Lane Four

Author: Lyza-Jane DeVera

  • Why You Need a Credit Consumption Strategy Before Rolling Out Agentforce

    Why You Need a Credit Consumption Strategy Before Rolling Out Agentforce

    Teams are beginning to realize the potential of integrating Agentforce into their go-to-market workflows—it’s fast becoming a valuable layer in modern Salesforce architecture. But when the conversation shifts to cost, there’s sometimes a disconnect and we’ve seen this a few times with our own customers. 

    Agentforce doesn’t follow the familiar SaaS model of predictable monthly or annual subscriptions, and it definitely doesn’t run on goodwill and server space. It’s metered. Every prompt, every interaction, and every automated task taps into a credit-based system. And those credits? Well, they can disappear faster than you’d expect, especially when AI is being used at scale. The consumption model feels invisible at first. Until it doesn’t.

    So, what exactly are you paying for?

    Think of credits like digital tokens that get eaten every time you interact with the AI in different ways. Right now, there are three big credit buckets to know about if you’re bringing Agentforce into your Salesforce org.

    Conversation Credits

    These are the heavy-hitters, being the most expensive credit type. One “conversation” means a full 24-hour exchange between a user (customer or employee) and your AI. It doesn’t matter if they send one message or twenty. Once that back-and-forth begins, you’re on the meter.

    Why’s it pricey? Because this is where most of the magic happens—customer interactions, support resolutions, sales nudges, all that front-facing functionality that makes Agentforce shine. But it adds up fast. And that’s why you want your AI use to be doing something worthwhile; it’s either saving your agents’ time or driving new revenue, not just answering FAQs it could’ve pointed to in a knowledge base.

    Data Cloud Credits

    These credits are consumed whenever Agentforce uses Data Cloud; a common example would be the collection and storage of agent analytic data. Salesforce calculates credit consumption based on how Data Cloud is used and the amount of data that is processed. Businesses typically find that their service agreements include a respectable number of Data Cloud credits so this is usually not a significant extra cost.

    Einstein Request Credits

    Now this one surprises people. Every time your org calls on a large language model (LLM) such as asking Agentforce to summarize a case, draft a note, or generate content, you burn through Einstein Request credits. And since these tasks can happen multiple times per conversation, they stack quickly. Most agreements include a good number of Einstein Request credits, though they are typically not as plentiful as Data Cloud credits.

    The reality is that no two orgs will burn through credits the same way. It depends on how your Agentforce flows are designed, how your users behave, and even how you’ve configured your automations.

    A team that’s using Agentforce for customer triage and escalations will chew through conversation credits differently than one using it for sales coaching. So it’s not just about tracking consumption but rather about designing an Agentforce strategy with consumption in mind to begin with.

    We’ve seen more than a few teams get hit with the jump scare here. The use cases for AI are obvious, the team’s excited about it, but the reality of consumption-based pricing tends to come in late. So here’s the key takeaway: before you roll out Agentforce and turn your team loose, get clear on how it’s being used and what it can actually deliver when it’s running with purpose and intention.

    We’ve been building and refining tailored AI strategies across industries for the past eight months, and we know how to make AI pull its weight without burning through credits like it’s Monopoly money. So if you want the AI you’re excited about today to still make sense when the invoice lands next quarter, let’s chat.

    Let’s chat!

  • Salesforce CPQ Is Phasing Out: Here’s How Lane Four’s Getting Ahead of It

    Salesforce CPQ Is Phasing Out: Here’s How Lane Four’s Getting Ahead of It

    You’ve probably heard the buzz: Salesforce is winding down CPQ. And if your revenue operations are anywhere near Salesforce’s ecosystem, that might’ve caused a little heartburn. No one likes surprises, especially when they involve core parts of your business’s tech stack.

    We’re not going to act like this caught us off guard though—or that we scrambled when the news dropped. It’s tech. Change is part of the deal. At Lane Four, we’re always willing to move forward and find the best approach to mitigate any questions from our clients.

    So, we’ve been tracking the shift toward Revenue Cloud Advanced (RCA—previously known as Revenue Lifecycle Management or RLM) for a little while now. And while Salesforce isn’t making this transition optional, we’re not sitting around waiting for someone else to figure it out.

    We’re already building the bridge.

    What We’re Building 

    Our in-house architecture and development team members have started building a custom solution that helps migrate existing CPQ customers into Revenue Cloud Advanced. This will be the newest addition to the Lane Four Labs packages, and the goal’s simple: make the jump from CPQ to RCA feel more like a controlled hop than a flying leap into the unknown.

    We’ll be straight with you—this isn’t a final-state, abundantly polished, plug-and-play rollout. It’s still taking shape. But so is Salesforce’s RCA toolkit, and that’s exactly how dependable solutions come to life: by building, testing, and improving in real time. You could say it’s one of those “trust the process” moments, and we’re all in.

    Lane Four Labs CPQ-to-RCA Package: Code script (Not final)

    We’re actively designing tools to automate key parts of the process and are already testing early iterations within selected client orgs. The first iteration of our migration tooling is already functional: it pulls and inventories your CPQ configuration. Think of it like a pre-move checklist, but digital and automatic. From there, we’re layering in smart filtering, data transformation, and recommendations for how to move each piece into RCA.

    Lane Four Labs CPQ-to-RCA Package: Billing Object Mapping and Notes (Not final)

    How Migration Works Right Now

    Here’s the current migration process, which is subject to change as we learn, test, and improve:

    1. Run the Lane Four CPQ Migration Tool
      We pull every bit of your existing CPQ data and configuration automatically.

    2. Clean House (if you want to)
      Want to filter or delete old items cluttering your system? Now’s the time. Some of this is automatic already; more is on the way.

    3. Review + Recommend
      We take a close look at your CPQ configuration objects—like price rules, QCP scripts, and the rest—and map out how they align with Revenue Cloud Advanced. Some elements transfer over easily and can already be migrated automatically. Others need a few adjustments to fit the new model. Either way, we walk through everything with you, step by step. (And yes, we’re actively working on automating more of this.)

    4. Data Transformation & Re-upload
      We’ll transform your CPQ data into its new Revenue Cloud Advance form and handle the (re)upload.

    5. Test & Launch
      Final tweaks, tests, and…go live and enable!

    Lane Four Labs CPQ-to-RCA Package:  Exporting CPQ Configurationss (early view of what’s already working)

    We’re continuing to automate more of the process every week. While the full picture will naturally evolve over time, one thing’s certain—we’re moving as fast as we can, learning quickly, and making the solution better with every iteration. Transparency matters, and if we’re asking you to trust us with your systems, we’d better be open about where things stand. Interested to know more or even get a head start on your Revenue Cloud transformation? Let’s chat!

    Let’s chat!

  • Lane Four Recognized as One of the 2025 Best Workplaces™ in Canada!

    Lane Four Recognized as One of the 2025 Best Workplaces™ in Canada!

    We’re excited to share that Lane Four has been named one of the Top 100 Best Workplaces™ in Canada [Under 100 Employees] for 2025—for the second year in a row! This prestigious recognition is especially meaningful because it’s based on direct feedback from our team members, who are the heart of everything we do. The rankings reflect the quality and consistency of the workplace experience across all teams and roles, evaluated by the Great Place To Work® Trust Index™ employee survey.

    The Trust Index™ delves into the core aspects of a great workplace, such as trust in leadership, camaraderie among colleagues, and loyalty to the company. It also takes into account the diversity of responses across different demographics, ensuring that our workplace stands out for its inclusivity and equitable culture.

    “We actively strive to create an environment where everyone feels welcomed, respected, and empowered to share their lived experiences and contribute their unique skills and perspectives.”

    Andrew Sinclair, CEO

    This remarkable achievement speaks volumes about the commitment of every member of the Lane Four team. We’re proud to be part of such an inspiring workplace, and we remain dedicated to fostering an environment where everyone thrives.

    About Great Place to Work®:

    Great Place to Work® is the global leader in defining and recognizing high-trust, high-performance workplace cultures. With a mission to improve society by helping companies transform their workplace environments, they provide benchmarks, frameworks, and expertise for creating and maintaining exceptional cultures. The Best Workplaces™ in Canada list is part of the world’s largest annual workplace study, representing the voices of 11 million employees across more than 50 countries. The winners are chosen exclusively based on employee input—there’s only one way to make this list: your employees must put you there.

    The results will be published in The Globe and Mail’s Report on Business section on April 4, 2025.

    Our company is built on a foundation of core values and guiding principles that we hold dear. These values—Be Real, Be Human, Be Curious, and Do Good Work—guide our every move. They’re not just words on a page; they’re the essence of who we are. And our guiding principles—Empowerment, Mentorship, Accountability, and Authenticity—ensure that we stay true to our values in every aspect of our organization. From our internal culture to how we interact with clients and partners, these principles are at the heart of everything we do.

    Check out www.greatplacetowork.ca

    Follow Great Place To Work® on Facebook , LinkedIn and Twitter and use #BestWorkplacesCA

    About Lane Four:

    At Lane Four, our people are at the heart of everything we do. And at our core, is a set of deeply held values that shape how we show up every day—Be Real, Be Human, Be Curious, and Do Good Work. These aren’t just catchy phrases; they’re the spirit behind every decision we make. Supporting these values are the principles we live by: Empowerment, Mentorship, Accountability, and Authenticity. Together, they keep us aligned and grounded, influencing everything from our internal culture to the way we build relationships with clients and collaborators. This is what drives us—what keeps our work meaningful and our connections genuine.

    As a fast growing consulting firm, we know that our success is driven by the talent, passion, and well-being of our team. That’s why we invest deeply in programs that nurture professional growth, celebrate individual contributions regularly, and create a strong sense of belonging. 

    From our commitment to diversity, equity, and inclusion—including initiatives like our Women in Tech and Business community—to our robust professional development days, we strive to create an environment where everyone can thrive and grow. Our wellness program reflects our belief that personal health fuels professional excellence, offering support across nutrition, fitness, mental health, and community engagement. Regular team socials, tenure recognition, and a holistic approach to compensation and benefits round out an employee experience that’s as meaningful as the work we deliver. 

    Being named one of the Best Workplaces in Canada is a reflection of this unwavering dedication to our people—and another milestone we’re proud to celebrate together.

    Let’s chat!

  • Getting Clear With Your AI Journey: What Agentforce Should Do Starts with Why You Need It

    Getting Clear With Your AI Journey: What Agentforce Should Do Starts with Why You Need It

    Whether you’re still kicking the tires on Agentforce or already signed on the dotted line and waiting to put it to work, here’s something worth pausing for: where you start matters just as much as how you start. Are you implementing AI just to keep up with the buzz, or do you actually have solid use cases in mind which helped you make the decision to move forward? Agentforce is an example of a powerful AI tool, but without the right starting point and a clear approach, even the best AI tool in the market can turn out to be the most chaotic for your organization. (And yeah—we can help you avoid that. Just saying.)

    Everyone’s excited about AI right now. That’s fair…it’s exciting! But too many teams want to rush into purchase and implementation to keep up without slowing down to ask the more strategic question: what’s the right use case to start with, and is our data, workflows, and processes even ready for it?

    If you’ve ever seen an AI assistant confidently give the wrong answer, you already know—bad data doesn’t just confuse people, it erodes trust.

    Don’t Try Running Before You Can Crawl

    Start Simple. Build Smart.

    Let’s say you’re still in decision mode. You’re mapping out where Agentforce could have the biggest impact, whether it’s support, sales, customer success, maybe even internal operations.

    Or maybe you’ve already made the call to move forward, and now you’re facing a blank slate with the team asking, “Okay, what should we have it do first?”

    Here’s how we approach that at Lane Four: start small, with safe, low-risk use cases.

    Think about things like:

    • Helping customers locate existing knowledge base articles
    • Answering common questions using public product specs
    • Surfacing store hours, shipping timelines, or FAQ-style content

    These are great starting points because they’re built on information that’s already public and (hopefully) already accurate. It lets you test the AI in a controlled way, build confidence, and fix any weird edge cases before there’s anything sensitive on the table.

    Now here’s the part most teams overlook until it bites them: data governance. But when you’re implementing an AI assistant inside Salesforce, governance and deciding how to manage your company’s data isn’t optional. It’s the framework that makes the entire system usable and safe.

    You need to know:

    • What exactly the AI can see
    • What types of data are off-limits (think PII, financials, private notes)
    • Who controls access, and who reviews what it’s saying
    • How to set thresholds for when a human should review the response


    You’d be shocked how often a misconfigured field permission leads to AI responses that pull the wrong customer history—or worse, someone else’s.

    What About Those Bigger, More Valuable Use Cases?

    Once you’ve got a couple of safe wins under your belt and see ROI from your initial implementations, that’s when you can start thinking bigger.

    We’ve said this before, but in case you missed it, we usually map out use cases in ascending order of risk:

    • Start with public data/knowledge articles → easy wins, low stakes
    • Then move to region-specific or semi-structured data/knowledge articles → more topic-specific and available to larger group of verified users
    • Then non-sensitive user data → past orders, case history
    • Finally, secure use cases → authenticated experiences, promo logic, upsell paths, even payment flows


    At each level, with the risk increasing, the value can also go up, but     only if you have won the slow and steady race every step of the way. What we mean with this is if you haven’t mindfully built the right guardrails and rush into implementation without a clear approach, that overall value starts to come with baggage your team didn’t sign up for.

    When you’re ready to move into higher-risk use cases, it’s also more than okay to keep them monitored by a human team member/agent for a while. In fact, we recommend it until the AI proves it can handle things consistently and your team feels that they can trust it to run solo. Agentforce doesn’t need to be 100% autonomous from the jump. A human-in-the-loop model works great for approval workflows, refund logic, or anything revenue-impacting. Sometimes, even just a quick review before the AI sends a message is all you need to sleep better at night.

    A Friendly Reminder on Salesforce Org Hygiene

    You know how someone repurposed a field five years ago and no one documented it? Or how one object has 47 picklist values, but only five are still in use?

    Yeah. That’s the stuff AI loves to trip over.

    Agentforce works with what it’s given, so now’s the time to get your house in order. Categorize those knowledge articles. Run that field audit. Clean up old record types. Review permission sets. It’s like prepping your kitchen before bringing in a professional chef; doesn’t matter how good they are if the pantry’s chaos and they don’t know where to find the ingredients.

    So, Where Should You Start?

    If you’re evaluating use cases, keep one eye on value and the other on risk. Think: “What’s helpful and safe enough to test first?”

    If you’ve already bought in, but don’t want to stumble out of the gate, focus on something that lets you learn fast without wrecking anything. Either way, build your roadmap around data quality, governance, and management from day one. This will allow things to scale smoothly, once your foundation’s solid.

    And if you’re curious how other Salesforce teams are doing it, or just want a reality check on what not to do when implementing Agentforce, check out our other blogs or reach out to us. We’d love to chat!

    Let’s chat!

  • Structuring Agentforce the Smart Way: New Agent vs. New Topic

    Structuring Agentforce the Smart Way: New Agent vs. New Topic

    If you’ve ever stood in front of the Agent Builder screen wondering whether you should spin up a brand new Agent or just tack on another Topic—yeah, you’re not alone. It’s a decision that hits right at the core of Agentforce architecture. And while there’s no magic formula, there is a logic to it. 

    So, let’s break it down; when do you create a new Agent, and when do you just add more topics to the one you’ve already got?

    Think of Agents in Terms of Audience, Not Just Function

    Let’s clear something up: Agents aren’t just chatbots with fancy names. Each Agent is a decision-maker designed to serve a specific audience—and that distinction matters more than you might think. Sure, you might have multiple Agents operating within the same department or business function, but if they’re speaking to different user groups, using different systems, or serving different intents, they should probably be separated.

    So when you’re standing at the fork in the road—“Do I add a new Topic or spin up a new Agent?”—start by asking: “Who is this Agent meant to serve?” 

    If the audience changes, odds are, so should the Agent.

    Take Support, for example. You might have one Agent handling password resets and troubleshooting for existing customers. But now let’s say someone wants to ask about open roles or the application process. Technically, you could add a Topic about careers—but that audience isn’t a customer anymore; it’s a job seeker. Different needs, different voice, different context. That’s a solid cue for a separate Agent.

    Audience often correlates with platform, too. If you’re embedding one Agent inside your customer support portal and another on your external careers page, there’s a good chance those Agents shouldn’t be the same—even if they both roll up under the same department. Bottom line? Grouping Agents by internal function or department isn’t enough. Agent boundaries should be drawn around who they’re supporting and how they need to respond. Get that part right and the architecture will start to fall into place a whole lot easier.

    When a New Topic Makes More Sense

    Now, on the flip side, when you’re still swimming in the same pool, but just need another lane, add a Topic. What do we mean by this?

    Topics are the Agent’s “What do I do with this?” filter. Every user utterance is evaluated against the Agent’s Topics. So, if your service Agent’s job, for example, is already to support your app, and users now need help configuring notifications, or connecting their email, you don’t need a new Agent. You need more Topics.

    Each Topic includes:

    • A Classification Description (how to know when this Topic applies),
    • A Scope (what it covers), and
    • Instructions (what the Agent actually says or does)


    It’s kind of like giving your Agent a script for different customer questions. You’re not hiring a new rep, you’re just giving the current one more training.

    Rule of Thumb:
    Different Mindset or Audience? New Agent.
    Different Question? New Topic.

    That’s a solid litmus test. So, ask yourself:

    • Is this a fundamentally different kind of conversation?
    • Does it need different integrations, different data, or a different tone?

    Then it’s likely Agent-worthy.

    But if it’s just a branch of the same core task—same systems, same persona—stick with your existing Agent and expand the Topics.

    And hey, sometimes the answer’s hidden in the actions. If the new Flow you’re building taps into different systems or needs a different set of variables, that may push you into new Agent territory whether you planned on it or not.

    A Quick Note on Actions (and a Common Pitfall)

    Here’s a twist most people learn the hard way: Agents can only execute one Action per utterance. There was chatter about this loosening up around Spring ‘25, so if it’s been a while since you last checked, it might be worth poking around to see what’s changed.

    And if you’ve updated your Flow inputs or outputs, don’t forget: you’ll need to recreate the Action from scratch. The Agent won’t just magically figure out the changes. Include version numbers in your Action names now, and save yourself a headache later.

    TL;DR—but like, with context

    If you’re:

    • Crossing functional boundaries with a different audience → Create a new Agent
    • Just expanding a known use case → Add a new Topic

    Agentforce is powerful, but it thrives on clarity. Clean Agent boundaries and well-scoped Topics keep your architecture flexible, your maintenance sane, and your end-users happy. And honestly, that’s the real win. If you’re curious about how to set these up, check out this article. Need more guidance mapping our your AI journey? Let’s chat.

    Let’s chat!

  • Agentic Architecture 101: Custom Topics, Actions, and Best Practices

    Agentic Architecture 101: Custom Topics, Actions, and Best Practices

    When it comes to getting the most out of Agentforce, the real magic happens under the hood and more specifically, in how you design Topics and configure Actions for the Agents you’ve built. These aren’t just check-the-box setup tasks. They’re the logic and structure that guide every interaction your Agent handles. In this post, we’ll walk through the mechanics of setting up custom Topics and Actions inside Salesforce, highlight what actually matters during configuration, and share some best practices based on our own implementations—because getting it almost right is usually what breaks things later.

    Topics

    Topics are the first layer in an Agent’s decision-making. Every time the user makes an utterance, the Agent evaluates what topic it belongs to. A topic has three important components, all of which are evaluated by the AI:

    1. Classification Description: How an agent should determine when to use this topic
    2. Scope: A high-level description of what the agent can do in this topic
    3. Instructions: Specifics on what, when, and how an agent should perform certain actions once its decided to use this topic

    You can create new Topics from the Topics tab on the Agent’s page in Setup…

    …or from the Topics section of Agent Builder.

    You can add new or existing Actions to a Topic.

    You also don’t necessarily need to have any Actions on a topic. A common use case for this is where you want strict control over how an Agent answers a particular question or set of questions. You don’t want it to search through Knowledge and then summarize a relevant article; you want the same “hard coded” answer every time.

    In this case, you can create a Topic for this scenario, and provide instructions on exactly what you want the Agent to say.

    Actions

    Actions are how an Agent interacts with the rest of Salesforce. Any kind of Read, Create, Update, or Delete is done through an Action.

    There are a few types of Agent Actions:

    1. Flow
    2. Apex
    3. Prompt Template
    4. API
    5. Predictive Model
    6. Service Catalogue Item

    Flow Actions can call any “Autolaunched (No Trigger)” Flow.

    Variables that are marked as Available For Input, or Available For Output in the target Flow become the Inputs and Outputs for the Action.

    When you create the Action, you will need to provide instructions for the action as a whole, as well as for each input and output.

    The agent is pretty smart at determining what information from the conversation to plug in to each input, but you sometimes may need to provide explicit instructions on “this information goes into this input” on the topic-level instructions too.

    Another important thing to understand is Session Variables. These contain information from the Messaging Session. The Messaging Session ID is generally the most important information in here, and you will need to pass it in to Flows which create Cases, in order to link the Messaging Session to the Case.

    The Messaging Session can also be used to carry information from a Pre-Chat form, or about a logged-in portal user.

    *Note: Not all the Messaging Session fields are automatically included in the Session Variables. You may need to use the Edit Included Fields button to add other fields to the Session Variables

    In the following example Flow, the Agent sets the CaseSubject, CaseDescription, InputEmail, and MessagingSessionID inputs.

    The Flow queries Contacts to see if one already exists with the email address provided.

    If a Contact is found, a Case is created with that Contact linked. If no Contact is found, a Case is created with the provided email address entered in the SuppliedEmail field normally used by email-to-case.

    The Case Number is queried from the just created Case and assigned to the outputText variable, and the Messaging Session with an ID that matches the MessagingSessionID variable is linked to the created case.

    Prompt Template works similarly. They execute a Prompt Template, passing certain information into inputs, and getting an output once the Template has been run.

    A Prompt Template is a much longer and more complex set of instructions sent to an LLM for analysis. Templates can include merge fields and even search results. 

    The most common use case right now is creating a custom knowledge base lookup action. The standard “Answer Questions with Knowledge” action actually uses a standard Prompt Template, but that template can be overridden, and you can also create custom actions that call custom templates.

    Apex Actions can call any Apex class with an invocable method. On the Apex side, this works just like invocable methods for Flows. The parameters in the method will define the inputs, and whatever the method returns is the output. Remember that the inputs and outputs must be collections, even if the variables are singular in practice. 

    Best Practices

    • Currently, an Agent can only perform a single action for every user utterance. Chaining actions together is not currently possible. This capability is supposed to be coming some time around Spring 2025, so for now, keep this in mind when designing your solution. 
    • If you add or change the input or output variables on a Flow, the inputs and outputs on the Agent Action do not update. You will need to re-create the Action from Scratch. For this reason, we suggest including version numbers in your Action names for now. 
      • There is a way to update the Agent Action using the CLI.
    • In practice, we have found that it is possible to end a conversation from the Agent’s side by instructing it to “execute the End Conversation action”. 
      • This does work, but it seems to be unsupported functionality, and is very picky on precise wording in the instructions. Use caution and test thoroughly if using this technique. 
    • The order you enter instructions in isn’t necessarily respected when the instructions are given to the LLM. If you have certain steps the Agent must complete in sequence, include them all in one instruction block.
    • Consider having immediate requests from the User to create a case route through your Knowledge topic first instead. Have the Agent ask the User what the subject of the case will be, but instead of creating the case right away, first perform a knowledge search, and ask if the results answer the user’s question. If not, go ahead and create a case. If it does, we just diverted a case from being created.

    AI in Salesforce—especially when it comes to Agentforce—is still evolving fast. And like most teams out there, we’re learning as we go. The Lane Four advantage? We’re moving quickly, implementing what we know, testing constantly, and doing our best to share what we’re seeing in real time. If things still feel a bit fuzzy, it’s because the space is changing—fast. But we’re committed to moving just as quickly, being transparent along the way, and doing everything we can to stay one step ahead. Thinking the same way and curious about what makes the most sense for your AI strategy? Let’s talk.

    Let’s chat!

  • Lane Four’s Key Takeaways from TrailblazerDX 2025

    Lane Four’s Key Takeaways from TrailblazerDX 2025

    Another TrailblazerDX is in the books for the Lane Four team, and as always, the Salesforce ecosystem has a lot to talk about. From long-awaited updates to some we weren’t even expecting (but gladly welcome), this year’s conference brought announcements that will shape the way admins, consultants, and developers work with Salesforce tools—especially Agentforce (to no one’s surprise).

    If you missed the event or just need a refresher, our team of architects and devs gathered the most meaningful takeaways from TDX 2025. Let’s get into it.

    Source: Lane Four
    Left to right: Mathieu Hubbard, Anthony Mottola, Amy Blackburn, Nupur Patel, Pete Gilbert

    For Anyone Working with Agentforce…

    Agentforce 2DX

    Agentforce isn’t undergoing a full rebrand, but Agentforce 2DX signals that Salesforce isn’t done evolving its AI-powered assistant. While some of the announcement felt like a marketing refresh, a few key improvements hint at bigger things on the horizon. Here’s what Agentforce 2DX is expected to bring:

    • Proactive engagement based on data changes. Think of it as an AI that doesn’t just react—it takes action when something significant happens.
    • Autonomous operation within business processes. The long-term vision seems to be AI that handles workflows without needing constant human input.
    • Multi-interface interactions with rich content and media. Agentforce could soon move beyond simple chat responses, incorporating more visual and interactive elements.

    That said, don’t expect to see these features overnight. The timeline points to at least April 2025, and given how AI rollouts tend to go, we wouldn’t be surprised if some capabilities a bit longer to materialize or become fully reliable. For now, developers and admins shouldn’t expect any immediate workflow disruptions—though they might want to start thinking about how these new capabilities could fit into their long-term strategy.

    Free Developer Edition for Agentforce

    One of the biggest barriers to working with Agentforce has been the lack of a free, hands-on environment. Until now, sandboxes consumed production credits, discouraging experimentation and opportunities to learn. That changes with the new Agentforce Developer Edition, which makes it possible for admins and developers to explore features without worrying about extra consumption costs.

    Variables Within Agents

    Finally! This one’s for developers building Agentforce solutions and clients who need stricter response control. This long-overdue feature gives teams better control over how Agentforce handles prompts and responses. With variables, developers can fine-tune behaviours, making debugging easier and ensuring responses align with business needs.

    Custom LWCs Inside Agentforce

    For developers and clients who need custom UI elements in Agentforce, this update allows you to fully customize how Agentforce responses appear in Salesforce. While no official release date was given, the ability to embed Custom Lightning Web Components (LWCs) inside Agentforce will be a major advantage for teams that need tailored “visualizations” in addition to the conversational elements. 

    CLI for Agentforce

    A Command Line Interface (CLI) for Agentforce is coming! Before this, deploying and managing AI-powered agents without CLI support has been cumbersome, especially for teams with strict deployment processes. Now, with CLI for Agentforce, teams can integrate it into CI/CD pipelines, automate deployments, and manage configurations more efficiently. While this update isn’t as flashy as some of the others, it removes a significant pain point for developers—one that should have been addressed in the initial release. For organizations that rely on structured deployment workflows, this will make Agentforce a much more viable and scalable solution.

    Agentforce Test Tooling in VS Code

    Testing Agentforce interactions inside VS Code sounds exciting, but in reality, this looks like a reskinned version of the existing testing center. For developers and QA teams, this will improve usability but won’t dramatically change how testing is done—at least not yet. Expect more meaningful updates when Salesforce introduces more advanced testing capabilities, like chaining multiple test cases, which could make AI model validation much more effective.

    Interaction Explorer

    One of the biggest challenges in early Agentforce projects was the lack of transparency into how AI interactions were happening. Debugging, optimizing, and refining Agentforce behavior was often a guessing game.

    That’s why Interaction Explorer is one of the most exciting updates to our team. This tool will give teams deeper insights into Agentforce activity, making it easier to troubleshoot issues, track AI decision-making, and refine agent performance.

    For developers, consultants, and admins, this solves a major visibility problem, providing insights that help ensure AI-driven processes align with business needs. The timeline for release (~Spring 2025) is ambitious, and given the current reporting challenges in Agentforce, there’s some skepticism about how well it will work at launch. But if it delivers? This could be a massive step forward for AI-powered operations inside Salesforce.

    AgentExchange

    Salesforce is introducing AgentExchange, a marketplace where users can share and discover Agentforce use cases, templates, and best practices. In theory, this could become a valuable resource for admins and ISVs looking to accelerate AI adoption.

    The big question? Will the community embrace it? If AgentExchange gains traction, it could shorten development time, improve collaboration, and make Agentforce more accessible to a wider audience. But if adoption is low, it may end up as another underutilized Salesforce feature.

    For now, admins and consultants should keep an eye on it—but we don’t expect it to replace hands-on experience anytime soon.

    Modular Flow Design

    One of the recurring themes at TDX 2025 was modularity in automation, particularly in Flow design. Miss the session? Check it out on Salesforce+! The emphasis is now on breaking down complex workflows into smaller, reusable subflows—a shift that will make Agentforce-driven processes cleaner, more efficient, and easier to manage.

    For admins and consultants building flows, this reinforces Salesforce’s push for simplicity and scalability. The more modular the design, the more effective Agentforce can be in executing workflows without unnecessary complexity. This isn’t a revolutionary change, but it’s a best practice shift that will make long-term maintenance much easier.

    Agents Can Read Business Rules Engines

    A major update for Revenue Cloud users—Agentforce can now read Business Rules Engines (BRE), just as it does with flows. Since BRE houses pricing logic and validation rules, this means Agentforce can now analyze and apply pricing conditions dynamically.

    For admins and consultants managing pricing structures, this reduces manual intervention and ensures AI-driven interactions align with existing pricing frameworks. This is a fairly big deal for teams that rely on BRE for complex pricing and validation logic—AI-powered pricing guidance is no longer just a concept, but a reality.

    Agentforce Agent for Setup

    Managing user permissions has always been one of the most tedious tasks for Salesforce admins. The new Agentforce-powered setup assistant is designed to automate permissions comparisons, troubleshoot access issues, and even suggest configurations.

    For admins handling user management, this could save countless hours—especially when onboarding or troubleshooting why one user has different access than another. Future enhancements, like cloning users and object/field creation, could take this even further. This update brings AI-powered efficiency to one of the most frustrating parts of Salesforce administration—and that’s a big win.

    Prompt Builder 2.0

    Salesforce’s Prompt Builder 2.0 brings a redesigned UI and new functionality that makes it easier to develop, iterate, and manage prompts. One of the most significant improvements is the ability to compile prompts to resolutions without sending them to the LLM, which means no consumption credits are used during the initial testing and refinement phase. This drastically reduces the cost of prompt development, making AI-driven interactions more accessible.

    For admins and consultants, the update provides greater flexibility in aligning prompt inputs with flow logic, removing previous limitations that required direct flow input bindings. This opens up new possibilities for customizing AI-driven processes without costly workarounds. For clients hesitant to experiment with AI due to cost concerns, this change could make Salesforce’s AI tools a far more attractive option.

    Agentforce for Developers (A4D) Enhancements

    Salesforce is bringing AI automation to LWC development with key updates to Agentforce for Developers (A4D), set to start rolling into GA between Summer 2025 and Winter 2026. One of the biggest improvements is A4D Aura Conversion (GA in Winter ’26), which will automatically analyze and convert Aura components into LWCs while preserving their original functionality. This will ease the transition away from Aura and help developers modernize applications with minimal manual work.

    In addition, A4D Prompt LWC (GA in Winter ’26) introduces natural language-based LWC generation, allowing developers to describe components in plain text and have A4D generate the necessary code, including missing imports, methods, and functionality. To further streamline development, A4D Test Case Generation (GA in Summer 2025) will scan existing code and automatically create unit tests, significantly reducing the time spent on test coverage. These updates will enable developers to work faster, reduce errors, and focus more on innovation rather than repetitive coding tasks.

    Headless Agents via Agentforce API

    Agentforce is breaking free from traditional chat interfaces with headless agents, allowing AI to be embedded directly into workflows, applications, and backend processes. These agents handle real-time text-based prompts and responses without needing a visible UI—perfect for custom automation and scaling operations without disrupting user experiences.

    The new Agentforce API makes integration straightforward: just create a Connected App and add it to the Service Agent Configuration. A Service Agent is required for setup, and these agents can be used as invocable actions in Apex and Flow, making automation even easier. For developers, admins, and consultants, this update removes friction from AI deployments, making it faster and more flexible to integrate smart automation into existing systems.

    General Callouts For Salesforce Admins and Consultants…

    While Agentforce stole the spotlight, there were a couple other takeaways that make everyday admin work smoother.

    Improved User and Permission Management

    Managing users and permissions in Salesforce has historically been a cumbersome process, but upcoming updates promise to simplify and enhance control over access management. Some of the biggest improvements include:

    • Permission Set Summaries for better visibility into assigned permissions.
    • Field History Tracking for Users, set to enter beta this summer and go GA in winter.
    • More flexible permission set assignments, including the long-requested ability to set expiration dates on permissions.


    For admins responsible for user management, these changes will make it significantly easier to track, audit, and adjust permissions without tedious manual work. While much of this is still forward-looking, the roadmap indicates that Salesforce is committed to improving the admin experience in a meaningful way.

    Dark Mode (Yes, Really)

    After more than a decade of requests, Salesforce is finally introducing Dark Mode with the Cosmos theme. The impact? Functionally minimal, but psychologically significant. While it doesn’t change how the platform works, it does provide a more comfortable viewing experience for users who spend long hours in Salesforce.

    For admins, consultants, and developers, this may not be a workflow-altering update, but for those accustomed to staring at bright screens well past midnight, we’re celebrating with you. Sometimes, it’s the little things that make the biggest difference.

    Source: Lane Four
    Left to right: Anthony Mottola, Mathieu Hubbard, Pete Gilbert, Amy Blackburn, Nupur Patel

    For Data Specialists…

    Tableau Next

    Salesforce is retiring CRMA (formerly Einstein Analytics) and replacing it with Tableau Next, a BI platform built on Hyperforce and Data Cloud architecture. Unlike its predecessor, Tableau Next is designed to be natively integrated with Agentforce, allowing businesses to take direct action from within their data analysis workflows. It also supports the same connectors as Data Cloud, including Zero Copy, ensuring seamless data movement without unnecessary duplication.
    For BI teams, admins, and consultants who have relied on CRMA, this will require an adjustment period, as the tools and workflows will differ significantly despite some compatibility. For clients still using CRMA, this update signals that migration planning should begin sooner rather than later—though Salesforce has yet to announce an official sunset timeline.

    Search and Retrievers

    Salesforce is enhancing its retriever search capabilities, allowing for more precise, customizable, and context-aware data queries. Updates include:

    • Enhanced filtering on retrievers 
      • Pre filter-retrievers (currently available)
      • Ranking Factors; recency, relevancy and popularity (coming soon)
    • Search across multiple retrievers, expanding search scope across various data sources
      • Ability to combine and join DLOs, making cross-source searches more powerful


    For consultants and developers working with Data Cloud, these improvements mean greater precision in how data is queried and surfaced—a critical need for organizations with strict data governance policies. For clients with highly specific audience segmentation needs, these changes unlock new possibilities for refining search parameters where previous limitations existed.

    And last, but most definitely not least (since this conference was all about these individuals)…

    A Few More Takeaways For Salesforce Developers

    Formula Evaluation in Apex

    Salesforce is introducing Formula Evaluation in Apex, allowing developers to store formulas in Custom Metadata Types and evaluate them dynamically in Apex. This means business logic can be modified without requiring a developer to update or deploy new code—an especially useful enhancement for business-critical automations that need frequent adjustments.

    For admins and developers, this brings more flexibility and responsiveness to workflows. Instead of hardcoding conditions in Apex, teams can update logic on the fly, ensuring that automations adapt to evolving business requirements without waiting on development cycles.

    Javadoc-Style Code Documentation in Apex

    While developers have always been able to document Apex code with comments, Salesforce is now adding official support for Javadoc-style annotations, along with linting validation and auto-generation tools in VS Code. This change helps Agentforce for Developers by making it easier to understand method functionality at a glance, without digging through code.

    Most modern IDEs will display these docs when hovering over methods, offering context on parameters, expected outputs, and usage recommendations. While self-documenting code should always be the goal, this feature improves code maintainability—especially for teams working on large-scale, collaborative projects.

    Apex Cursors

    Currently, using OFFSET in SOQL queries limits developers to retrieving only the first 2,000 records—making it difficult to work with large datasets efficiently. Apex Cursors, now in beta, change that by allowing processing of up to 50 million records per cursor.

    For developers working with massive data operations, this is a huge performance boost. Cursors are also serializable, meaning they can be passed to Queueable classes—allowing for chained processing of large datasets without hitting governor limits. This update is essential for high-volume automation and reporting use cases.

    Delegated Polling for Async Actions

    Salesforce didn’t announce this one directly, but a TDX session that our team attended highlighted a powerful design pattern for delegating polling while waiting for deployments to complete. In traditional CI/CD workflows (e.g., GitHub Actions, Bitbucket Pipelines), the system continuously polls for deployment status—often wasting valuable compute time while waiting for tests to finish.

    With delegated polling, CI/CD pipelines can offload the polling task and receive a notification once the deployment is complete, significantly reducing unnecessary resource consumption. This is especially valuable for teams working under monthly execution time limits on CI/CD platforms. Developers looking to implement this approach can reference the example implementation here.

    For teams focused on deployment efficiency, this pattern saves time, reduces costs, and optimizes CI/CD workflows, making Salesforce deployments far less resource-intensive.

    TrailblazerDX 2025 delivered a mix of long-awaited improvements, promising new features, and a few announcements that left us wondering, “Will this really work as advertised?” Now, we wait to see how these changes roll out. If Salesforce sticks to its roadmap, 2025 is shaping up to be a pivotal year for the ecosystem. Want to experience the possibilities with trusted guidance? Let’s chat. 

    Let’s chat!

  • Why Data Governance Matters in Your Agentforce Implementation

    Why Data Governance Matters in Your Agentforce Implementation

    Bringing AI functionality into your Salesforce org—especially with a powerful tool like Agentforce— can feel like stepping into the future. But before you start letting AI handle customer interactions and internal processes, there’s a big question to answer: “Is your data ready for it?”

    Without a solid data governance strategy, even the most advanced AI can misfire. You might have the best algorithms in place, but if your AI is pulling from outdated, incomplete, or overly permissive data structures, you’re setting yourself up for a mess. That’s why we keep emphasizing data governance—it should be the foundation of any Agentforce implementation. So let’s break it down: why it matters and how to get it right.

    Who Sees What? The Critical Role of Data Access Controls

    Salesforce has long been a leader in data security, offering layers of built-in permissions and controls to keep information where it belongs. The new Agentforce AI features are simply layered on top of this existing security framework, meaning if there are cracks in your data access settings, AI will inherit them. 

    At the beginning of most AI projects, we perform an audit of our client’s Salesforce org—essentially a health check of the system—to determine AI readiness and recommend changes, if necessary. This is a reality check—are users seeing [only] what they need to? The most common red flag we see is that user profiles have more access than they should. Simply implementing Salesforce best practices with respect to data access will help to ensure that AI applications are sharing appropriate information with each user. Among users who should technically have the same level of access, there may be several different audiences who need different information. Fine-tuning access controls is as much about security as it is about ensuring AI delivers accurate, relevant responses.

    For Example: Let’s say your customer service reps handle different regions, but there’s no restriction on which cases they can access. If AI pulls insights from one region and serves them up to another, you could end up with compliance issues or just plain confusion. 

    When it comes to data hygiene, we love to use the phrase, “Garbage In, Garbage Out”. In the current context, we’re saying, “AI is only as good as the information it has to work with”. If it’s pulling from outdated or miscategorized knowledge articles, it’s going to give outdated or misleading answers. This is where structured knowledge management becomes essential.

    As a company that jumped in headfirst with Agentforce implementations, we’ve learned a thing or two about the role data hygiene and governance play in AI success. One key insight? Many organizations eager to adopt AI weren’t as prepared when it came to structuring and organizing the data that powers it. So, what did we do? We developed a method to help AI applications recognize the most common wording associated with a topic. To make this work, our clients update each knowledge article with a summary that includes alternate search terms—especially for product names, features, technical terms, acronyms, and industry jargon—ensuring AI can retrieve and interpret information accurately.

    Key Reminders when Organizing Knowledge Articles to Increase AI Readiness

    • Ensure knowledge articles are accurate and current: AI can’t fact-check in real-time. If an article hasn’t been updated in months or years, it could be spreading incorrect information.
    • Categorize content properly: If knowledge articles aren’t labelled correctly, AI may struggle to find the best answer, leading to irrelevant or incomplete responses.
    • Use alternate search terms: This is to prepare for people describing things in different ways. For example, an article titled “How to make a deposit” should include alternate search terms like “adding funds”, and an article about Salesforce’s Field Level Security feature should have “FLS” as an alternate search term. Including common synonyms, abbreviations, and acronyms in knowledge article summaries makes AI more effective.

    Implementing AI isn’t as simple as flipping a switch—it’s about making sure the right data reaches the right people at the right time. A well-structured governance strategy doesn’t just keep things running smoothly; it prevents security risks and misinformation from becoming major roadblocks.

    If you’re assessing your company’s AI readiness or already rolling out Agentforce, start by evaluating your data governance strategy. Who has access to what? How is your knowledge base structured? Is your AI learning from clean, well-organized data?

    AI is here to stay, but whether it works for you or against you depends on how well your data is managed. Let’s get it right from the start. Need some help with planning or execution? Let’s chat.

    Let’s chat!

  • Salesforce Spring ’25 Release: Lane Four’s Expert Highlights

    Salesforce Spring ’25 Release: Lane Four’s Expert Highlights

    Salesforce’s Spring ’25 Release is here and you can bet we’ve spent the last few weeks putting them to the test. From Flow updates that simplify documentation and execution to AI-driven workflow improvements, this release has plenty to offer and we’re eager to share.

    After combing through them, we’re breaking down the most impactful changes—along with the areas that (we think) still need work. After all, being a trusted partner means giving you the real story, right? That said, this isn’t an exhaustive list. If you work in the ecosystem, you’re no stranger to the fact that Salesforce moves fast, and more changes are still rolling out. We’re focusing on the features that most impact the work we’re doing with our current clients, but there may be additional updates that are crucial for your team.

    For the full list of enhancements, check out the official Salesforce Spring ’25 Release Notes and see if there are any updates that could make a difference for your org!

    For reference, here’s our Winter ’25 Salesforce Release Highlights to see how things are evolving.

    Flow Updates: More Power, Less Effort

    Direct Linking to Flow Components (Auto-Layout Only) → Ever struggled to explain exactly where a problem is in a Flow? Now, you can generate direct links to specific Flow elements in Auto-Layout. This is huge for documentation and collaboration. No more vague instructions or unnecessary back-and-forth. Just send the link, and your teammate (or future self) lands exactly where they need to.

    Screen Flow with Dynamic Action (Beta) → Salesforce is introducing auto-triggered screen actions in Screen Flows, allowing actions to fire based on user inputs without clicking “Next.” This means better interactivity and a more seamless user experience. However, since this is still in beta, we’d recommend testing carefully before rolling it out to end users.

    Join Collections with the Transform Element → Transform vs. Flow Loops: when do you use each? Salesforce is making it easier to process bulk data with the Transform element, reducing the need for loops in some cases. However, loops aren’t obsolete—you’ll still need them when complex logic is required. The takeaway? Use Transform when you can, Loops when you must.

    Attachments on Send Email → A long-overdue update, Flows can now send emails with attachments. This is a major win for automated communications—think invoices, reports, or personalized files—without having to rely on custom Apex or third-party workarounds.

    Flow Version History Inside Flow Builder → Say goodbye to external documentation for tracking Flow versions. With Flow Version History, you can now view previous versions directly inside Flow Builder. This makes troubleshooting and auditing much smoother, especially for complex automations.

    AI-Powered Productivity with Einstein and Agentforce

    Einstein for Flow & Formula Generation → The idea is great: Admins get AI assistance when creating complex formulas. But based on our testing, it’s not always accurate, so take AI-generated suggestions with a grain of salt.

    Agentforce Service Planner → This new agent type provides a to-do list for service teams. However, admins still need to create and configure these lists manually. While it’s a step toward better agent workflow management, it’s not a plug-and-play solution just yet.

    Agentforce Conversations in Omni-Channel Supervisor → Omni-Channel Supervisors can now monitor Agentforce conversations in real-time. This improves visibility into AI-powered interactions, which is great for quality control and training.

    Einstein Article Recommendations → While this feature could be very impactful for customer support, it too still needs some work. Right now, the recommendations aren’t always accurate, and adoption will depend on Salesforce fine-tuning the AI’s ability to pull truly relevant knowledge articles.

    Experience Cloud: Performance & Usability Upgrades

    (Beta) Experience Delivery → A major infrastructure upgrade for LWR sites, promising faster load times, better scalability, and improved security. If you’re building with the Build Your Own (LWR) template, this is worth a closer look.

    File Upload in LWR Sites → LWR sites can now support file uploads via the Lightning Web Component. No more need for Aura workarounds—finally!

    Record List Component (Beta) → A Record List component for LWR sites allows admins to display and filter Salesforce records. Think of it as a simplified list view builder for Experience Cloud.

    Source: Salesforce

    Unified Knowledge with Data Cloud → This allows organizations to connect internal and third-party knowledge bases directly to Data Cloud. This not only enhances AI-powered responses but also ensures that service teams have access to more accurate, well-rounded information for better customer support.

    Key Features:

    • First & Third-Party Knowledge Integration: Aggregate knowledge from both internal and external sources using Data Cloud connectors.
    • Retrieval-Augmented Generation (RAG) Updates: Strengthens AI-driven responses (e.g., Agentforce, Service Replies) by grounding them in a broader, more reliable knowledge base.
    • Increased Article Size Limit: Knowledge articles can now be up to 100 MB (previously capped at 131,000 characters). However, articles larger than 25 MB won’t be searchable for performance reasons.
    • Knowledge Article DMO – Scale enterprise-level knowledge, including structured and unstructured data (e.g., PDFs, transactional records).


    Note:     This isn’t a paid feature, but it does consume Data Cloud and Einstein credits—so keep an eye on usage.

    Revenue Cloud: Smoother Sales & Pricing Workflows

    Pricing Logs → Debugging pricing issues is now easier with detailed logs in the Pricing Operations Console—a big win for RevOps teams.

    Save Product Configurations → Reusing product configurations means faster quoting and ordering, cutting down on repetitive work.

    Improved Assignment Rules for DRO → Enhanced rule-based logic for Document Generation in Revenue Operations (DRO) makes manual task assignments more efficient. This reduces admin workload and ensures tasks are routed to the right people automatically.

     

    Revenue Cloud+

    Advanced Configurator (Closed Beta) → For those dealing with complex product setups, a new constraint engine makes setting up complex product validation rules more intuitive. Each product can now have its own configurator, with two setup options:

    • A Flow-like UI with drag-and-drop components for Admins—perfect for those who love a visual builder
    • An Advanced Configurator for developers needing more control

    Dynamic Revenue Orchestrator (DRO) → More flexibility in fulfillment plans—instead of waiting for order activation, you can now trigger fulfillment plans from Flow actions. Staged Assetization allows you to create and activate plans at any point in the process.

    Approval Orchestration Flow (No CPQ or Revenue Cloud license required!) → Salesforce is finally making approvals easier without needing a CPQ or Revenue Cloud license. New features include:

    • Groups as approvers—because sometimes, decisions need multiple eyes
    • Email notifications—so no more “I didn’t see it” excuses
    • Email-based approvals—approve directly from your inbox, no login required

    Revenue Cloud+ Billing & Other Account-Level Updates

    Consumption Management Add-On SKU → For companies selling usage-based products, new tools for managing usage-based selling, include:

    • Pack Products let you bundle usage-based offerings together
    • Account Wallets track usage progression at the account level, giving customers (and your billing team) a clear view of consumption

    Source: Lane Four

    On-Demand Invoice Generation for Accounts & Orders → Need an invoice right now? Instead of waiting for scheduled runs, admins can now generate invoices instantly for specific accounts or orders. This is a highly valuable update for companies with custom billing cycles or one-off charges.

    Suspend & Resume Billing for Accounts & Billing Schedule Groups → Whether it’s for contract pauses, seasonal customers, or account changes, you can now suspend and resume billing as needed—without canceling subscriptions or causing billing headaches.

    Bulk Invoicing Made Easier with New Invoice Scheduler Features → Ad Hoc Invoice Run: Need to generate invoices outside of the normal schedule? Now you can run a one-time bulk invoice process to handle last-minute changes.

    We’re sure the most recent Spring ’25 Salesforce releases deliver real gains in efficiency, but we’re always here to keep it real with you. Features like direct Flow linking, email attachments, and version history offer immediate value, while AI-driven tools are looking like they may still need some refining before they’re fully reliable.

    For admins and consultants, the key isn’t just knowing what’s new—it’s identifying what’s ready for production versus what needs more testing. Things are evolving fast, but successful implementation still depends on aligning updates with real business needs. Stay tuned as we continue to share more detailed findings as we continue to test out these new features!

    In the meantime, have questions on how these updates fit into your RevOps and GTM strategy? Let’s chat.

    Let’s chat!