The Present and Future of AI in Sales and GTM: A Deep Dive with Jason Lemkin and Kyle Norton, CRO at Owner

Artificial intelligence has officially entered the sales department, taken one look at the CRM, and asked why there are 47 duplicate contacts named “John.” It is not here merely to draft cold emails faster. The real opportunity is much bigger: AI is changing how companies identify demand, qualify buyers, prepare sellers, manage pipeline, retain customers, and build a modern go-to-market engine.

That was the central theme of a conversation between SaaStr founder Jason Lemkin and Kyle Norton, Chief Revenue Officer at Owner. Their discussion matters because it moves past the usual “AI can write an email” advice and toward a harder, more useful question: How should a revenue organization actually operate when AI is part of the team?

For sales leaders, founders, CROs, and GTM operators, the answer is not to replace every human with a chatbot wearing a tiny virtual blazer. It is to redesign work so AI handles repetitive, data-heavy, time-consuming tasks while people spend more time building trust, solving real business problems, and guiding buyers through complicated decisions.

Why AI in Sales and GTM Is Different This Time

Sales technology has always promised more productivity. CRM systems promised cleaner records. Sales engagement platforms promised better sequences. Conversation intelligence tools promised better coaching. Most of those promises were helpful, although many reps still ended their day manually updating fields while wondering whether “next steps” counts as a complete sentence.

AI changes the equation because it can work across those systems. It can read call transcripts, summarize account activity, identify buying signals, suggest next-best actions, enrich account records, draft follow-up messages, organize customer feedback, and surface risks before a deal quietly wanders into the wilderness.

That does not mean every AI workflow will produce instant revenue. In fact, companies that treat AI as a magical vending machine often end up with more tools, more dashboards, and more chaos. The strongest AI sales strategies begin with a business problem, not a shiny demo.

Jason Lemkin and Kyle Norton’s perspective is especially useful because it focuses on the operational side of AI. The important question is not, “Which AI tool should we buy?” It is, “Which parts of our go-to-market motion are slowing down revenue, frustrating customers, or wasting seller time?”

The Core Lesson: AI Should Expand Revenue-Generating Time

One of the clearest ideas from Norton’s GTM philosophy is that sales teams should spend a much larger share of their day doing actual revenue work. That sounds obvious until you look at the average seller’s calendar. Research, note-taking, CRM updates, internal handoffs, forecasting, pipeline reviews, and proposal edits can consume hours that should have gone toward customer conversations.

At Owner, AI has been used to reduce administrative work around CRM updates, notes, and pipeline management. The goal is not to make salespeople busier just for the thrill of seeing more green dots in Slack. The goal is to give them more time to listen, diagnose, advise, and close.

AI Should Remove Friction, Not Add Another Login

The best AI sales systems disappear into the workflow. A rep finishes a discovery call, and the system captures notes, updates key fields, summarizes objections, identifies action items, and creates a follow-up draft. The rep reviews it, adjusts the parts that need human judgment, and sends it.

That is far more valuable than asking every seller to open five different tools, copy information between them, and pretend this is “streamlined.” If AI adds clicks, it is not automation. It is just paperwork with better branding.

AI Can Turn Conversations Into Actionable Revenue Data

Sales calls contain a gold mine of information: competitor mentions, pricing concerns, product gaps, stakeholder names, implementation risks, urgency signals, and the dreaded phrase, “We need to think about it.” AI can help capture and classify this information at scale.

For example, an AI-enabled system can flag when a buyer repeatedly mentions a missing integration, when a rep promises something the product team has not approved, or when a decision-maker has not appeared in the process. Instead of discovering these issues during a painful forecast meeting, leaders can address them while the deal is still alive.

AI Should Improve Qualification Before It Improves Outreach Volume

One of the easiest ways to misuse AI is to send more bad outbound messages faster. Nobody asked for a thousand generic emails that begin with, “I noticed your company is doing amazing things in the space.” That sentence has caused more inbox damage than most people realize.

A better use of AI is to improve account selection and qualification. AI can analyze firmographic data, product usage, intent signals, website activity, hiring trends, customer fit, and historical win patterns. This gives sellers a better reason to contact a prospect in the first place.

In other words, AI should make outbound more relevant before it makes outbound more frequent.

What Owner’s Approach Reveals About AI-Native GTM

Owner serves independent restaurants, a market where many customers are busy operating real businesses rather than spending their afternoons comparing software categories. That makes GTM efficiency especially important. A restaurant owner wants to know whether a platform will help them grow direct orders, strengthen customer relationships, and reduce dependence on expensive third-party channels. They do not want a thirty-slide presentation about “synergistic workflow transformation.” Neither does anyone else, frankly.

Norton’s approach suggests that AI can create meaningful advantages when it is attached to a specific operating model. Instead of treating AI as an isolated project, a company can connect it across sales, onboarding, customer success, revenue operations, and demand generation.

Centralized AI Beats Scattered Experimentation

One important distinction is the difference between decentralized experimentation and centralized AI infrastructure. Giving every employee access to an AI assistant may increase curiosity, but it does not automatically create a scalable GTM system.

When each rep builds personal prompts, personal trackers, and personal automations, useful knowledge often stays trapped in individual workflows. One seller may create an excellent account-research prompt, while another develops a better follow-up structure. If those assets are never standardized, the organization never compounds its learning.

A centralized approach creates shared data, reusable prompts, consistent rules, approved messaging, common evaluation standards, and connected systems. The goal is not to eliminate creativity. It is to make the best ideas available to the entire revenue team instead of leaving them in one person’s browser history.

AI Needs Context Before It Can Be Trusted

AI performs best when it has high-quality context. That includes product information, pricing guidance, customer stories, ideal customer profiles, competitor positioning, security policies, implementation timelines, and historical deal outcomes.

Without context, AI can produce polished nonsense at remarkable speed. It may write a beautiful email that references the wrong company, misunderstands a prospect’s industry, or confidently invents a customer problem. The grammar may be perfect. The embarrassment will also be perfect.

That is why a strong AI sales stack requires a well-maintained context library. Revenue teams need a single source of truth for the information agents and sellers use in prospecting, qualification, deal support, onboarding, and customer communication.

The Future of AI in Sales: From Assistant to Revenue System

The future of AI in sales will not be one giant robot salesperson closing every deal while humans relax on a beach. It will be a network of specialized AI agents handling different parts of the revenue process under human oversight.

AI Agents Will Handle More Operational Work

Sales teams are likely to rely on AI agents for account research, lead routing, data enrichment, follow-up reminders, opportunity hygiene, conversation summaries, renewal risk monitoring, and next-best-action recommendations. These systems will become increasingly connected to CRM platforms, product data, call recordings, support tickets, and marketing activity.

The key word is orchestration. The future CRO will not simply manage salespeople, territories, and quotas. They will also manage workflows, data quality, agent performance, automation guardrails, and the human-to-AI ratio across the revenue organization.

Salespeople Will Become More Strategic, Not Less Important

As buyers gain easier access to information, salespeople will need to offer something harder to automate: judgment. Buyers can ask AI to compare vendors, summarize reviews, create implementation checklists, and explain product categories. What buyers still need is help navigating internal politics, aligning stakeholders, calculating business value, reducing risk, and making confident decisions.

The modern seller will spend less time acting as a walking brochure and more time acting as a guide. The best reps will understand the customer’s business deeply enough to challenge assumptions, uncover hidden risks, and help the buying committee reach consensus.

GTM Teams Will Shift From Funnels to Signals

Traditional GTM reporting often focuses on stage-based funnels: leads, meetings, opportunities, pipeline, and closed revenue. Those metrics will still matter, but AI allows teams to pay closer attention to live signals.

Signals can include a pricing-page visit, a product trial milestone, a support-ticket pattern, a new executive hire, a competitor mention during a call, a sudden drop in product usage, or a buyer who keeps opening implementation documentation at midnight. AI can connect these signals and suggest what should happen next.

This creates a more responsive go-to-market motion. Instead of running generic campaigns on a fixed calendar, teams can react to buyer behavior with more precision and better timing.

The Risks: AI Can Scale Mistakes as Easily as Success

AI can make a strong sales process stronger. It can also make a weak sales process louder, faster, and more difficult to clean up. Companies should not automate workflows they do not understand.

Bad Data Produces Bad Recommendations

If CRM records are incomplete, product data is outdated, or customer information is scattered across systems, AI will struggle to produce reliable outputs. Before launching ambitious agents, teams should audit their data sources, permissions, field definitions, and ownership rules.

A simple principle applies: an AI system cannot become more trustworthy than the information it receives.

Human Review Still Matters in High-Stakes Moments

AI-generated messaging can be useful for research, first drafts, and routine follow-up. However, high-value accounts, sensitive customer issues, pricing commitments, legal claims, and executive communications should still receive human review.

Trust is difficult to earn and absurdly easy to lose. A seller who sends an inaccurate AI-generated message to a strategic prospect may not get another chance to explain that “the model misunderstood.” Buyers do not care. They just see a vendor that failed to do basic homework.

Governance Is a Revenue Issue, Not Just an IT Issue

AI governance is often treated as a compliance topic. It is also a revenue topic. If customers do not trust how a company handles their data, recommendations, communications, or security, sales cycles become harder and retention becomes more fragile.

Revenue leaders should work closely with legal, security, product, and IT teams to define what AI can access, which actions require approval, how outputs are evaluated, and how sensitive data is protected. The winning AI strategy is not reckless automation. It is responsible leverage.

A Practical 90-Day AI Sales and GTM Plan

Companies do not need to rebuild their entire revenue organization overnight. A sensible AI GTM strategy can begin with a narrow, measurable workflow.

Days 1–30: Find the Revenue Friction

Interview sellers, sales managers, customer success leaders, and revenue operations teams. Ask where time disappears, where information gets lost, and which tasks create the most frustration. Review call-to-close workflows, CRM hygiene, lead routing, proposal creation, onboarding handoffs, and renewal processes.

Choose one painful problem with a clear metric. Examples include reducing CRM update time, improving lead-response speed, increasing qualified meetings, shortening proposal turnaround, or identifying churn risk earlier.

Days 31–60: Build, Test, and Measure

Create a controlled AI workflow with defined inputs, expected outputs, review requirements, and success metrics. Train the system on approved company knowledge rather than random internet scraps. Test it with a small group of users before releasing it broadly.

Measure business outcomes, not vanity metrics. A thousand AI-generated emails do not matter if qualified pipeline falls. Ten saved hours per rep do not matter if those hours turn into longer internal meetings. Measure revenue-generating activity, conversion quality, sales-cycle movement, retention signals, and customer satisfaction.

Days 61–90: Standardize What Works

Once a workflow proves useful, document it. Create templates, data standards, training materials, escalation rules, and quality checks. Put ownership in the hands of a cross-functional team that includes sales, RevOps, marketing, customer success, and technical leaders.

The goal is to create a repeatable system, not a heroic one-off experiment performed by the one employee who knows how to make three tools talk to each other.

Conclusion: AI Will Not Replace GTM Strategy, but It Will Expose Weak GTM Strategy

The conversation between Jason Lemkin and Kyle Norton offers a useful reality check for revenue leaders. AI is not simply a productivity feature. It is becoming a core operating layer for modern sales and go-to-market teams.

The companies that benefit most will not be the ones that buy the most AI tools. They will be the ones that redesign workflows, centralize useful knowledge, improve data quality, train their teams, and protect buyer trust. AI can make average processes faster, but it can make excellent processes extraordinary.

The future belongs to human-AI revenue teams where agents handle the repetitive work, systems surface the right signals, and people focus on the moments that require judgment, empathy, credibility, and business insight. That is not the end of sales. It is sales finally getting to spend more time selling.