The Autonomous Seller

Ross Sylvester Founder, CEO -- Adrata

---

Humans are not wired to sell in today's enterprise market. Here's the data that proves it -- and what comes next.

---

The Setup

For most of history, selling was about relationships, judgment, and timing. A rep who could read a room, remember a buyer's kids' names, and show up with the right proposal at the right moment had an edge. That edge worked when buying was simple, when decisions were made by one or two people, and when the seller controlled the flow of information.

That world is gone.

Today, the average complex B2B purchase involves 11 individual stakeholders -- sometimes flexing up to nearly 20 for enterprise-level deals.1 In 2015, Gartner pegged the average at 5.4. Buying committees have doubled in a decade. Each stakeholder arrives with their own research, their own priorities, their own internal politics. And they're not waiting for your sales rep to educate them. Buyers are 70% through their decision journey before they ever speak to a salesperson.2 By the time the phone rings, the decision is already taking shape -- and your rep wasn't in the room when it happened.

No human being can track 11 stakeholders across 50 active opportunities simultaneously, remember every conversation, detect every shift in sentiment, and execute with perfect consistency across all of them. This is not a question of talent or effort. It is a question of cognitive architecture. The human brain was not built for this scale of parallel processing.

This is the same argument Chamath Palihapitiya made about investing. For decades, the best investors were the ones with the best judgment -- the Warren Buffetts, the Peter Lynches, the George Soroses. They could read markets, synthesize information, and act on instinct honed by decades of pattern recognition. But as markets became faster, more complex, and more data-rich, the edge shifted. The machines didn't need to be smarter than the best humans. They just needed to be more consistent than the average ones.

Sales is now undergoing the same phase transition.

---

The Proof: What the Data Says

The numbers are not ambiguous. They are catastrophic.

Only 25-28% of B2B sales reps hit quota in 2024 -- the lowest figure in six years.3 This is not a blip. This is a structural collapse. Historically, quota attainment hovered around 60-70%. In 2024, average attainment dropped to 43%, down from 53% the year before.4

The cause is not that reps got worse. The cause is that quotas got harder while the selling environment got more complex. Sales quotas rose 37% from 2023 to 2024 while attainment plummeted.4 QuotaPath's 2024 Compensation Trends Report found that 91% of organizations missed quota expectations.4 Nine out of ten. That is not underperformance. That is a system failure.

Meanwhile, sales reps spend only 25-28% of their time actually selling.5 The rest goes to CRM updates, internal meetings, prospecting research, email composition, and administrative overhead. Bain's 2025 Technology Report found that AI could effectively double active selling time by eliminating routine tasks -- suggesting that the current model wastes roughly half of every rep's productive capacity on work that adds zero direct revenue value.5

The SDR model is even more broken. Average SDR tenure: 16-18 months.6 Average ramp time: 3.1-3.2 months.6 Peak performance occurs in years two and three. Most SDRs leave before they peak. Companies are spending six figures to recruit, train, and ramp a rep who will leave before reaching full productivity. The annual fully-loaded cost of a human SDR runs $110,000-$150,000. An AI sales agent costs $10,000-$12,000 per year -- a 70-80% reduction -- while operating 24/7 with zero ramp time.7

Put differently: the cost per effective selling hour for a human SDR, once you account for the 72% of time spent on non-selling work, is roughly $240-320. For an AI agent, that number approaches $0.70-4.00.7

These are not projections. These are current economics.

---

A Brief History of Sales Methodology

To understand where autonomous selling is going, you need to understand the intellectual history that brought us here. The evolution mirrors what happened in quantitative finance -- a slow march from intuition to data to algorithms to full autonomy.

1936 -- Dale Carnegie and the Relationship Era

Dale Carnegie published How to Win Friends and Influence People in 1936. It has sold more than 30 million copies and remains one of the bestselling nonfiction books in history.8 Its thesis was simple: the seller's edge was charisma. Smile. Remember names. Make the other person feel important. Listen more than you talk.

For decades, this was the dominant sales philosophy. It worked because buying was personal. Decisions were made by individuals, not committees. Information asymmetry favored the seller. If you could build rapport, you could close deals. The Rolodex was the technology. The golf course was the platform.

1988 -- Neil Rackham and the First Data-Driven Methodology

Neil Rackham's SPIN Selling represented a fundamental break. Rackham and his team at Huthwaite analyzed 35,000 sales calls across 23 countries over 12 years, spending over $1 million on research.9 It was the largest observational study of selling behavior ever conducted.

His central finding destroyed the prevailing orthodoxy: traditional closing techniques -- the assumptive close, the alternative close, the urgency close -- were negatively correlated with success in large, complex sales. What worked instead was a structured questioning methodology: Situation, Problem, Implication, Need-Payoff. Consultative questioning outperformed traditional closing.

SPIN Selling was the equivalent of Harry Markowitz's Modern Portfolio Theory in finance. It didn't just propose a new approach. It proved, with data, that the old approach was wrong. Sales stopped being an art and started becoming a science.

1996 -- MEDDIC and the Sales Equivalent of Quant Investing

In 1996, Dick Dunkel, working under SVP John McMahon at PTC, created MEDDIC.10 PTC was one of the most successful enterprise software companies of its era, but it had a problem: attrition. The sales organization was roughly 300 people strong, and they couldn't hire fast enough to replace the reps they were losing.

Dunkel and his colleagues analyzed their wins and losses and discovered that, without fail, every deal's outcome could be attributed to six specific elements: Metrics, Economic Buyer, Decision Criteria, Decision Process, Identify Pain, and Champion. They codified these into a qualification framework.

The result: PTC grew from $300 million to $1 billion in revenue in four years.10

MEDDIC was the sales equivalent of quantitative investing. It replaced gut feel with a systematic, repeatable framework. It didn't eliminate the need for skilled sellers -- it gave them a map. And like early quant strategies, it created an enormous edge for those who adopted it while the rest of the market still relied on instinct.

1999 -- Salesforce and the Digitized Rolodex

Marc Benioff and Parker Harris founded Salesforce in 1999 with the tagline "The End of Software."11 It was the first cloud CRM. It digitized the Rolodex, gave managers pipeline visibility, and created a system of record for customer relationships.

Salesforce is now a $37.9 billion revenue company with a 20.7% CRM market share -- more than its four closest competitors combined.11 It transformed how sales organizations operate.

But here is what Salesforce did not do: it did not add intelligence. A CRM is a database. It records what happened. It does not tell you what will happen, what should happen, or why a deal is stalling. For all its power, Salesforce was a digital filing cabinet. The human still had to do all the thinking.

2011 -- The Challenger Sale

Matt Dixon and Brent Adamson studied 6,000 sales reps across 90 companies and published The Challenger Sale in 2011.12 Their central finding: 40% of top performers in complex sales were "Challengers" -- reps who teach, tailor, and take control of the conversation. Relationship builders, the Dale Carnegie archetype, were the worst performers in complex B2B environments.

The performance gap was staggering. In complex sales, star performers outperformed the average by 200%.12 And the differentiator was not empathy or likability. It was the ability to push back, reframe the buyer's understanding of their own problem, and deliver commercial insight the buyer couldn't have developed on their own.

The Challenger Sale proved that in complex enterprise deals, the seller who teaches wins. Relationships alone are not enough when 11 stakeholders need to reach consensus.

---

The Five Levels of Autonomous Selling

Chamath described five levels of autonomous investing, from manual stock-picking to fully algorithmic portfolio management. Sales is following the same trajectory. Here are the five levels -- and we are currently transitioning from Level 4 to Level 5.

Level 1: Manual Selling (Pre-1990s)

Rolodex. Handshakes. Lunches. Golf. The Dale Carnegie era.

The seller IS the system. There is no technology, no process, no data. Success equals personal charisma plus persistence. The best sellers are the ones who can work the longest hours, remember the most names, and make the most phone calls.

The edge was largest when markets moved slowly enough for humans to keep up. One buyer, one decision-maker, one relationship. When the seller controlled the information, the seller controlled the deal.

Level 2: CRM-Assisted Selling (1999-2010)

Salesforce launches in 1999. Then HubSpot, Microsoft Dynamics, and dozens of others. For the first time, sales organizations have a shared digital record of every contact, every deal, every interaction.

Pipeline visibility improves dramatically. Managers can see the forecast. Reps can log their activities. But there is zero intelligence. The CRM records data. It does not interpret it, predict outcomes, or recommend actions. No human judgment is replaced -- just better record-keeping.

This is the equivalent of digitizing a paper trading ledger. Useful, but not transformative. The human still makes every decision.

Level 3: Automated Selling (2010-2018)

Salesloft launches in 2011. Outreach in 2014. The era of email sequences, cadences, and multi-step automated workflows begins.

For the first time, a rep can set up a 12-touch outbound sequence and have it execute automatically. Emails go out on schedule. Follow-ups trigger based on opens and clicks. The SDR model -- popularized by Aaron Ross in Predictable Revenue -- becomes the standard factory model for pipeline generation.13

But these systems have zero contextual awareness. They execute predetermined rules. They don't know if the prospect just got promoted, if the company just raised a round, or if the economic buyer changed. They are the "algorithmic trading" of sales -- rules-based, with no learning and no adaptation.

The automation increased volume but did nothing for quality. Response rates began their long decline. Buyers started drowning in templated outreach. The arms race was on.

Level 4: AI-Augmented Selling (2018-2024)

This is where the data starts getting interesting.

Gong pioneers conversation intelligence. The company records, transcribes, and analyzes sales calls at scale, identifying patterns in what top performers say and do differently. Gong has raised $584 million in total funding, reached a $7.25 billion peak valuation, and generated $332 million in revenue in 2024.14

Clari builds an AI-powered revenue platform that delivers 98% forecast accuracy and, according to a Forrester Total Economic Impact study, generated $96.2 million in value for enterprise customers with a 398% ROI -- achieving payback in under six months.15

In December 2025, Gong released its second annual State of Revenue AI report, analyzing 7.1 million sales opportunities across 3,600+ companies and surveying more than 3,000 global revenue leaders. The headline finding: teams embedding AI as a core driver of their go-to-market strategy generate 77% more revenue per rep -- a six-figure increase over teams that don't.16

Bain's 2025 research confirms the pattern: early AI deployments in sales show 30%+ improvement in win rates.5

At Level 4, AI listens, scores, recommends, and coaches. But humans still drive. The AI is the co-pilot, not the pilot. It tells the rep what to do. The rep still has to do it.

Level 5: Agentic AI Selling (2024-Present)

This is where the phase transition begins.

11x builds AI SDR "Alice" and AI phone agent "Mike." The company raised a $50 million Series B led by Andreessen Horowitz in November 2024, valued at approximately $350 million. A single AI agent replaces the output of multiple human SDRs, operating across languages, time zones, and channels simultaneously.17

Artisan AI builds "Ava," an AI BDR that automates prospecting, messaging, and meeting booking. The company raised a $25 million Series A in April 2025, backed by Y Combinator, HubSpot Ventures, and Glade Brook Capital. By the time of the raise, Artisan had 250 paying customers and approximately $5 million ARR.18

Clay builds the data infrastructure layer for AI-powered go-to-market. The company raised a $100 million Series C at a $3.1 billion valuation -- six times its valuation from just over a year earlier. Clay reached $100 million in ARR in 2025, tripling revenue year-over-year. Its customer list includes OpenAI, Anthropic, and thousands of revenue teams building AI-native workflows.19

The Gartner predictions tell the rest of the story:

• By 2028, AI agents will outnumber human sellers by 10x.20
• AI agents will intermediate more than $15 trillion in B2B spending by 2028.20
• By 2028, 60% of B2B seller work will be executed through generative AI technologies, up from less than 5% in 2023.21

At Level 5, the AI doesn't just recommend -- it acts. It researches prospects, crafts personalized outreach, qualifies inbound leads, books meetings, follows up, handles objections, and in some cases, runs entire early-stage sales conversations. The human enters the loop for high-stakes moments: the executive alignment call, the negotiation, the close. Everything else is handled by agents.

---

The Medallion Fund of Sales

Jim Simons founded Renaissance Technologies in 1982. His flagship Medallion Fund, launched in 1988, has delivered average annual returns of 66% before fees and 39% net -- the most successful track record in the history of investing.22 Over its entire history, the fund has had effectively one losing year (1989). A $100 investment in 1988 would have grown to approximately $398.7 million by 2018.22

As Gregory Zuckerman documented in The Man Who Solved the Market, Medallion's edge was not one brilliant algorithm. It was the systematic replacement of human judgment with mathematical models across every part of the investment process -- from signal detection to trade execution to risk management. The key insight: the edge came from how many parts of the process could be handed to machines and how those interconnected pieces worked together.

"The Medallion Fund didn't just automate trading. It automated thinking."

Sales has its Medallion moment.

MEDDIC at PTC was the first proof point -- a systematic, data-driven sales machine that took a company from $300 million to $1 billion in four years.10 It was the sales equivalent of early quantitative investing: a framework that replaced intuition with process and delivered outsized returns.

The modern equivalent is what Gong's data reveals: organizations embedding AI at every stage of the sales workflow -- from prospecting to qualification to coaching to forecasting -- are generating 77% more revenue per rep than those who don't.16 This is the beginning of a Medallion-like divergence. The gap will widen.

The parallel is precise. Medallion's edge was never one model. It was the integration of hundreds of models, each handling a different piece of the investment process, all connected and continuously learning from each other. In sales, the equivalent is an AI stack that handles research (Clay), outreach (11x, Artisan), conversation analysis (Gong), forecasting (Clari), and coaching -- all feeding data back into each other.

The firms that build this integrated stack will compound their advantage just as Renaissance did. Everyone else will be trading on instinct in a market that no longer rewards it.

---

What This Means for Revenue Leaders

The forward-looking data is clear:

• By 2028, 60% of B2B seller work will be executed through generative AI conversational interfaces.21
• By 2027, 95% of seller research workflows will begin with AI, up from less than 20% in 2024.23
• Sellers using AI are 3.7x more likely to meet quota than those who don't.24
• 81% of sales teams are already experimenting with or have fully deployed AI tools, up from less than a third in 2023.25

The Gong data shows the compounding effect. Revenue organizations using AI in 2024 reported 29% higher sales growth than their peers. By 2025, that advantage had grown to 77% more revenue per rep. The gap is accelerating.16

But here is the counterpoint that every honest analysis must include:

By 2030, 75% of B2B buyers will prefer sales experiences that prioritize human interaction over AI.26

Gartner's Colleen Giblin put it directly: "After several years of increasing interest in self-serve and AI-driven sales, we're now beginning to see a reversal, with more buyers expressing a desire for authentic human engagement, especially in complex or high-stakes transactions."26

This is not a contradiction. It is a design specification.

The future is not full replacement. The future is a radical reallocation of human attention. AI handles the 75% of the sales workflow that is research, administration, outreach, qualification, and follow-up. Humans handle the 25% that requires judgment, empathy, creativity, and trust -- the executive conversation, the negotiation, the moment when a $2 million deal hinges on whether the buyer trusts the person across the table.

The organizations that understand this will build what I call the centaur model -- human judgment augmented by machine execution at every other stage. The organizations that try to automate everything will discover that buyers still want to look someone in the eye before signing a seven-figure contract. The organizations that resist automation will be buried by competitors whose reps are 3.7 times more likely to hit quota.

---

What the Smartest Investors See

The venture capital market has already placed its bets.

In July 2024, Andreessen Horowitz published "Death of a Salesforce," authored by Zeya Yang, Marc Andrusko, and Angela Strange.27 The thesis was unambiguous:

"AI will so fundamentally reimagine the core system of record and sales workflows that no incumbent is safe."

a16z argued that the next generation of sales platforms will be multi-modal -- text, image, voice, video -- containing every customer insight from across the company. AI-native platforms will extract more insight from a customer and their mindset than any current tool can. These systems will automate early-stage tasks like prospecting and qualification while providing live coaching and driving decisions across sales, marketing, and product teams.27

Sequoia has backed both Gong ($7.25 billion valuation) and Clay ($3.1 billion valuation), with partner Sonya Huang championing the thesis that the AI application layer will capture more value than the infrastructure layer.28 Their portfolio reflects a conviction that every industry requiring humans to create original work -- from marketing to sales -- is up for reinvention.

The pricing model shift is equally telling. Traditional sales compensation runs 10-15% of deal value in AE commissions. AI agent platforms are experimenting with outcome-based pricing: 3-5% take-rates on closed deals.29 If an AI agent can deliver comparable conversion rates at one-third the cost, the economic incentive to shift is overwhelming.

a16z's December 2024 enterprise newsletter made this explicit: AI is driving a shift from seat-based to outcome-based pricing across the entire enterprise software stack.29 The companies that charge per seat are the incumbents. The companies that charge per outcome are the insurgents.

---

The Uncomfortable Truth

Here is what most sales leaders do not want to hear.

The quota crisis is not cyclical. It is structural. When 91% of organizations miss quota and only 25% of reps hit their number, the problem is not the reps. The problem is that the selling model -- built for a world of single-threaded, relationship-driven deals -- has been deployed into a world of 11-stakeholder, consensus-driven, information-rich buying processes that no human can navigate at scale without machine assistance.

The reps who succeed are not the ones working harder. They are the ones working with AI. Gartner's data is definitive: sellers who partner with AI are 3.7x more likely to meet quota.24 That is not a marginal improvement. That is a categorical difference.

And the gap will widen. Gong's year-over-year data shows the advantage of AI-augmented selling growing from 29% to 77% in a single year.16 This has the same shape as Medallion's compounding advantage over discretionary traders. Once the machine learning flywheel starts turning -- more data, better models, more wins, more data -- the advantage compounds.

The question for every revenue leader is not whether to adopt AI. The question is whether you are building for Level 4 or Level 5. The companies that treat AI as a feature -- a chatbot bolted onto an existing workflow -- will get Level 4 results. The companies that redesign the entire revenue process around AI agents, with humans deployed at the highest-leverage moments, will get something closer to Medallion-level performance divergence.

The autonomous seller is not coming. The autonomous seller is here. The only question is how long it takes you to recognize it.

---

Further Reading

• "How to Win Friends and Influence People" by Dale Carnegie (1936) -- The foundational text of relationship selling. Still relevant for the human moments that AI cannot replace.
• "SPIN Selling" by Neil Rackham (1988) -- The first data-driven sales methodology, based on 35,000 observed sales calls. Proved that questioning strategy beats closing technique.
• "Crossing the Chasm" by Geoffrey Moore (1991) -- The technology adoption lifecycle. Essential for understanding where AI in sales sits today (early majority).
• "The Challenger Sale" by Matt Dixon and Brent Adamson (2011) -- Why teaching, tailoring, and taking control outperforms relationship-building in complex sales. Based on 6,000 rep study.
• "Predictable Revenue" by Aaron Ross and Marylou Tyler (2011) -- The SDR model that Salesforce used to add $100 million in recurring revenue. The playbook that defined Level 3.
• "The Hard Thing About Hard Things" by Ben Horowitz (2014) -- Building when there is no playbook. Relevant for every leader navigating the AI transition.
• "The Man Who Solved the Market" by Gregory Zuckerman (2019) -- Jim Simons and Renaissance Technologies. The definitive account of how machines replaced human judgment in investing.
• "Death of a Salesforce" by Zeya Yang, Marc Andrusko, and Angela Strange, Andreessen Horowitz (2024) -- The VC thesis on why AI will fundamentally reimagine every sales workflow.

---

Sources

Footnotes

1. Gartner, "The New B2B Buying Journey" (2023-2024). Average complex B2B purchase involves 11 individual stakeholders, up from 5.4 in 2015. https://www.gartner.com/en/sales/insights/b2b-buying-journey ↩

2. Gartner; Forrester, "B2B Buyer Journey Research" (2023-2024). Buyers are 57-70% through their decision process before engaging sales. https://www.gartner.com/en/sales ↩

3. Salesforce, "State of Sales Report" (2024). Only 28% of sales reps hit annual quota, the lowest figure in six years. https://www.salesforce.com/resources/research-reports/state-of-sales/ ↩

4. QuotaPath, "2024 Compensation Trends Report." 91% of organizations missed quota expectations. Sales quotas rose 37% from 2023 to 2024 while average attainment dropped from 53% to 43%. https://www.quotapath.com/blog/sales-teams-miss-quota/ ↩ ↩2 ↩3

5. Bain & Company, "AI Is Transforming Productivity, but Sales Remains a New Frontier," Technology Report 2025. Sales reps spend approximately 25% of time on direct selling. AI can double active selling time and deliver 30%+ improvement in win rates. https://www.bain.com/insights/ai-transforming-productivity-sales-remains-new-frontier-technology-report-2025/ ↩ ↩2 ↩3

6. The Bridge Group, "2024 SDR Metrics & Compensation Report." Average SDR tenure 16-18 months, average ramp time 3.1-3.2 months. https://blog.bridgegroupinc.com/sdr-metrics ↩ ↩2

7. SuperAGI, "AI vs Human SDRs: A Comparative Analysis of Costs, Productivity, and Results in 2025." Human SDR fully loaded cost $110K-$150K annually vs. AI agent $10K-$12K. https://superagi.com/ai-vs-human-sdrs-a-comparative-analysis-of-costs-productivity-and-results-in-2025/ ↩ ↩2

8. Carnegie, Dale. How to Win Friends and Influence People. Simon & Schuster, 1936. Over 30 million copies sold worldwide. ↩

9. Rackham, Neil. SPIN Selling. McGraw-Hill, 1988. Based on analysis of 35,000 sales calls across 23 countries over 12 years. ↩

10. MEDDICC, "Who Created MEDDIC?" and Sales MEDDIC Group, "Origin of MEDDIC." Dick Dunkel created MEDDIC at PTC in 1996 under John McMahon. PTC grew from $300M to $1B in four years. https://meddicc.com/resources/who-created-meddic; https://www.salesmeddic.com/blog/origin-of-meddic ↩ ↩2 ↩3

11. Salesforce FY2025 Annual Report. $37.9 billion revenue. IDC: 20.7% CRM market share in 2024. https://www.macrotrends.net/stocks/charts/CRM/salesforce/revenue; https://www.salesforce.com/news/stories/idc-crm-market-share-ranking-2025/ ↩ ↩2

12. Dixon, Matthew and Brent Adamson. The Challenger Sale. Portfolio/Penguin, 2011. Study of 6,000 reps across 90 companies. 40% of top performers were Challengers; stars outperform average by 200% in complex sales. ↩ ↩2

13. Ross, Aaron and Marylou Tyler. Predictable Revenue. PebbleStorm, 2011. The SDR model used by Salesforce to add $100M in recurring revenue. ↩

14. Gong funding and revenue data. $584M total funding, $7.25B peak valuation (Series E, June 2021), $332M revenue in 2024. https://getlatka.com/companies/gong; https://www.gong.io/press/gong-raises-250-million-in-series-e-funding-at-7-25-billion-valuation ↩

15. Clari, Forrester Total Economic Impact Study (2024). 398% ROI, $96.2M in value, 98% forecast accuracy. https://www.clari.com/press/clari-revenue-ai-delivered-96-million-in-value-to-enterprise-customers/ ↩

16. Gong Labs, "State of Revenue AI Report" (December 2025). Analysis of 7.1M sales opportunities across 3,600+ companies. Teams using AI generate 77% more revenue per rep. https://venturebeat.com/ai/gong-study-sales-teams-using-ai-generate-77-more-revenue-per-rep; https://www.gong.io/press/new-gong-labs-research-finds-ai-is-now-a-trusted-decision-maker-in-revenue-teams ↩ ↩2 ↩3 ↩4

17. 11x, Series B announcement (November 2024). $50M Series B led by Andreessen Horowitz. ~$350M valuation. AI SDR "Alice" and AI phone agent "Mike." https://techcrunch.com/2024/09/30/11x-ai-a-developer-of-ai-sales-reps-has-raised-50m-series-b-led-by-a16z-sources-say/; https://www.globenewswire.com/news-release/2024/11/11/2978485/0/en/11x-Raises-a-50-Million-Series-B-Led-by-Andreessen-Horowitz-to-Accelerate-the-Era-of-Digital-Workers.html ↩

18. Artisan AI, Series A announcement (April 2025). $25M Series A. AI BDR "Ava." 250 paying customers, ~$5M ARR. https://www.artisan.co/blog/artisan-series-a ↩

19. Clay, Series C announcement (August 2025). $100M Series C at $3.1B valuation, led by CapitalG. $100M ARR in 2025, tripling year-over-year. 10,000+ customers. https://techcrunch.com/2025/08/05/clay-confirms-it-closed-100m-round-at-3-1b-valuation/; https://www.clay.com/series-c ↩

20. Gartner, Press Release (November 2025). "By 2028, AI agents will outnumber sellers by 10x." AI agents will intermediate >$15 trillion in B2B spending by 2028. https://www.gartner.com/en/newsroom/press-releases/2025-11-18-gartner-predicts-by-2028-ai-agents-will-outnumber-sellers-by-10x-yet-fewer-than-40-percent-of-sellers-will-report-ai-agents-improved-productivity; https://www.digitalcommerce360.com/2025/11/28/gartner-ai-agents-15-trillion-in-b2b-purchases-by-2028/ ↩ ↩2

21. Gartner, Press Release (September 2023). "By 2028, 60% of B2B seller work will be executed through conversational user interfaces via generative AI sales technologies." https://www.gartner.com/en/newsroom/press-releases/2023-09-21-gartner-expects-sixty-percent-of-seller-work-to-be-executed-by-generative-ai-technologies-within-five-years ↩ ↩2

22. Zuckerman, Gregory. The Man Who Solved the Market: How Jim Simons Launched the Quant Revolution. Portfolio/Penguin, 2019. Medallion Fund: 66% average annual returns before fees, 39% net, since 1988. https://www.cornell-capital.com/blog/2020/02/medallion-fund-the-ultimate-counterexample.html ↩ ↩2

23. Gartner, "The Role of Artificial Intelligence (AI) in Sales" (2025). "By 2027, 95% of seller research workflows will begin with AI." https://www.gartner.com/en/sales/topics/sales-ai ↩

24. Gartner, Press Release (September 2024). "Sellers who partner with AI are 3.7 times more likely to meet quota." Survey of 1,026 B2B sellers. https://www.gartner.com/en/newsroom/press-releases/2024-09-16-gartner-sales-survey-reveals-sellers-who-partner-with-ai-re-three-point-seven-times-more-likely-to-meet-quota ↩ ↩2

25. Gong, "State of Revenue AI Report" (2024). 89% of revenue organizations use AI-powered tools, up from 34% in 2023. https://www.gong.io/press/revenue-organizations-using-ai-in-2024-reported-29-percent-higher-sales-growth-than-their-peers-according-to-new-report-from-gong ↩

26. Gartner, Press Release (August 2025). "By 2030, 75% of B2B buyers will prefer sales experiences that prioritize human interaction over AI." https://www.gartner.com/en/newsroom/press-releases/2025-08-25-gartner-says-by-2030-that-75-percent-of-b2b-buyers-will-prefer-sales-experiences-that-prioritize-human-interaction-over-ai ↩ ↩2

27. Yang, Zeya, Marc Andrusko, and Angela Strange. "Death of a Salesforce: Why AI Will Transform the Next Generation of Sales Tech." Andreessen Horowitz, July 31, 2024. https://a16z.com/ai-transforms-sales/ ↩ ↩2

28. Sequoia Capital portfolio. Gong (valued at $7.25B) and Clay ($3.1B valuation). Sonya Huang, Partner, AI application layer investment thesis. https://sequoiacap.com/people/sonya-huang/; https://sequoiacap.com/podcast/training-data-amit-bendov/ ↩

29. Andreessen Horowitz, "AI Is Driving A Shift Towards Outcome-Based Pricing," December 2024 Enterprise Newsletter. https://a16z.com/newsletter/december-2024-enterprise-newsletter-ai-is-driving-a-shift-towards-outcome-based-pricing/ ↩ ↩2