GTM Strategy

AI Won’t Save a Broken GTM—But It Will Scale What Works

Buying AI is a technology decision. Creating value from AI is an operating-model decision.

By Big Wheel Performance · 2026-07-16

Every founder, CEO, board member, and revenue leader seems to be asking the same question right now:How do we use AI to grow faster?It is a fair question, but I think it is the wrong place to start. The better question is: Do we actually have a go-to-market system worth scaling? Because AI is not going to fix a broken GTM engine; it is going to accelerate whatever is already happening. If your pipeline is filled with weak, poorly qualified opportunities, AI will help you create more of them, faster. If your sales stages are inconsistent, AI may give you better-looking dashboards, but the underlying process still will not convert. If your ideal customer profile is broad enough to include almost anybody with a budget, AI can help you reach thousands more companies at a fraction of the cost. Unfortunately, many of them will still be the wrong companies. And if your positioning sounds like everybody else in your category, AI can generate an endless supply of polished content that nobody remembers.That is not leverage. That is inefficiency operating at machine speed.<h2>We Have Seen This Movie Before</h2>AI may feel different from every previous wave of business technology. In many ways, it is.The speed, capacity, and potential impact are unlike anything most of us have experienced during our careers, but the way companies are responding to it feels very familiar.I have spent more than 30 years helping build, lead, and improve commercial organizations. During that time, I have watched companies invest heavily in CRM systems, marketing automation, sales-enablement platforms, revenue operations, intent data, predictive analytics, digital transformation, and just about every other technology that promised to change how businesses grow.Many of those technologies created enormous value…but many did not.The difference was rarely the quality of the software, but rather the quality of the operating model underneath it.The companies that benefited were usually the ones that understood their market, had a clear sales process, maintained reasonable data discipline, defined ownership, and established a consistent operating cadence.The companies that struggled often used the technology to avoid dealing with those issues.They bought a CRM instead of defining a sales process. They bought marketing automation instead of clarifying their audience and message. They bought forecasting software instead of enforcing pipeline discipline. They built RevOps teams without first determining what the revenue organization was supposed to optimize.  Now we are watching the same pattern repeat itself with AI. The technology is more powerful, and the temptation to skip the fundamentals is even greater.<h2>The Problem Isn’t Adoption. It’s Sequence.</h2>AI adoption is no longer a fringe activity.McKinsey’s 2025 global survey found that 88% of respondents said their organizations were regularly using AI in at least one business function. Yet only about one-third said their companies had begun scaling their AI programs across the organization, and just 39% reported any measurable enterprise-level EBIT impact. That gap matters.Most companies now have access to AI, but far fewer have figured out how to turn that access into meaningful economic value.The issue is not a lack of tools. It is the order in which those tools are being introduced.Companies are:Deploying AI-generated outbound before clearly defining their ICP, implementing AI forecasting before establishing consistent pipeline stages.Automating account research before deciding which accounts actually deserve attention.Generating more content before validating whether their positioning means anything to the buyer.Deploying customer-health models without defining who owns the intervention when an account is at risk.<ul><li></li></ul>The technology gets added first. The process gets questioned later—usually after the expected results fail to appear.McKinsey found that redesigning workflows was one of the practices most closely associated with bottom-line impact from AI. Its research also found that only a minority of companies had fundamentally redesigned the workflows where AI was being introduced.That is the distinction too many organizations are missing.Buying AI is a technology decision. Creating value from AI is an operating-model decision.You cannot simply bolt intelligence onto a weak process and expect the process to become intelligent. Structure still has to come before scale. You cannot automate your way to clarity.<h2>AI Is an Amplifier, Not an Architect</h2>AI is remarkably good at processing information, finding patterns, summarizing activity, recommending actions, and executing repeatable work.What it cannot do is:<ul><li>Make the most important strategic decisions for the business.</li><li>Decide which market you should pursue.</li><li>Determine which customers are economically attractive and strategically valuable.</li><li>Define why a buyer should choose you instead of a competitor.</li><li>Create accountability across Sales, Marketing, Customer Success, Delivery, and Finance.</li><li>Fix a culture where people protect weak opportunities because nobody wants pipeline coverage to decline.</li><li>Decide whether a stalled deal deserves more effort or needs to be disqualified.</li><li>Replace the leadership judgment required to make hard choices.</li><li></li></ul>AI can research 1,000 accounts. It cannot tell you whether those 1,000 accounts belong in the right market.AI can write a personalized email for every prospect. It cannot rescue a value proposition that is generic, irrelevant, or disconnected from the buyer’s priorities.AI can tell you that a deal has lost momentum. It cannot make the seller ask the uncomfortable question.AI can flag a customer as a churn risk. It cannot create ownership when Sales and Customer Success disagree about who is responsible for the relationship.The technology performs inside the architecture you give it. If the architecture is strong, AI can become an extraordinary force multiplier. If the architecture is weak, AI simply helps the weakness move faster.<h2>Activity Has Never Been the Same as Progress</h2>One reason companies are attracted to AI is that its output is immediately visible.More emails get written.More prospects get researched.More content gets produced.More calls get summarized.More contacts get scored.More reports get generated.The organization feels faster.That feeling can be deceptive. Output is easy to measure. Progress is harder.A team can double the number of outbound messages it sends while producing fewer qualified conversations. Marketing can dramatically increase content production without creating more market awareness, trust, or demand. Sales can automate CRM updates while continuing to enter poor opportunities, inaccurate close dates, and inconsistent stages. Customer Success can generate hundreds of account alerts without improving retention because nobody acts on them.We should not confuse the increased production of work with the increased creation of value. AI has lowered the cost of activity. It has not lowered the cost of bad strategy. And it has not made judgment optional.<h2>Where AI Actually Creates Value</h2>This is not an argument against AI. I am a strong believer in its potential. Once the fundamentals are in place, AI can improve the economics of a revenue organization in ways that were not possible even a few years ago. The key is applying it to a defined business constraint inside a functioning system.<h3>Better Account Prioritization</h3>AI can analyze firmographic data, leadership changes, hiring activity, funding events, technology adoption, engagement, intent signals, and historical conversion patterns far more efficiently than a human team.That can help companies distinguish between accounts that fit the ICP and accounts that may actually be ready to act, but fit and timing are different.An organization can show a great deal of online activity and still be a poor potential customer. Another can resemble your best customers but have no urgency, budget, or internal sponsorship.AI becomes valuable when the organization has already defined:<ul><li>Which customers are most likely to receive meaningful value.</li><li>Which customers are most likely to be profitable.</li><li>Which business conditions create urgency.</li><li>Which signals actually correlate with buying behavior.</li><li>Which accounts the company should deliberately avoid.</li><li></li></ul>Without that clarity, AI is not prioritizing your market. It is ranking signals without understanding the strategy behind them. It can be extremely precise and still be strategically wrong.<h3>Better Seller Preparation</h3>AI can summarize company information, identify likely priorities, analyze earnings calls, suggest discovery questions, find relevant customer stories, and prepare sellers for meetings.That matters because buyers are also using AI.HubSpot’s 2025 sales research found that buyers are increasingly using AI to research solutions, compare vendors, and even prepare RFPs before speaking with a seller. Gartner subsequently reported that 69% of B2B buyers still preferred to validate AI-generated information with a sales representative.That creates an important shift.The buyer may be more informed before the first conversation, but that does not make the seller less important. It changes what the buyer needs from the seller. They do not need somebody to repeat information they could find online. They need judgment and context.They need someone who can challenge assumptions, understand the business problem, navigate risk, and connect the solution to an economic outcome.AI can help prepare the seller for that conversation, but it cannot make the seller credible. Technology can improve readiness, but it cannot replace trust.<h3>Better Pipeline and Forecast Management</h3>AI can identify missing stakeholders, declining engagement, unusual stage duration, unrealistic close dates, and patterns associated with opportunities that historically slipped or were lost.That can make pipeline reviews faster, more objective, and more useful, but only when the data underneath the analysis is reliable.If one seller advances an opportunity after completing a demo while another waits until the buyer confirms a business case, the same stage represents two completely different levels of buyer commitment.If close dates are changed every few weeks to keep opportunities in the forecast, AI learns from manipulated data.If weak deals remain open because leadership is afraid of losing pipeline coverage, AI does not eliminate the fiction. It incorporates the fiction into its prediction.Salesforce’s 2026 State of Sales report found that 46% of sales professionals using AI agents said data-quality issues were hurting their sales efforts.AI does not turn poor CRM data into truth. It turns poor data into a more confident recommendation.<h3>Better Retention and Expansion</h3>The existing customer base may provide one of the strongest opportunities for near-term AI value.AI can monitor product usage, support activity, engagement, sentiment, payment behavior, adoption milestones, stakeholder changes, and other signals that may indicate risk or opportunity.It can:<ul><li>Identify potential churn sooner.</li><li>Surface expansion opportunities earlier.</li><li>Help teams prepare for renewals, customer reviews, and account-planning discussions.</li><li></li></ul>But a signal only creates value when the company knows what to do with it. A churn alert means little when nobody owns the intervention. An expansion recommendation goes nowhere when the account executive and customer-success manager disagree about responsibility. A cross-sell opportunity is unlikely to convert when the customer has not realized value from the original purchase.AI can make the opportunity visible, but the operating model must determine what happens next.<h3>Better Productivity</h3>AI can reduce the time sellers spend on research, CRM updates, call summaries, proposals, routine emails, internal searches, and administrative work.HubSpot reported that 84% of sales professionals using AI said it saved time and optimized processes. Gartner found that AI tools were saving sellers an average of 4.8 hours per week.That sounds like a significant productivity gain.But Gartner also found that 72% of sales organizations were failing to consistently reinvest that saved time into higher-value selling activities.That may be one of the most important findings in this entire conversation.Saving time does not automatically create value.The organization has to decide what people should do with the capacity they get back.<ul><li>Should sellers spend more time with customers?</li><li>Develop deeper account plans?</li><li>Pursue executive relationships?</li><li>Improve follow-up?</li><li>Coach other team members?</li><li>Build new pipeline?</li><li></li></ul>If leadership does not define the higher-value behavior, the saved time will disappear into more internal meetings, more low-quality outreach, or simply more unused capacity. Productivity is not measured by how quickly a task is completed. It is measured by what the company accomplishes with the capacity it creates.<h2>The Question Every Leadership Team Should Ask</h2>Before investing heavily in AI across the revenue organization, I would ask one question:If we doubled the output of our current GTM system tomorrow, would we generate materially more profitable revenue?Not more emails.Not more MQLs.Not more meetings.Not more contacts.Not more opportunities entered into the CRM.More profitable revenue.If the answer is clearly yes, AI may be exactly the right investment.The system works, and the business needs more throughput, greater consistency, faster insight, or better economics.But if the answer is no—or even “we are not sure”—then the constraint is probably structural.Maybe:<ul><li>The ICP is too broad.</li><li>The company’s message creates interest but no urgency.</li><li>Sellers are advancing opportunities based on their own activity rather than buyer commitment.</li><li>Marketing is rewarded for lead volume while Sales is responsible for revenue.</li><li>The company is winning customers it cannot successfully implement, serve, or retain.</li><li>Every seller is following a different process.</li><li>The forecast is negotiated rather than derived from evidence.</li><li></li></ul>None of those problems improve when you increase activity. More volume into a broken funnel does not create scale. It creates more waste to manage.<h2>What I Would Fix Before Scaling AI</h2>Before deploying AI broadly across GTM, I would make sure five things are in place.<h3>1. A Defensible ICP</h3>Know who you are built to serve—and who you are not.Understand the problems you solve exceptionally well, the conditions that create urgency, the customers most likely to adopt successfully, and the accounts most likely to generate attractive long-term economics.A real ICP is not just a description of where you have sold. It is a decision about where you should invest next.<h3>2. Positioning Buyers Actually Care About</h3>AI can create hundreds of variations of your message, but it cannot make the underlying message matter.The company still needs to clearly explain:<ul><li>What problem it solves.</li><li>Why that problem matters now.</li><li>What makes its approach different.</li><li>What measurable outcome it creates.</li><li>Why the buyer should believe it can deliver.</li><li></li></ul>HubSpot’s 2026 marketing research makes a related point: as AI increases content volume, brands without a distinct point of view are becoming harder to notice and remember.The danger is no longer just creating bad content. It is creating perfectly competent content that nobody cares about.<h3>3. Pipeline Stages Based on Buyer Commitment</h3>Pipeline stages should reflect what the customer has done, agreed to, or committed to; they should not merely record seller activity.“Discovery completed” tells me what the seller did.“The buyer confirmed the problem, business impact, decision process, and next action” tells me what changed in the opportunity.That distinction is what makes forecasting, coaching, conversion analysis, and AI recommendations useful.<h3>4. Clear Ownership Across the Customer Lifecycle</h3>The organization must be clear about:<ul><li>Which segments it is pursuing.</li><li>Which motion applies to each segment.</li><li>Who owns the account at each stage.</li><li>How Marketing, Sales, Partners, Delivery, and Customer Success work together.</li><li>What information must transfer at each handoff.</li><li>Which decisions can be automated.</li><li>Which decisions require human judgment.</li><li></li></ul>Without clear ownership, AI does not create efficiency; it creates duplicate work, conflicting communication, and faster confusion.<h3>5. An Operating Cadence Tied to Outcomes</h3>AI cannot become another side project owned by a handful of enthusiastic users. It has to become part of how the business operates.Leadership needs to define:<ul><li>Which business constraint AI is intended to improve.</li><li>Which workflow will change.</li><li>Which data must be consistently captured.</li><li>Which behaviors are expected.</li><li>How the benefit will be measured.</li><li>Who owns adoption.</li><li>Who owns ongoing improvement.</li><li></li></ul>The tool should follow the diagnosis, the workflow should follow the desired outcome, and the investment should be evaluated based on business impact—not usage statistics.<h2>Start With the Constraint, Not the Demo</h2>One of the most expensive mistakes a company can make is starting with an impressive AI demonstration and then looking for a business problem to justify buying it.The order should be reversed.<ol><li>Start with the constraint.</li><li>Define the desired outcome.</li><li>Understand the current workflow.</li><li>Fix the structural issues.</li><li></li><li>Then determine where AI can improve speed, quality, consistency, or economics.</li></ol>A company struggling to generate pipeline may not need an AI outbound platform. It may need a narrower market, stronger differentiation, better proof, or a smarter channel strategy.A company missing its forecast may not need predictive forecasting software. It may need clearer stage criteria, better deal inspection, and the discipline to remove bad opportunities.A company experiencing churn may not need a sophisticated customer-health algorithm. It may need a better implementation process, clearer customer-success criteria, and stronger accountability for adoption and outcomes.The tool should solve a business problem. It should not become another layer of software attempting to cover up a leadership or management problem.<h2>Build the System. Then Scale It.</h2>The companies creating the most value from AI will not necessarily be the ones that bought the most tools or adopted them first. They will be the ones that redesigned how work gets done.They will: <ul><li>Focus AI on high-value workflows instead of scattering it across dozens of experiments.</li><li>Combine clean data, clear ownership, measurable outcomes, and disciplined execution.</li><li>Reinvest the productivity they create.</li><li>Understand where automation adds value and where human judgment still matters.</li><li></li></ul>Gartner found that sales organizations providing sellers with AI-enabled next-best actions were 2.6 times more likely to achieve commercial growth. But the lesson is not simply that every company should install a next-best-action tool. The research emphasized redesigning seller workflows so that AI recommendations become part of how sellers operate.That is the difference between adding AI and becoming AI-enabled.The flywheel does not begin with the technology.It begins with:A clear market. A relevant value proposition. A disciplined process. Clear ownership. Clean data. Consistent execution.Then AI can make the system faster, smarter, and more scalable.<h2>The Bottom Line</h2>AI may be the most powerful productivity and growth lever available to B2B revenue organizations today, but a lever only works when it has something solid to push against.Without a clear ICP, meaningful positioning, buyer-based pipeline stages, reliable data, ownership, and operating discipline, AI accelerates inconsistency.With those foundations in place, it can improve prioritization, strengthen seller preparation, identify risk sooner, increase retention, and create significant economic leverage.So I would stop beginning the conversation with:Where can we add AI?I would start with:Where do we already have a sound system that AI can make materially better—and where do we need to rebuild the system first?AI is becoming easier to buy every day. Competitive advantage is still hard to build.The winners will not be the companies producing the most activity, installing the most applications, or talking the loudest about AI. They will be the companies with the clearest strategy, the strongest leadership, the cleanest operating model, and the discipline to turn technology into measurable business value.AI will not create that foundation for them, but once the foundation is there, it can dramatically change what is possible.Build the system. Then scale what works.Sources and Further Reading<ul><li>McKinsey & Company, The State of AI: Global Survey 2025.</li><li>McKinsey & Company, The State of AI 2025: Agents, Innovation, and Transformation.</li><li>Salesforce, State of Sales, Seventh Edition, 2026.</li><li>Gartner, AI-Enabled Next Best Actions and Commercial Growth, May 2026.</li><li>Gartner, AI Saves Sellers Nearly Five Hours Per Week, May 2026.</li><li>Gartner, B2B Buyers Turn to Sales Representatives to Validate AI-Generated Insights, May 2026.</li><li>HubSpot, 2025 State of Sales Report.</li><li>HubSpot, The Age of the AI-Powered Buyer, October 2025.</li><li>HubSpot, 2026 State of Marketing Report.</li></ul>