that includes why Salesforce are taking longer to deliver a feature (dependencies) and the tradeoffs they are making internally. On the outside, we don’t know how complex an idea would be to implement. From our surface-level point of view, we don’t truly know the intricate dependencies between features and functionality.

IdeaExchange prioritization list takes the user wish list and marries it up with reality – how Salesforce product development actually happens, taking complexity into account. Paired with True to the Core’ sessions, we’re getting a deeper understanding of where Salesforce are heading (and when!). 

Follow the hashtag #TTTC to keep up with the conversation on Twitter.