A Board and CEO Briefing on Agentic AI, the Evolution of the Sales Professional, and the Future of CRM
Executive Summary: Selling Has Become an Intelligence Problem
The sales productivity crisis is often framed as a talent problem, a quota problem, or a pipeline problem. All of these forces matter, but they obscure an equally important structural issue: most sales organizations are still running a twenty-first-century revenue mandate on tooling built to log the past rather than shape the future.
The numbers make the challenge visible.
Research from Salesforce’s State of Sales work, corroborated by Forrester, finds that sellers spend only about 30 percent of their time actually selling. The rest is consumed by administration and CRM data entry (roughly 20 percent), internal meetings (15 percent), prospect research (15 percent), email (10 percent), and other overhead. Microsoft’s own analysis of seller workflows puts the number even more starkly: sellers spend only about nine hours a week on genuinely high-impact work. Top performers sell roughly 34 percent of their time; underperformers sell just 23 percent — an 11-point gap that tracks almost exactly with quota attainment.
The arithmetic is unforgiving. Revenue leaders cannot simply hire their way out of this gap, and they cannot count on traditional CRM adoption to close it, because the CRM is where most of the lost time already goes. Something in the operating model has to change.
This is where agentic AI, and specifically what SalesChoice calls Zero-Query Sales Intelligence, moves from an interesting capability to a strategic necessity.
Zero-Query Sales Intelligence describes AI that anticipates a seller’s or a leader’s needs before they ask — synthesizing operational, behavioral, financial, customer, and environmental signals to deliver contextual guidance directly inside the flow of work. It is the opposite of a chatbot that waits to be queried.
Insight arrives ambiently, at the moment it is useful, rather than on demand.
Gartner’s research puts the scale of the shift in perspective. By 2028, the firm predicts, AI agents will outnumber human sellers by roughly ten to one, and agents will directly influence an estimated $15 trillion in B2B purchasing activity. Yet in the same forecast, Gartner cautions that fewer than 40 percent of sellers will report that AI agents actually improved their productivity. The risk is not too little automation — it is “agent sprawl”: more digital activity without more revenue, deployed on top of the same fragmented data and unclear accountability that made CRM painful in the first place.
Canada has a compelling vantage point on this shift in SalesChoice, the Toronto-based B2B sales intelligence company founded in 2011 by Dr. Cindy Gordon to replace black-box forecasting with auditable, explainable AI.
Its evolution toward Zero-Query Sales Intelligence — part of what the company describes internally as the Gordon AI Maturity Curve, progressing from process automation through predictive, generative, behavioral and emotional intelligence toward ambient and ultimately symbiotic human-machine intelligence — offers a concrete, governed template for what agentic selling can look like in practice.
The significance of this shift extends well beyond any single vendor or platform. Agentic AI does not need to replace sales professionals to revolutionize selling; it needs to absorb the 70 percent of the job that was never actually selling, surface the right signal at the right moment, and free sellers to do the parts of the job — judgment, empathy, negotiation, trust-building — that no agent can yet authentically do.
That could make agentic AI, and the intelligence layer that governs it, the most consequential change to the sales profession since the CRM itself was invented.
1. The Sales Profession Is Reaching Its Productivity Breaking Point
For three decades, sales technology has focused largely on capturing activity: contacts, opportunities, stages, forecasts. Those records remain necessary, but a perfectly logged pipeline still has to convert into revenue, and that conversion depends on judgment, timing, and relationship — the parts of selling a system of record was never built to help with.
That is where the current model begins to strain.
The modern sales day is highly fragmented. Seller, CRM, inbox, calendar, call-recording tool, forecasting spreadsheet, and enablement platform each hold a piece of the truth, and reconciling them consumes exactly the hours reps report losing to administration. Deals stall not for lack of interest but for lack of the right nudge at the right moment — a follow-up that never happened, a champion who went quiet, a risk signal buried in a transcript no one reread.
Other functions confronted a similar fragmentation problem earlier and solved it by layering intelligence on top of their systems of record: finance moved from ledgers to real-time analytics, supply chains moved from spreadsheets to predictive planning. Sales is now attempting the same move, later and under more pressure, because the cost of the status quo compounds directly into missed quota.
It helps to think of how a sales professional’s relationship with AI has actually evolved, rather than treating “AI in sales” as a single moment.
SalesChoice’s own maturity framework is a useful map of that evolution. The earliest stage is process automation — workflow rules, lead routing, basic reporting — which digitized the job without making anyone smarter. The next stage is predictive intelligence: lead scoring and forecasting models that tell a seller where to focus, which is where most enterprise sales AI still sits today. Generative intelligence follows — AI that drafts emails, summarizes calls, and produces content, the copilots that became mainstream over the past two years. Behavioral intelligence layers in buyer intent and engagement signals, reading what accounts are actually doing rather than what they say. Emotional intelligence extends that further, reading the seller’s own state — engagement, burnout risk, confidence — because a disengaged rep and a disengaged buyer both leak revenue. Ambient, or Zero-Query, intelligence is the current frontier: guidance that arrives without being requested. And the aspirational peak, which SalesChoice calls symbiotic intelligence, is a genuine partnership in which machines handle execution at scale, and humans supply judgment, trust, and accountability.
Most sales organizations, if they are honest about where they sit on that curve, are still oscillating between the second and third stages — predictive scores and generative drafts — while shopping for agentic tools built for the sixth. That gap is exactly where Gartner’s “agent sprawl” risk lives.
CINDY’S PERSPECTIVE
We need to stop treating “AI adoption” as a single milestone a sales organization either has or hasn’t reached. It is a maturity curve, and skipping stages is where most agentic AI investments quietly fail.
If a sales organization cannot yet trust its pipeline data enough to act on a predictive score, it has no business handing that same data to an autonomous agent and asking it to email a customer. Leadership’s task is to be honest about which stage the organization is actually in, fix the data and trust foundation first, and only then layer on autonomy — sequenced deliberately, not because a vendor demo looked impressive.
2. SalesChoice Is Showing What Zero-Query Sales Intelligence Looks Like in Practice
One of the clearest examples of Zero-Query Sales Intelligence in practice is SalesChoice itself, a B2B sales intelligence company built around what it calls “Auditable AI Forecasting on Salesforce.”
SalesChoice set out early to solve a specific credibility problem: sales forecasts generated by black-box models that executives could not explain, audit, or trust enough to act on. That founding constraint — explainability first — has shaped the company’s evolution ever since, including its move toward ambient, Zero-Query intelligence: systems that anticipate what a revenue leader or seller needs before they ask, synthesizing operational, behavioral, financial, customer, and environmental signals into real-time contextual guidance delivered inside the workflow, rather than requiring a query, a dashboard login, or a prompt.
Importantly, SalesChoice is not attempting to make its AI more autonomous by making it less transparent. The platform is built around five specific commitments: an intuitive interface accessible to sellers who are not data scientists, transparent accuracy scoring so users know how much to trust a given prediction, insights ranked for different learning styles, visibility into the data underlying each model’s output, and embedded coaching that turns an insight into a concrete next action rather than a number on a dashboard.
That hybrid posture — ambient and proactive, but auditable and explainable — may ultimately prove more durable than the more autonomous, less transparent agentic tools now flooding the market.
SalesChoice extends the same logic to organizational health through MoodInsights, a companion platform that gives employees a daily, anonymous channel to report how they feel and gives leaders a dashboard to act on that signal before it shows up in a forecast miss. The company reports that organizations using the tool have seen turnover fall by roughly 40 percent, productivity rise by about 18 percent, mistakes fall by roughly 62 percent, absenteeism drop by about 50 percent, and customer advocacy rise by around 12 percent. Those figures are specific to SalesChoice’s own deployments and should be read as a directional case rather than a universal benchmark.
Even so, the underlying logic matters for any board evaluating agentic AI: a forecast is only as trustworthy as the seller behind it, and an agent is only as safe to automate as the process it is automating is well understood, well governed, and well measured.
CINDY’S PERSPECTIVE
SalesChoice’s approach isn’t about the ambient intelligence layer. It’s the discipline to keep that layer explainable while everyone else in the category is racing toward autonomy first and transparency later.
Sales is a trust-sensitive function for good reason: forecasts move capital allocation, board expectations, and hiring plans, and a wrong number delivered with false confidence is worse than no number at all. Building Zero-Query intelligence that is also auditable requires more discipline than simply wiring an LLM into a CRM and calling it an agent.
It is also worth noting that this approach has been built, from the founding, by a female CEO in a category — enterprise AI infrastructure — that still skews heavily toward male-founded vendors. As the sales technology stack comes to combine forecasting, behavioral data, emotional intelligence, and autonomous agents, the leadership bench needs to broaden along with the toolset.
3. Agentic AI Is What Turns Prediction Into Autonomous Action
The agent attracts attention because it is visible: it drafts the email, books the meeting, updates the record. The more consequential shift is what sits behind it — the difference between a system that answers when asked and a system that acts on your behalf, continuously, toward a goal.
Generative AI and agentic AI are frequently conflated, but the distinction matters enormously to a board evaluating risk. A generative copilot drafts a message and waits for a human to send it. An agent researches the account, drafts the message, sequences the follow-ups, updates the record and escalates only when it hits a defined boundary — a multi-step, goal-directed, tool-using system operating with far less moment-to-moment human review.
Gartner expects that shift to move quickly: task-specific AI agents are projected to appear in 40 percent of enterprise applications by 2026, up from under 5 percent in 2025. Sales is one of the categories moving fastest, because prospecting and qualification are exactly the repetitive, high-volume, pattern-based work agentic systems handle well.
A new generation of AI “SDR” platforms — Artisan, 11x, Regie.ai and others — now run outbound prospecting largely autonomously: researching accounts, personalizing outreach at scale, sequencing follow-ups and qualifying leads before a human ever engages. One vendor in the category, Artisan, ran a widely covered — and widely criticized — billboard campaign in 2025 urging companies to “stop hiring humans” for sales development roles, a stance the company later complicated by hiring a human specifically to manage that campaign.
That anecdote is amusing, but it also illustrates a serious point: even the most aggressive automation vendors have found that judgment, escalation, and accountability still need a human in the loop somewhere. The interesting design question is not whether that human exists, but where in the process they sit.
Where agentic AI is embedded directly into a seller’s existing workflow rather than bolted on as a separate tool, the results are notable. Microsoft reports that sellers with high Copilot and agent usage inside its own sales organization closed 20 percent more deals, saw a 13 percent lift in lead-to-opportunity conversion, and generated 9.4 percent higher revenue per seller — evidence that the productivity gain comes from integration and workflow fit, not from the presence of an agent alone.
The emerging architecture is increasingly clear. Predictive and ambient intelligence tell an organization where to focus. Generative AI produces the content. Agents execute the repeatable steps. Humans supervise the exceptions, own the relationship, and make the calls an agent should not be trusted to make alone.
CINDY’S PERSPECTIVE
This is the point at which CROs need to think beyond deploying an agent. A prospecting bot is a tool; a prospecting bot connected to verified data, explainable scoring, escalation rules, and human oversight is a governed production system — a fundamentally different asset, and a fundamentally different risk profile.
Too many organizations buy an agent and call it transformation. Real transformation happens when territory design, compensation, hiring plans and escalation paths change around the agent — not when the agent is simply pointed at the existing process.
Sales leaders should therefore be asking a bigger question: which parts of this job should never be autonomous, and have we actually written that boundary down — or are we finding out where it is the first time an agent crosses it in front of a customer?
4. The CRM Market Is Moving From System of Record to System of Action
SalesChoice is important because it shows what governed, explainable agentic intelligence looks like in a live enterprise deployment. The AI SDR category is important because it shows autonomous execution moving from pilot to production. The broader CRM market is where the stakes of this shift become unmistakably visible.
ServiceNow’s CEO, Bill McDermott, has described autonomous AI agents as an “extinction-level event” for traditional CRM, predicting that agentic platforms will eliminate as much as 90 percent of conventional CRM systems by 2029 as software shifts from organizing customer data to acting on it directly. That is a deliberately provocative claim from a company positioned to benefit from it, but it captures a real anxiety now showing up in enterprise software valuations: Salesforce’s stock fell roughly 30 percent over the course of 2026 amid investor concern that Agentforce and its agentic peers were not yet proving out fast enough relative to newer, agent-native challengers, even as the company reported more than a billion dollars in Agentforce-related bookings.
Salesforce, Microsoft and HubSpot are each racing to answer the same question from inside their incumbent platforms. Salesforce’s Agentforce uses an “Atlas” reasoning engine to coordinate multiple agents across lead generation, account research and service. Microsoft frames its answer as “agentic CRM in the flow of work” — agents embedded in Outlook, Teams and Excel rather than requiring a separate interface, with the seller-productivity results cited earlier as its internal proof point. HubSpot’s Breeze layer adds autonomous agents on top of its existing assistant and data-enrichment tools. Smaller incumbents — Zoho, Freshworks, Creatio and others — are adding comparable agent layers at a faster and cheaper release cadence.
At the same time, a wave of agent-native challengers is attacking the category from outside the incumbents entirely. Zero, marketed explicitly as “the zero-click CRM,” aims to eliminate manual data entry by having agents research leads, monitor buying signals, and keep the record current automatically — a direct challenge to the premise that a CRM should require a human to feed it. Newer entrants such as Attio and Clay are pursuing a similar thesis: a system of record that acts, rather than one that merely stores.
It is worth holding two facts in tension here, because a board briefing that only repeats the more dramatic prediction is not a useful board briefing. First, the shift toward agentic, action-taking systems is real and already reflected in product roadmaps, capital allocation, and public-market sentiment. Second, forecasts of imminent, total disruption should be treated with real skepticism: analysts also project human-staffed support and service roles will keep growing through 2029, and survey data continues to show a majority of customers still prefer a human on first contact for anything consequential. Transformation here is more likely to be uneven and gradual than a single extinction event.
That is likely how this will actually unfold. No single vendor will “win” agentic CRM outright. Organizations will increasingly automate prospecting, qualification, forecasting, renewal risk, and customer research through different agents from different vendors, stitched together — which is precisely why data quality, integration discipline, and governance matter more than which logo sits on the license.
CINDY’S PERSPECTIVE
The question is no longer whether AI agents can take action inside a CRM.
They can.
The more important question is whether an organization’s data, governance and escalation rules are mature enough for that action to be trustworthy — at enterprise scale, across regions, compliance regimes and customer relationships that took years to build.
That is why Gartner’s finding is the one boards should sit with longest: sales leaders who overhaul their data and workflow infrastructure before scaling agents are roughly five times more likely to see real ROI than those who simply add agents on top of the systems they already have.
This is also why leaders should resist innovation theatre. An agent that closes a demo flawlessly is not transformation. A system that measurably reduces cycle time, lifts conversion and holds up under audit on a real enterprise pipeline is transformation.
The category needs far more measurement and far less “extinction-level” hyperbole.
5. Trust, Governance and the Human Edge Could Become the Real Differentiator
The governance opportunity may ultimately be as consequential as the productivity opportunity.
Gartner’s research contains a warning that deserves board-level attention: 60 percent of Chief Sales Officers say revenue outcomes are now driven largely by factors outside their direct control. Layering autonomous agents onto that already-diffuse accountability, without clear boundaries on what an agent may say, promise, or commit to on the company’s behalf, does not reduce that risk. It compounds it.
An agent that can research an account and draft an email is low-risk. An agent that can quote a price, confirm a delivery date, or commit to contract terms without human review is a different category of exposure entirely — legal, financial, and reputational — and one that few sales organizations have explicitly governed for, because the tooling arrived faster than the policy did.
This is where explainability stops being a nice-to-have and becomes the actual product of a responsible agentic AI strategy. An auditable model that shows its reasoning and its data lineage can be reviewed, corrected, and defended to a regulator, an auditor or a customer. An opaque agent that simply acts, however impressively, cannot — and “the model did it” is not an answer a board or a courtroom will accept.
The human dimension deserves equal weight. As agentic AI commoditizes execution — the drafting, the scheduling, the routine follow-up — the capabilities that remain scarce are precisely the ones machines cannot authentically replicate: empathy built on shared experience, trust earned through consistency over time, moral judgment about which deals to walk away from, and the imagination to question a prospect’s stated requirement rather than simply optimizing against it. Those are not soft skills to be trained away as automation scales. They are about to become the primary differentiator between sales organizations, because everything else will be commoditized by the same agents everyone else is buying.
If governance and human judgment are designed into the agentic layer from the outset, rather than bolted on after an incident, the old trade-off between speed and trust could begin to weaken — organizations could move faster and remain more accountable, not less.
CINDY’S PERSPECTIVE
For too long, AI governance in sales has been treated as a compliance checklist added after a platform is already live.
The greater opportunity is to design accountability into the agent architecture itself — explainable scoring, clear escalation boundaries, human review at the moments that actually carry risk — so that trust becomes a feature of the system rather than a promise made about it after the fact.
That is a far more durable proposition for boards, regulators and customers alike, and it is a genuine competitive advantage for whichever organizations build it first rather than being forced into it after a public failure.
Executive Conclusion: Selling Is Entering Its Agentic Moment
Agentic AI will not solve the sales productivity crisis on its own. Compensation design, territory planning, hiring, and pipeline discipline all remain critical — but none of those factors remove the underlying arithmetic: sellers are spending 70 percent of their time on work that isn’t selling, and the tooling built to fix that has, for two decades, mostly added to the burden.
Sales organizations need substantially more seller capacity while the profession itself is under real pressure — from burnout, from turnover, from AI-driven anxiety about the SDR role in particular. That creates a simple operating reality: revenue leaders must learn to generate more pipeline and more forecast accuracy with the people, data, and trust they already have.
SalesChoice offers one of the clearest examples of what a governed version of that transition can look like, evolving from auditable forecasting toward ambient, Zero-Query Sales Intelligence without sacrificing explainability. Microsoft, Salesforce and HubSpot are showing how incumbent platforms are racing to embed agentic capability directly into the seller’s existing workflow. Artisan, 11x and Regie.ai are demonstrating that autonomous prospecting has moved from pilot to production, controversy and all. ServiceNow’s extinction-level provocation, however overstated, has forced the entire category to answer a harder question about what a CRM is actually for.
Collectively, these signals suggest that selling is beginning to cross an important threshold.
Pipelines increasingly begin as signals. AI translates that signal into a prioritized next action. Agents execute more of the repeatable outbound and administrative work. Ambient intelligence surfaces risk and opportunity before a human asks. Skilled sellers supervise, build trust, exercise judgment, and do the relationship work that still requires a person.
In other words, selling is beginning to industrialize its intelligence layer, the way construction is beginning to industrialize its physical one.
The winners will not necessarily be the companies that deploy the most agents. They will be the organizations that pair agentic execution with governed, explainable intelligence and disciplined data foundations faster than their competitors.
For boards and CROs, the strategic question is therefore not whether agentic AI is coming to sales. On Gartner’s own trajectory, it will outnumber human sellers ten to one within two years. The real question is whether today’s sales leaders will redesign their operating model, their governance and their talent strategy around that shift — or discover its boundaries for the first time in front of a customer, a regulator or a board.
Cindy’s Closing Perspective
I have spent much of my career building explainable AI in a category that often prizes speed over scrutiny. The pattern with agentic AI in sales is now familiar to me: first the technology looks unrealistic, then it becomes interesting, then it feels threatening — a billboard telling companies to stop hiring people will do that — and eventually it becomes part of how the profession works.
Agentic AI in sales is moving through those stages faster than almost any technology I have watched enter this industry.
We should not exaggerate what it can do. Agents cannot yet exercise judgment about which relationship to protect at the cost of a quarter’s number, cannot authentically build the trust that closes a complex enterprise deal, and cannot substitute for a sales culture that is actually well led. No model compensates for weak leadership or unclear accountability.
But complacency here would be the far greater mistake.
Sellers are burning out under work that was never selling. Buyers increasingly expect anticipatory, not reactive, engagement. Boards are asking harder questions about AI governance with every quarter that passes. These realities create a powerful reason to rethink how revenue intelligence is built — not just bought.
The vendors that deserve the closest attention are therefore not simply those with the most autonomous agents. They are the organizations willing to do the harder work of making that autonomy explainable, auditable, and genuinely trustworthy.
That has been SalesChoice’s work from the beginning.
Perhaps the defining question for sales leaders is no longer whether they can afford to experiment with agentic intelligence. Perhaps it is whether they can afford to keep running a revenue engine that only tells them what already happened.
Lady Whistledown’s Sales Society Papers
Dearest Gentle Reader,
It has come to this author’s attention that the newest member of the sales development team neither asks for a base salary nor requests Fridays off.
It is, in fact, not a person at all.
And, one must admit, it has arrived at a rather indelicate moment for the profession. Sellers already spend a scandalous seven-tenths of their working hours on matters other than selling, quotas remain stubbornly unforgiving, and one imagines rather a lot of very expensive software has, until recently, mostly added to the paperwork rather than subtracted from it.
Enter the agents.
One ambitious young company took out billboards across San Francisco last year instructing the public, in no uncertain terms, to “stop hiring humans” for sales development. Society was not amused. There were, this author is reliably informed, actual threats issued over a billboard, which seems rather a lot of feeling to spend on a marketing campaign. The delicious twist, dear reader, is that the very same company shortly thereafter hired an actual human being — to manage the fallout from the campaign that told everyone else not to hire humans.
How scandalous. How entirely predictable.
Meanwhile, in altogether more sober company boardrooms, one chief executive has taken to describing the traditional customer relationship management system as facing an “extinction-level event,” while his rival’s share price wobbled a genuinely un-genteel thirty percent over the year, and a scrappy young platform calling itself simply “Zero” promises to eliminate the very data entry that made everyone despise the old systems in the first place.
There will naturally be those who fear the agents intend to replace the sellers entirely. This author suspects the more interesting possibility is that they may finally free the good ones from the eleven-point gap between winning and merely surviving quota — the very gap, Gartner tells us, that separates a top performer from an also-ran.
The greater disruption may therefore not be the arrival of the agent itself.
It may be the realization that revenue can be organized very differently — provided somebody, somewhere, remains willing to be held accountable when the agent is wrong.
One should watch the explainable players particularly closely. Society is endlessly enthusiastic about autonomy once somebody else has absorbed the risk of it. It is rather less generous to the leaders who must first persuade regulators, boards and customers that a system nobody can fully explain deserves their trust.
Leadership, dear reader, always appears most obvious in retrospect.
For now, the agents are reading the pipeline, the forecasts are beginning to explain themselves, and some of the industry’s largest players are quietly redesigning how revenue gets made.
The quota, as they say, is beginning to reset.
Until our next gathering, dear reader,
Lady Whistledown
Research Bibliography
Gartner. “Gartner Predicts AI Agents Will Outnumber Sellers 10 to 1 by 2028, Yet Fewer Than 40% of Sellers Will Say Agents Improved Productivity.” Press release, 2026.
Gartner. “Gartner Predicts By 2028 AI Agents Will Outnumber Sellers by 10X — Yet Fewer Than 40% of Sellers Will Report AI Agents Improved Productivity.” Press release, 2025.
Digital Commerce 360. “Gartner: AI Agents Will Command $15 Trillion in B2B Purchases by 2028.” 2025.
Gartner. “Gartner Predicts 40% of Enterprise Apps Will Feature Task-Specific AI Agents by 2026, Up from Less Than 5% in 2025.” Press release, 2025.
Landbase. “Why Sales Reps Spend Less Than 30% of Their Time Selling (And What to Do About It).” Citing Salesforce State of Sales and Forrester research, 2026.
Microsoft Dynamics 365 Blog. “Agentic CRM in the Flow of Work: How AI Is Transforming Sales and Rebuilding Customer Trust.” 2026.
CRM.org. “CRM Is Dead? Long Live Agentic AI.” Newsletter, 2026.
AIMultiple. “Top 8 Agentic CRM Platforms in 2026.” 2026.
Zero. “Zero: The Zero-Click CRM.” Corporate website.
Parameter. “Salesforce (CRM) Stock Tumbles 30% in 2026 Amid AI Disruption Concerns.” 2026.
Artisan AI. “Why We Put ‘Stop Hiring Humans’ on Billboards.” Corporate blog.
Yahoo Finance. “Artisan Told Companies to ‘Stop Hiring Humans’ for Sales. Now It’s Hiring One for Just That Job.” 2026.
The Enterprise World. “Dr. Cindy Gordon, SalesChoice.” Profile interview.
SalesChoice. “The Last Competitive Advantage: Why Human Intelligence Will Become More Valuable in the Age of Artificial Intelligence.”
SalesChoice. “MoodInsights.” Product overview.
