Nithya Konduru

Nithya Konduru is a content strategist and growth marketer with a background in biomedical engineering and medical science. She specializes in SEO, demand generation, and content strategy across healthcare and health tech, helping organizations translate complex topics into high-performing, conversion-focused content. She has led content and growth initiatives across startups and scale-ups, driving significant increases in organic traffic and user acquisition. Nithya brings a data-driven, user-first approach to building content systems that support both visibility and business growth.

Three corporate workers on a laptop
Blog, Salesforce CPQ

Salesforce CPQ End-of-Sale: What RevOps Teams Need to Do Now

June 9, 2026

If you’re running revenue operations on Salesforce CPQ, you’ve probably seen the announcement. In March 2025, Salesforce officially declared CPQ “End of Sale.” No new customers. No new features. No product investment. The product is frozen. What most organizations haven’t done yet is act on it. Fourteen months in, the majority of CPQ customers are still in wait-and-see mode. That window is closing, and the cost of waiting is no longer theoretical. Where Most Teams Are Right Now (And Why That’s a Problem) The majority of organizations still on Salesforce CPQ filed the EOS announcement under “deal with later.” That made sense in Q2 2025. It doesn’t make sense anymore. Here’s the timeline playing out in real time: 2025 to 2026: Support slows. Bug resolution takes longer. Salesforce begins redirecting partner and internal resources away from CPQ 2026 to 2027: Aggressive Salesforce campaigns begin pushing Agentforce Revenue Management migrations. Expect upsell pressure, reduced legacy discounts, and expiring incentive offers 2027 to 2028: Formal End of Life announcement expected, with migration deadlines attached 2029 to 2030: Full support sunset, based on typical enterprise software patterns. Salesforce has not confirmed an official EOL date The companies that start now get to make a deliberate decision. The ones that wait until 2027 will be making a reactive one, under contract pressure, with fewer implementation partners available and less negotiating leverage. That window is closing faster than most executives realize. The Honest Truth About Migrating to Agentforce Revenue Management Salesforce rebranded Revenue Cloud Advanced to Agentforce Revenue Management at Dreamforce 2025. It’s a meaningful evolution, with AI agents embedded directly into quoting, contracting, billing, and renewal workflows. The platform is genuinely compelling for organizations that are ready for it. For a full breakdown of how it compares to legacy CPQ, read our guide here. But Salesforce’s sales motion can obscure some realities worth knowing before you commit. It’s a reimplementation, not an upgrade. Agentforce Revenue Management is built natively on Salesforce core using standard objects. Legacy CPQ runs as a managed package. These are fundamentally different architectures. Your pricing rules, product bundles, CPQ scripts, and Quote Line Editor customizations don’t carry over. Everything gets rebuilt. The timeline is longer than the pitch deck suggests. Most enterprise migrations run 12 to 18 months. Organizations with complex product catalogs, multi-territory pricing, or deep ERP integrations should plan for the longer end of that range, and budget for it accordingly. Licensing costs will go up. Agentforce Revenue Management is priced at a premium over legacy CPQ. Factor that into your business case before you start building one. The platform is still maturing. The Winter ’26 and Spring ’26 releases have added meaningful improvements: multi-order creation from a single quote, new pricing formulas, and deeper Agentforce AI integration for forecasting. But some capabilities that CPQ customers take for granted are still being developed on the new platform. Evaluating ARM today is different from evaluating it in 2023. None of this means Agentforce Revenue Management is the wrong answer. For many organizations, it’s exactly the right one. But going in with clear eyes about the scope is what separates a successful migration from a troubled one. 5 Things RevOps Teams Need to Do This Year 1. Audit Your CPQ Environment Before Anyone Touches Anything This is the step most teams skip, or do too quickly, and it’s the one that causes the most pain downstream. Before any vendor conversation, document: Every active product bundle, pricing rule, and quote template currently in use All custom scripts, flows, Apex triggers, and third-party integrations connected to CPQ Where quotes break down and require manual workarounds: spreadsheets, offline approvals, side tools Your Salesforce renewal date and current contract terms Two things come out of this audit. First, you get a realistic read on how complex your migration actually is. Second, and more importantly, you get clarity on what your business actually needs today versus what it needed when CPQ was first configured. Most teams discover their CPQ environment was built for a version of the business that no longer exists. That’s useful information before you spend 18 months rebuilding it. 2. Get the Right People Aligned Before You Schedule a Single Demo The technical work is rarely what derails a CPQ migration. It’s the organizational work. Sales leadership, the CFO, and RevOps need to be aligned on three things before any vendor is brought into the conversation: Why now. Not “because EOS happened.” That’s not compelling to a CFO. The real reason is that legacy CPQ is already limiting revenue velocity, forecasting accuracy, and AI readiness in ways that have a measurable cost What winning looks like. “Quotes still generate” is not a success metric. Faster cycle times, cleaner revenue data, fewer approval bottlenecks, and a quote-to-cash process that scales with the business are How this connects to the broader strategy. Boards in 2026 want to see AI-ready infrastructure, tighter alignment between sales and finance, and unified revenue forecasting. A CPQ migration done right addresses all three. Done poorly, it’s just a disruptive IT project Executive sponsors who understand the strategic dimension will make decisions faster and defend the investment better when the project hits its inevitable rough patches. 3. Ask These Questions Before You Commit to Any Path Most RevOps teams jump to platform evaluation before they’ve answered the questions that should drive the decision. These are the ones that matter: Is Salesforce our long-term system of record, or are we operating across multiple CRMs? Are we moving toward usage-based, subscription, or consumption pricing in the next two years? How M&A-active is the business? Acquisitions stress CPQ environments in ways that become visible only after they close What’s the current state of our product catalog data? Is it clean enough to migrate, or does it need a full rebuild first? Do we have the internal bandwidth to run an 18-month implementation alongside business as usual? The answers don’t point to one universal path. They point to your path. 4. Treat Data

Three people looking at a sceen
Blog, AI, AI Agents

Why Most MedTech AI Pilots Never Reach Production

May 14, 2026

Most MedTech AI pilots do not fail because of the model. They fail because the organization was never operationally ready to scale them. That pattern is becoming increasingly clear across the enterprise AI market:• RAND Corporation estimates that more than 80% of AI projects fail to deliver business value• Gartner reports that only 48% of AI projects make it into production• MIT research found that 95% of generative AI pilots produce no measurable P&L impact, often due to workflow and integration issues rather than model performance The pilot works in a controlled environment:• curated data• dedicated resources• executive attention Production is different. That’s where AI collides with fragmented data, disconnected workflows, unclear ownership, regulatory constraints, and low frontline adoption. At N28 Technologies, we’ve seen the same pattern repeatedly: organizations invest heavily in AI capability while underinvesting in the operational foundation required to support it. That’s the real scaling problem. 1. AI-Ready Data Is Still Rare Most MedTech organizations have data. Few have operationally usable data. Clinical systems, CRM platforms, service data, device telemetry, and commercial operations often live in disconnected environments with inconsistent identifiers and fragmented governance. This is becoming one of the biggest enterprise AI bottlenecks. Gartner has repeatedly identified poor data quality and weak governance as leading causes of AI project failure. A pilot can work around this with curated datasets. Production cannot. At N28 Technologies, we consistently see AI success tied directly to the quality of the operational data layer underneath it. AI scales the quality of your data foundation — good or bad. 2. AI Pilots Ignore Workflow Reality Many AI pilots fail because they are designed outside the operational systems people use every day. Standalone dashboards and disconnected interfaces may work in a pilot environment, but they rarely drive long-term adoption. MIT research on enterprise GenAI deployments found that flawed integration into existing workflows was one of the primary reasons AI projects failed to generate measurable business impact. If AI recommendations are not embedded directly into the workflows teams already trust, usage drops quickly. For MedTech organizations, that means AI must connect directly into systems like Salesforce, Health Cloud, Service workflows, and commercial operations platforms. The organizations scaling AI successfully are designing around workflow execution, not model experimentation. 3. Governance and Ownership Arrive Too Late MedTech organizations often treat governance as a post-pilot exercise when it should shape the architecture from day one. This challenge is becoming more acute as AI regulation expands across healthcare, medical devices, and enterprise software. Gartner recently projected that more than 40% of agentic AI projects could be abandoned due to rising costs, governance gaps, and unclear business value. Regulatory review, auditability, model ownership, retraining processes, escalation paths, and compliance workflows all need to be part of the deployment strategy early. Pilots usually have sponsors. Production systems require operational owners. Without clear ownership, AI systems slowly lose trust, degrade operationally, and eventually disappear from frontline workflows. 4. Success Metrics Focus on Accuracy Instead of Operations A 92% accurate model that nobody operationalizes is still a failed deployment. Too many AI pilots optimize for technical performance while ignoring operational outcomes. The organizations successfully scaling AI are measuring:• reduced cycle times• increased throughput• faster service resolution• improved workflow completion• stronger user adoption Not just model accuracy. This shift matters because enterprise AI is increasingly moving from “insight generation” toward workflow execution. AI only creates enterprise value when it changes operational execution. The Real Question Most MedTech organizations are asking: “Where should we use AI?”The better question is:“What operational foundation does AI need in order to scale?”Because scaling AI is not primarily a model problem.It is a workflow, data, governance, and execution problem.That is where successful deployments are won or lost. At N28 Technologies, we believe the future of enterprise AI is not standalone models or disconnected copilots.It is AI embedded directly into operational workflows where teams already execute work.That requires more than technology. It requires readiness.