As buyer behavior shifts with the rise of artificial intelligence, Deloitte Digital and Salesforce can help sales organizations build an AI-ready architecture that connects experience, orchestration and data.
Composable, modular, open
The priorities of chief revenue officers (CROs) haven’t changed. Leaders are still responsible for quickly onboarding new sellers, aligning incentives, designing and delivering effective coaching, and driving sales methodology adoption. Even with new tools, these challenges can feel just as difficult as they did five years ago.
What has changed is the buyer. Buyers increasingly use artificial intelligence (AI) to research vendors, compare solutions and narrow their options long before they contact a seller to discuss an opportunity. By the time the buyer engages, the consideration set is often already defined. In fact, 95% of the time, the winning vendor is already on the shortlist from day one, and buyers wait until two-thirds of the way through their evaluation before contacting a seller.¹
That doesn’t necessarily mean a competitor made a more compelling case. Often, buyers simply know which vendor they prefer before the process even starts.
That shift in buyer behavior requires sales organizations to rethink how they operate. Sellers need to reach buyers earlier and approach the conversation with something the buyer cannot get from a search.
This is where AI can change how a sales organization operates. However, in our experience, the organizations that have seen the biggest returns are the ones that redesigned their processes and tech stack around AI, not the ones that bolted AI onto existing processes.
Deloitte Digital’s new POV, Accelerating sales with agentic AI, presents seven levers that leading organizations use to adapt their go-to-market models for AI. This article dives deeper into the technology lever.
Salesforce at the center of intelligent selling
Many organizations have invested heavily in Salesforce. Sellers use it every day, and core commercial processes are built around it. Now, the surrounding environment has changed. Frontier models, agent frameworks, and open protocols such as Model Context Protocol (MCP) arrived faster than most architectures were built to absorb.
And then in August 2026, Salesforce and Anthropic announced Claudeforce, a partnership that makes Claude the default reasoning model across Agentforce and Slack, and ships Salesforce in Claude, a plugin that lets sellers work the pipeline without opening Salesforce.² In this environment, “inside” and “outside” the platform are no longer meaningful distinctions.
With these shifts, Salesforce has become more than the system where sellers work; it can serve as the anchor of a broader AI-ready architecture.
Designing an AI-ready architecture
For most CROs and CIOs, the question isn't whether to invest in AI for sales. It's how to build a technology architecture that makes AI useful across the entire commercial model.
One approach gaining traction is a “headless” architecture, which lets organizations use their data and AI capabilities wherever the work happens, not only within a platform’s built-in user interface (UI). Headless 360 is Salesforce's implementation of that architecture.
A headless architecture consists of three key layers:
Each layer may contain several platforms and solutions, and Salesforce is central to all three.
DESIGNING FOR CHANGE
A composable, modular and open sales stack
Three principles keep a modernized sales stack from becoming next year's technical debt.
Composable means assembling the stack from best-fit components and connecting them through open interfaces, including the interface layer itself. Salesforce provides the CRM and agent workflow foundation, while other platforms support data, infrastructure or specialized AI needs.
Modular means that organizations can replace any component without rebuilding the entire environment. The market is moving too quickly for technology decisions that cannot be revisited. If changing one agent framework requires re-integrating six downstream systems, the architecture is still rigid and may limit evolution. With modular design, organizations can adopt new capabilities as they emerge without platform migration, keeping costs manageable.
Open is the principle that has evolved as hybrid architectures have taken flight. The Salesforce platform was designed as a complete platform, but completeness and composability are different design goals. Headless 360 prioritized composability; every capability can be accessed through an API, an MCP tool or a CLI command. The trade-off is that more of the data foundation comes from one vendor. In exchange, what you build on top remains yours to change.
From efficiency gains to structural advantage
AI agents that handle meeting preparation, research and pipeline hygiene can give sellers hours back every week. Organizations that connect planning, execution and reporting can shorten deal cycles and gain a clearer view of cost-to-serve.
Those improvements matter, but the bigger opportunity is to do things that weren't possible before, like continuously optimizing territories using live pipeline signals, identifying renewal risk based on customer conversations months before the renewal date, or surfacing connections across accounts and products that might otherwise remain hidden.
One organization we work with found that 70% of seller time was consumed by administrative work, including finding information, managing knowledge and navigating internal processes. After consolidating on a single experience surface over Salesforce, the organization did more than recover that time. It gave sellers access to context-aware intelligence that no amount of manual research could have assembled.
Building a technology architecture for what comes next
The commercial stack is becoming more open, composable and modular, and that makes the platforms at the center more valuable. AI alone will not determine sales organizations’ results; success will depend on the architecture, redesigned processes, transformed organization and context you provide to AI.
For most revenue teams, Salesforce is well-positioned to serve as that center of gravity, and the ones building durable advantage will connect it to everything else to scale AI. This flexibility will help organizations respond as buyer behavior, AI capabilities and commercial priorities evolve, without having to rebuild from the ground up.
This article is part of Deloitte Digital's Accelerating sales with AI series. The companion piece lays out the full seven-lever framework for building an AI-ready commercial operating model.
About the authors
Abe Awasthi leads Sales Transformation at Deloitte Digital, helping clients drive profitable growth and operational efficiency through commercial strategy, process innovation and AI-enabled technology. His expertise spans AI-first operating models, go-to-market transformation and agentic solutions.
Harry Datwani leads Deloitte Digital's Global Salesforce business where he works with C-suite executives to optimize their front office digital and AI investments and create differentiated experiences for sellers, partners and customers.
Design your human + digital sales strategy
Empower trusted data and AI with Deloitte Digital and Salesforce
Integrate AI Agents into your team with Agent Advantage™ for Salesforce
Sources
1. 6sense, 2025 B2B Buyer Experience Report, accessed September 1, 2026.
2. Salesforce, “Salesforce and Anthropic Announce Claudeforce: The #1 AI Meets the #1 AI CRM,” press release, August 27, 2026.
