The Physics of AI GTM
Ross Sylvester, Co-Founder & CEO, Adrata | Feb 2026 | ~12 min read
Every generation of sales technology promises to rewrite the rules. AI is different — not because it breaks the laws of go-to-market, but because it finally makes those laws visible.
I studied physics in college. Not because I wanted to build particle accelerators, but because physics offered something no other discipline did: a set of universal laws that explained why systems behave the way they behave. Not what happens. Why it happens.
Fifteen years in B2B sales taught me that go-to-market has its own physics. There are conservation laws, entropy gradients, speed limits, and gravitational fields. We just never had the instrumentation to see them clearly. CRM data was too coarse. Activity logs were too noisy. Pipeline reviews were impressionistic — vibes dressed up as forecasts.
AI changes the instrumentation. For the first time, we can observe the underlying dynamics of how deals actually move — and don't move — through a revenue system. What we're seeing isn't a new set of rules. It's the old rules, finally measured with precision.
Here are the five laws I've come to believe govern every AI-powered go-to-market motion. Understanding them won't give you a silver bullet. But it will keep you from violating fundamental constraints that no amount of technology can overcome.
The First Law: Conservation of Revenue Energy
The total effort required to close a deal is constant. AI doesn't reduce it. AI redistributes where it's spent.
In thermodynamics, the first law states that energy cannot be created or destroyed — only converted from one form to another. The same principle holds in revenue.
Every closed deal requires a fixed quantum of work: research, personalization, objection handling, trust-building, consensus formation, commercial negotiation, legal review. This total hasn't changed in twenty years. What has changed — dramatically — is where humans spend their portion of it.
The data makes this clear. Companies deploying AI across their GTM stack report saving an average of 12 hours per rep per week on administrative and research tasks. Pipeline generation is up 23%. Companies leveraging AI-driven CRM systems see up to 40% higher lead-to-opportunity conversion rates compared to static approaches. Revenue increases of 3% to 15% are common, with sales ROI improvements of 10% to 20%.
But here's the conservation law in action: win rates are simultaneously down 18% over the past two years. More pipeline, more activity, more AI — and fewer wins.
How is that possible? Because most organizations are using AI to amplify the wrong phase of the revenue cycle. They're pouring energy into top-of-funnel generation — more emails, more sequences, more "personalized" outreach — while the actual bottleneck is mid-funnel and late-funnel work that AI can assist with but cannot perform.
The effort didn't disappear. It shifted. And the organizations that shifted it to the wrong place are generating more pipeline that converts worse.
The practical implication for CROs: Stop measuring AI's impact by volume metrics — emails sent, meetings booked, pipeline created. Start measuring it by conversion efficiency at each stage. If your top-of-funnel is up 40% but your Stage 2-to-Stage 3 conversion is flat, you haven't saved energy. You've wasted it upstream while starving the stages that actually produce revenue.
The highest-performing teams we work with have inverted the typical AI allocation. They spend less AI energy on prospecting and more on deal intelligence: mapping buying committees, identifying stakeholder sentiment, surfacing risk signals, and arming champions with the materials they need to sell internally. They didn't reduce the total work. They used AI to move rep time from work that machines can do to work that only humans can do.
Conservation of revenue energy means every hour of rep time you free up must be deliberately reinvested — not in more activity, but in higher-quality activity at the stages where deals actually die.
The Second Law: Entropy in Pipeline
Without the continuous application of intelligence, pipeline naturally decays toward disorder.
The second law of thermodynamics states that in any closed system, entropy — disorder — always increases. Left alone, hot things cool, structures crumble, and organized systems drift toward chaos. The only way to counteract entropy is to apply energy from outside the system.
Your pipeline is a thermodynamic system. And it obeys the second law with ruthless consistency.
Consider the data. Lead databases lose 10% to 20% of their validity every quarter as contacts change roles, companies restructure, and buying priorities shift. Deals without a scheduled next step for more than seven days see win rates collapse by 65%. A single canceled meeting reduces deal progression by 18%; two cancellations reduce it by 58%. And across the industry, 86% of B2B purchases stall at some point during the buying process.
This is entropy. Without continuous, targeted energy input — a rep following up, a champion re-engaging, new information entering the deal — every opportunity degrades toward "no decision." The natural state of pipeline is not progress. It is decay.
Pre-AI, fighting pipeline entropy was an exhausting manual process. Reps maintained deal momentum through sheer force of will — follow-up calls, check-in emails, periodic nudges. The problem was that human attention is finite and unevenly distributed. The deals that got attention were the ones that were already moving. The deals that needed attention — the ones silently decaying — were invisible until it was too late.
AI changes the entropy equation in two fundamental ways.
First, it makes decay visible. Machine learning models can detect the early signatures of pipeline entropy — declining email engagement, lengthening response times, stakeholder disengagement, language shifts in communications — before human intuition catches them. When a deal's entropy rate spikes, the system can flag it while intervention is still possible.
Second, it reduces the cost of applying counter-entropic energy. Automated nudges, AI-generated status updates, predictive next-best-actions, and intelligent scheduling all reduce the effort required to keep a deal in motion. This doesn't eliminate the need for human energy — it ensures that human energy is directed at the deals where it matters most.
But here's where the second law bites back: AI can also accelerate entropy if deployed carelessly. Automated sequences that fire without context. "Personalized" emails that read as obviously machine-generated. Follow-ups that demonstrate no memory of prior conversations. Each of these interactions doesn't just fail to reduce entropy — it actively increases it by eroding the buyer's confidence that they're dealing with a competent, attentive seller.
The practical implication for CROs: Build an entropy dashboard. For every deal in pipeline, track the rate of decay: days since last meaningful engagement, stakeholder response latency, meeting cancellation rate, champion activity level. AI should power this dashboard and recommend interventions. But the interventions themselves — the energy that reverses entropy — must be substantive. A templated "just checking in" email is not counter-entropic energy. A message that references the buyer's latest earnings call and connects it to the business case you've been building together — that's energy.
The second law tells you that pipeline maintenance is not optional. It is a thermodynamic requirement. Organizations that treat pipeline as a set-it-and-forget-it asset are fighting physics.
The Third Law: The Speed Limit of Trust
AI can accelerate research, personalization, and process. It cannot accelerate trust. Trust has a human speed limit.
In physics, the speed of light is an absolute constraint. No matter how much energy you apply, you cannot make information travel faster than c. You can get asymptotically close, but the limit is real and non-negotiable.
Trust is the speed of light of go-to-market.
The data here is striking in its consistency. B2B buyers spend 70% of their buying journey doing independent research before ever engaging a seller. They average 16 interactions with the winning vendor across the full cycle. Opportunities that include live meetings close 32 days faster and at significantly higher win rates. And despite the automation revolution, Gartner projects that by 2030, 75% of B2B buyers will prefer sales experiences that prioritize human interaction over AI.
These numbers point to a constraint that no technology can override. Trust is built through demonstrated competence, consistent follow-through, and genuine understanding of the buyer's situation — all of which require time and human presence. You can use AI to arrive at the trust-building moment faster. You cannot use AI to make the trust-building moment shorter.
This creates a paradox that many AI-first GTM teams are discovering the hard way. AI lets you reach more buyers faster. But reaching buyers faster doesn't mean you can earn their trust faster. And deals where trust hasn't been established don't close — they stall, slip, or die in "no decision."
The 6sense Buyer Experience Report found that 81% of B2B buyers already have a preferred vendor at the time of first seller contact. That preference wasn't formed by outbound sequences or automated emails. It was formed through peer recommendations, analyst reports, community discussions, and content consumption — trust signals that accumulated long before a sales rep entered the picture.
Meanwhile, Salesforce research shows 61% of customers say advances in AI make it even more important for companies to be trustworthy. The irony is precise: the more AI a company deploys in its sales process, the higher the trust bar becomes. Buyers who sense automation become more skeptical, not less. They're looking for signals that a real human understands their real problem.
High-performing teams respect the speed limit. They use AI to compress everything around the trust interaction — research, prep, scheduling, follow-up, internal coordination — so that every moment of human-to-human contact is maximally valuable. The rep walks into the meeting already knowing the buyer's tech stack, recent executive changes, competitive landscape, and likely objections. That preparation is AI-driven. The conversation itself — the trust-building moment — is irreducibly human.
The practical implication for CROs: Audit your sales process for what I call "trust density" — the ratio of trust-building human interactions to total buyer touchpoints. If AI is increasing your total touchpoints but your trust density is declining, you're accelerating past the speed limit and the physics will catch up. The symptom is unmistakable: more pipeline, more activity, more first meetings — and a cratering Stage 2 to close-won rate. You're generating interest faster than you're building trust.
The speed limit also explains why the deals involving the most stakeholders often have the highest close rates when managed well. Each stakeholder interaction is a trust-building opportunity. A deal with 10 engaged stakeholders has 10 trust relationships in play — 10 people who can independently validate that your team is competent, responsive, and honest. Multi-threaded deals don't just de-risk the sales process. They compound trust. Gong Labs found that multi-threading boosts win rates by 130% in deals over $50K, and UserGems data shows multi-threaded deals achieve a 5x higher win rate. That's not a process advantage. That's a trust advantage, distributed across the buying committee.
The Law of Gravity: The Pull of the Status Quo
The larger the organization, the stronger the gravitational pull against change. AI cannot overcome gravity — but it can help you calculate the escape velocity.
In celestial mechanics, gravity is a function of mass. The more massive an object, the stronger its gravitational field, and the more energy required to escape its pull. Earth's escape velocity is 11.2 km/s. Jupiter's is 59.5 km/s. The physics doesn't care about your ambition. It cares about mass.
Enterprise sales has its own gravitational field, and it is called the status quo.
The research on status quo bias in purchasing decisions is unambiguous. Between 40% and 60% of B2B deals end in "no decision" — not lost to a competitor, but lost to inaction. And this isn't because buyers don't feel pain. Dixon and McKenna's analysis of 2.5 million sales conversations for The JOLT Effect found that of deals lost to inaction, only 44% were due to genuine status quo preference. The remaining 56% — the majority — were lost to indecision. Buyers who wanted to change, acknowledged the need for change, and still couldn't pull the trigger.
This is gravity. And it scales with organizational mass.
A 50-person startup can make a purchasing decision in a week. A single founder or VP can evaluate, decide, and sign. The gravitational pull of the status quo is weak because the organization's mass is small — fewer stakeholders, fewer processes, fewer institutional habits to overcome.
A 5,000-person enterprise is Jupiter. The average buying committee now includes 6 to 10 decision-makers, each bringing 4 to 5 pieces of independent research. Gartner reports that 74% of these buyer teams experience "unhealthy conflict" during the decision process. Procurement has its process. Legal has its process. Security has its process. Each process is a layer of organizational mass, and each layer increases the escape velocity required to break free from "the way we do things now."
The gravitational metaphor explains something that frustrates every enterprise rep: the close rate on large deals isn't just lower because they're more complex. It's lower because the gravitational field is exponentially stronger. A $25K deal at a mid-market company might face one layer of status quo inertia. A $500K deal at a Fortune 500 faces dozens of layers — and those deals take 270 days to close, more than 10x longer than deals under $1K.
So what role does AI play against gravity?
AI cannot overcome status quo gravity any more than a rocket can eliminate Earth's gravitational field. But AI can do what NASA does: calculate the precise escape velocity, optimize the trajectory, and ensure that every unit of thrust is applied in the right direction at the right time.
Concretely, this means:
Mapping the gravitational field. AI can identify every stakeholder, process, and institutional bias that constitutes the organizational mass working against change. Who are the decision-makers? What are their historical buying patterns? Where have similar deals stalled in this organization before? This is gravitational cartography — understanding the field before you try to launch through it.
Calculating escape velocity. For every deal, there is a minimum amount of business case, stakeholder alignment, and risk mitigation required to overcome the status quo. AI can estimate this based on the organization's size, industry, buying history, and committee composition. If you're under-investing relative to the escape velocity — if your business case lacks the thrust to overcome the gravity — AI can flag that before you waste six months on a deal that was never going to launch.
Sequencing the burn. Rockets don't fire all their fuel at once. They stage their burns — boosting at precise moments when the physics favor acceleration. AI can identify the optimal moments to apply energy to a deal: when a stakeholder's engagement peaks, when a competitive threat creates urgency, when budget cycles align, when organizational change creates a window of reduced gravitational resistance.
The practical implication for CROs: Stop blaming reps when enterprise deals stall. Start measuring the gravitational field. Every deal review should include an explicit assessment: what is the organizational mass working against this change, and does our current trajectory generate enough escape velocity to overcome it? If the answer is no, the correct response isn't "try harder." It's "apply thrust differently" — or redirect the energy to a deal where the physics are more favorable.
The hardest discipline in enterprise sales is walking away from a deal where the gravity is simply too strong. AI can make that calculus less emotional and more physical.
The Law of Relativity: Time Moves Differently for Buyers and Sellers
A sales cycle that feels like three months to a seller feels like three years to a buyer. And vice versa. This perceptual asymmetry is the hidden cause of most deal failures.
Einstein's special relativity demonstrated that time is not absolute. Two observers moving at different velocities experience time differently. An astronaut traveling near the speed of light ages more slowly than her twin on Earth. The clock ticks at the same rate for both. But the experience of time diverges.
Buyers and sellers exist in different reference frames, and their experience of the same deal timeline diverges in ways that are nearly as dramatic.
For the seller, a deal is one of 30 to 50 opportunities in their pipeline. Each individual deal occupies a few hours of attention per week. The three-month sales cycle feels like three months — a sustained but manageable effort among many.
For the buyer, the same deal is one of the biggest decisions they'll make this year. It requires coordinating across departments, building internal consensus, defending budget allocation, and putting their professional reputation on the line. Those three months feel like an eternity of internal meetings, slide decks, vendor evaluations, and political navigation. And the buyer isn't just managing your deal. They're doing their actual job — the one they were hired for — while squeezing vendor evaluation into the margins.
The data captures this perceptual gap precisely. Buyers only spend 17% of their total buying time meeting with vendors — all vendors. For any given seller, you're getting a fraction of that 17%. Meanwhile, 50% of sales reps think they aren't pushy — but 84% of buyers describe their experience with sellers as pushy. That isn't a communication problem. It's a relativistic one. What feels like "appropriate persistence" in the seller's reference frame feels like "relentless pressure" in the buyer's.
This explains one of the most puzzling patterns in B2B sales: deals that are progressing well from the seller's perspective suddenly go dark. The buyer doesn't respond for two weeks. The champion stops returning calls. The deal that was "90% committed" evaporates.
What happened? Nothing changed in the seller's reference frame. But in the buyer's reference frame, two weeks is half a budget cycle. A reorg happened. A competing priority emerged. The political landscape shifted. The buyer's experience of those two weeks was not silence — it was chaos that had nothing to do with your deal.
Three-quarters of B2B buyers now say they prefer a rep-free sales experience. But research also shows that self-service purchases lead to higher buyer's remorse. This is the relativistic paradox: buyers want to move at their own speed (which often feels faster internally), but when they do, they miss details that only a skilled human guide can surface.
AI's role in relativity is to synchronize the reference frames.
The best AI systems function as translation layers between the buyer's timeline and the seller's timeline. On the seller side, AI tracks buyer engagement signals — email opens, document views, website visits, stakeholder activity — to create a real-time map of how the deal is moving in the buyer's reference frame. When the buyer goes dark, the AI can distinguish between "they're busy with a board meeting this week" and "the champion just lost their internal sponsor" — information that would take a human rep days or weeks to surface.
On the buyer side, AI-powered tools can reduce the administrative burden of the evaluation process — automating vendor comparisons, generating internal business cases, and streamlining the approval workflow. This doesn't just save the buyer time. It compresses their experience of the deal timeline, reducing the perceptual gap.
The practical implication for CROs: Train your team to think in two clocks. Every deal should have a seller timeline and a buyer timeline, and the AI should be tracking both. When the two diverge — when the seller thinks the deal is on track but buyer engagement signals suggest otherwise — that's your early warning system. The deal isn't dying because the buyer lost interest. It's dying because time is moving differently in their reference frame, and your team didn't notice.
Build buyer-time metrics into your forecasting model. A deal where the buyer is engaging weekly is on a different clock than a deal where the buyer engages monthly — even if both deals have been in pipeline for the same number of calendar days. AI can measure this. Your CRM, left to its own devices, cannot.
Putting the Laws Together
These five laws compose a unified field theory of AI-powered go-to-market:
- Conservation: The work to close a deal is constant. AI redistributes it — invest in the stages that matter.
- Entropy: Pipeline decays without intelligence. AI makes decay visible and reduces the cost of fighting it.
- Speed Limit: Trust cannot be accelerated beyond a human rate. AI should compress everything else to maximize trust density.
- Gravity: Status quo resistance scales with organizational mass. AI calculates escape velocity and optimizes trajectory.
- Relativity: Buyer and seller time are not equivalent. AI synchronizes the reference frames.
The organizations that will win the AI GTM era are not the ones that deploy the most AI. They're the ones that deploy AI in accordance with these laws. Pouring AI into prospecting while ignoring pipeline entropy violates the second law. Automating every buyer touchpoint violates the third. Chasing enterprise deals without measuring gravitational resistance violates the fourth. Forecasting in seller-time without tracking buyer-time violates the fifth.
Physics doesn't reward effort. It rewards understanding. The laws don't care how hard you try or how much technology you deploy. They care whether your system respects the constraints that govern how deals actually move.
The good news: unlike actual physics, these laws were never visible before. Now they are. And visibility is the first step toward leverage.
The data referenced in this article draws from Gartner's B2B buying research (2023-2025), Forrester's State of Business Buying (2024), 6sense's B2B Buyer Experience Report (2024-2025), Ebsta and Pavilion's B2B Sales Benchmarks (2024), Dixon and McKenna's analysis of 2.5 million sales conversations for The JOLT Effect, Gong Labs' analysis of 1.8 million deals, UserGems' analysis of 5,000+ B2B SaaS opportunities, and Outreach's Sales 2025 Data Report.