AI in Sales Transformation

The anatomy of a sales call has changed dramatically. Last week, I shadowed some of HubSpot’s top reps and what struck me was how differently the best sellers work today. They’re using AI at every stage: before, during, and after the call. And the results are real. The brain: before the call. AI does the heavy research — scanning 10Ks, news, emails, and past calls to surface the insights that matter most. Tools like Breeze Assistant can prep a full company overview in seconds. According to our State of Sales Report, 74% of sellers say buyers are showing up to calls more informed than ever before. Salespeople need to be just as ready. The heart: during the call. AI notetakers capture everything: next steps, budget mentions, open questions, so reps can focus on listening, not typing or scribbling notes on the side.  Also, AI assistants surface the right case study or testimonial in real time, making every answer sharper and every example more relevant. That means as a sales rep you are more engaged and relevant. The muscle: after the call. AI follows through fast. It drafts personalized follow-up emails in your own voice, outlines next steps, and flags what needs attention. More time with customers and less time writing emails. The result: sellers who prepare better, connect deeper, and close faster. The anatomy of a great sales call used to be manual effort and hustle. Now, it’s human connection powered by intelligence.

The more I spend time building HUMAIN, the more convinced I become that the age of traditional enterprise sales is coming to an end. Relationship selling alone is no longer enough. In the AI era, value realization and solution selling matter far more than simply pushing products, licenses, or features. Most enterprises today are struggling with one fundamental challenge: they know AI is important, but they do not know how to operationalize it or realize measurable business value from it. Many organizations are still trying to apply AI on top of broken workflows, fragmented data, outdated operating models, and heavy bureaucracy. The future sales organization must look very different. The next generation of enterprise sellers must become: - deeply technical, - operationally aware, - capable of workflow redesign, - capable of discovering hidden inefficiencies, - and able to connect AI to real business outcomes. The conversation can no longer start with technology. It must start with: - What business problem are we solving? - What operational friction exists? - What workflow should disappear? - What can become autonomous? - How do we redesign the enterprise around intelligence and AI agents? In many cases, customers themselves may not even fully understand the root cause of their inefficiencies. This is why the future seller is evolving into something entirely different: part technologist, part operator, part strategist, part transformation architect. At HUMAIN, this transformation has honestly been one of the hardest challenges for me personally. Building AI products is difficult. Building AI infrastructure is difficult. But transforming the mindset of enterprise go-to-market teams may be even harder. I spend a surprising amount of time reading messages that come to me on LinkedIn because I am constantly searching for people who think differently: builders, systems thinkers, operators, problem discoverers, AI-native minds, people obsessed with solving hard problems rather than simply closing deals. The future AI field organization will not look like the traditional sales teams of the past. And I believe the companies that figure this out first will define the next decade of enterprise AI.

Last quarter, we spent $1,404,619 on AI tokens - an all-time high - and the ROI wasn’t what we expected… Most of the ROI didn’t come from “flashy AI”, it came from boring AI doing boring work at scale. Here’s where our spend went and what actually moved the needle: 1. Telling reps who to call today (and why) We’re using AI to sift through millions of signals and tell reps who to talk to today and why. The signals that we’ve found matter: Job changes (new decision makers = new opportunities), buying committee changes and intent signals (active web research and pricing page visits). The big ROI driver is helping our customers with daily prioritization so they don’t have to go fishing for actionable info. At ZoomInfo, We’ve seen a 25-33% increase in meeting quality and opp creation when AEs are sourcing using our AI tools. Win rates also jump from 16-20% to 30%. 2. Writing outreach that doesn’t sound automated We’re moving from “20 segments of 1,000” to 20,000 segments of 1. Not “VP IT at enterprise insurance” messaging… but John at State Farm, who we talked to last year, who competes with three of our customers, with context pulled in automatically. Customer ROI here ultimately comes from better response rates and higher close rates by being more relevant. Buyers care when you show you care. 3. Turning sales calls into usable data Every sales call (ours and customers) is recorded using @Chorus and becomes structured data: objection patterns, competitor mentions, deal risk, coaching moments. We’ve found the benefits of this are huge - 25-30% faster ramp time for new reps, and 10-15% larger deal sizes through better discovery and value articulation. The average rep sells more like the best rep. 4. Speeding up low-value engineering work Every engineer at Zoominfo has Intellij and VS Code w/ Cline. AI handles the unglamorous stuff: Boilerplate code, refactors, test coverage. We’ve seen ~25–30% faster execution on these routine tasks, which frees senior engineers to focus on system design and real product innovation. Our biggest lesson so far has been that if your data foundation is garbage, AI just helps you move faster in the wrong direction. You won’t get AI “working” until you have contextual customer/prospect data centralized, and you can actually build on top of it. We’re still early and we’re trying a lot of things but these have been the highest ROI drivers by a mile. If you’re testing AI in your GTM stack, drop a comment with what’s actually working for you - I’m all ears.

MIT just released ten free AI courses for leaders. An hour with them is the difference between approving AI and rubber-stamping it. Most leaders assume AI literacy is an engineering concern. It is not. It is a decision-making one. You do not need to code. You need to understand one layer deeper than the people selling to you. Here is what that one layer changes. Without it, you ask whether something is cutting-edge. With it, you ask what data it was trained on, what it fails at, what it costs to run at scale, and what decisions it actually automates. The first question can be answered by a brochure. The other four cannot be answered by anyone hiding behind language. That is the whole shift. And it takes about an hour per course. Ten MIT courses worth scanning. Not to become technical. To stop being routed around. 1. AI 101. The vendor-proofing starter. https://lnkd.in/gyJz7whc 2. Artificial Intelligence. Separate real from hype. https://lnkd.in/ggneRvcZ 3. Foundation Models and Generative AI. Spot overselling. https://lnkd.in/gNwgbtaE 4. Introduction to Machine Learning. Fund the right bets. https://lnkd.in/gT5HcRs5 5. Understanding the World Through Data. Ask sharper questions. https://lnkd.in/gVdj_EhG 6. Introduction to Deep Learning. Know what actually costs millions. https://lnkd.in/gq8PwrnP 7. ML with Python. See if your team is building or spinning. https://lnkd.in/gUdHfAhx 8. How to AI Almost Anything. Find competitive whitespace. https://lnkd.in/ghfgKgsx 9. Introduction to Algorithms. Defend AI decisions under scrutiny. https://lnkd.in/g2P-3ptd 10. AI in K-12 Education. See transformation before it reaches you. https://lnkd.in/gs9Fesqy Here is the part most miss. The risk of staying non-technical is not only that vendors oversell you. It is that your own organization stops bringing you the real decisions. When a leader cannot evaluate AI, teams learn to route around them. They pre-decide, then present. The signature becomes a formality. That is how an executive becomes a rubber stamp without ever noticing. Literacy is how you stay the one making the call, not the one being managed toward a yes. So pick one course this week. Spend the hour. You are not learning to build. You are learning to stay in the room. 💾 Save this list and start with the first course. ♻️ Repost so a leader in your network stays the decision-maker, not the rubber stamp. 🔔 Follow Gabriel Millien for daily insights on closing the AI Execution Gap

Here are my Top AI Mistakes over the course of my career - and guess what thebtakeawaybis - deploying AI doesn’t guarantee transformation. Sometimes it just guarantees disappointment—faster (if these common pitfalls aren’t avoided). Over the 200+ deployments I’ve done most don’t fail because of bad models. They fail because of invisible landmines—pitfalls that only show up after launch. Here they are 👇 🔹 Strategic Insights Get Lost in Translation Pitfall: AI surfaces insights—but no one trusts them, interprets them, or acts on them. Why: Workforce mistrust OR lack of translators who can bridge business and technical understanding. 🔹 Productivity Gets Slower, Not Faster Pitfall: AI adds steps, friction, and tool-switching to workflows. Why: You automated a task without redesigning the process. 🔹 Forecasting Goes From Bad → Biased Pitfall: AI models project confidently on flawed data. Why: Lack of historical labeling, bad quality, and no human feedback loop. 🔹 The Innovation Feels Generic, Not Differentiated Pitfall: You used the same foundation model as your competitor—without any fine-tuning. Why: Prompting ≠ Strategy. Models ≠ Moats. IP-driven data creates differentiation - this is why data security is so important, so you can use the important data. 🔹 Decision-Making Slows Down Pitfall: Endless validation loops between AI output and human oversight. Why: No authorization protocols. Everyone waits for consensus. 🔹 Customer Experience Gets Worse Pitfall: AI automates responses but kills nuance and empathy. Why: Too much optimization, not enough orchestration. 👇 Drop your biggest post-deployment pitfall below ( and it’s okay to admit them - promise) #AITransformation #AIDeployment #HumanCenteredAI #DigitalExecution #FutureOfWork #AILeadership #EnterpriseAI

Most AI workflows overpromise & undersell. But one of my favorites has (actually) driven hundreds of thousands in incremental revenue. The CEO of Zapier—who’s the homie—shared it with me, and I’ve been hooked ever since. Think of it as an AI SDR, who qualifies, organizes, and engages sales leads. Here are all of the steps my sales sidekick takes: 1) Extracts the name, email, company, role, and website for any lead that fills out a sales form on our website 2) Researches the lead online to gather the following info: - Company website & recent news - Linkedin profile and background - Company size, industry, and estimated funding/revenue/growth indicators - Specific pain points related to my company’s service 3) Compares lead info against ideal ICP criteria I’ve set: - US-based company - VP-level & up  - Revenue: $10m-$500m annually  - Company size: >50 employees 4) Scores the lead as “Great Fit,” “Possible Fit,” or “Poor Fit” based on ICP comparison 5) Adds a new record to our CRM with the following details: - Contact details (name, email, company, role) - Research findings (company size, revenue, industry) - ICP fit score - Date submitted 6) Conditional logic based on Lead Fit IF lead is “Great Fit” Draft a personalized email in Gmail incorporating: - Their specific company challenges identified in research - Relevant case studies from similar companies - Clear next steps for a discovery call IF lead is “Possible Fit” Send direct message in Slack to me with:  - A summary of lead and research findings - Reasons for uncertainty regarding ICP fit - A recommendation with supporting data - The question: “Should I draft a response email for this lead?” IF response is “yes”: follow great fit action  IF response is “no”: no response Update CRM for this lead based on action taken in Step 6. Let me know if you have any questions—and if you take it for a spin—let me know what you think. #ZapierPartner

𝗦𝗮𝗹𝗲𝘀 𝗰𝘆𝗰𝗹𝗲𝘀 𝗮𝗿𝗲𝗻’𝘁 𝗹𝗼𝗻𝗴. 𝗧𝗵𝗲𝘆’𝗿𝗲 𝗷𝘂𝘀𝘁 𝗹𝗼𝘀𝘁 𝗶𝗻 𝗻𝗼𝗶𝘀𝗲. Many sales leaders believe they can improve performance by adding more training, tools, or dashboards. But the reality doesn’t change. Deals continue to stall. Forecasts remain inaccurate. The problem isn’t a lack of effort or even talent. It’s the absence of a real-time execution layer that turns data into action. 𝟭. 𝗡𝗲𝘅𝘁-𝗦𝘁𝗲𝗽 𝗜𝗻𝘁𝗲𝗹𝗹𝗶𝗴𝗲𝗻𝗰𝗲 Reps no longer guess their next move. AI reads deal patterns and gives them precise, timely actions that move the pipeline forward. 𝟮. 𝗗𝗲𝗮𝗹 𝗥𝗶𝘀𝗸 𝗗𝗲𝘁𝗲𝗰𝘁𝗶𝗼𝗻 𝗮𝗻𝗱 𝗥𝗲𝗰𝗼𝘃𝗲𝗿𝘆 Most lost deals show early warning signs. AI detects when momentum drops and triggers recovery actions before the deal slips away. 𝟯. 𝗦𝗮𝗹𝗲𝘀 𝗖𝗮𝗹𝗹 𝗣𝗿𝗲𝗽 𝘄𝗶𝘁𝗵 𝗖𝗼𝗻𝘁𝗲𝘅𝘁 Reps walk into calls fully prepared. AI surfaces the key insights, talking points, and questions tailored to each persona and stage. 𝟰. 𝗔𝘂𝘁𝗼𝗺𝗮𝘁𝗶𝗰 𝗣𝗶𝗽𝗲𝗹𝗶𝗻𝗲 𝗣𝗿𝗶𝗼𝗿𝗶𝘁𝗶𝘇𝗮𝘁𝗶𝗼𝗻 Not every deal deserves equal attention. AI helps reps focus on the right opportunities at the right time. 𝟱. 𝗜𝗻𝘀𝘁𝗮𝗻𝘁 𝗥𝗲𝗽 𝗖𝗼𝗮𝗰𝗵𝗶𝗻𝗴 𝘄𝗶𝘁𝗵𝗼𝘂𝘁 𝗮 𝗠𝗮𝗻𝗮𝗴𝗲𝗿 Coaching doesn’t have to wait for a review. AI analyzes performance patterns and provides real-time guidance for improvement. 𝟲. 𝗣𝗹𝗮𝘆𝗯𝗼𝗼𝗸𝘀 𝗳𝗿𝗼𝗺 𝗥𝗲𝗮𝗹 𝗪𝗶𝗻𝘀 Winning deals leave a trail of patterns. AI turns those into living playbooks that adapt across industries and personas. 𝟳. 𝗪𝗼𝗿𝗸𝘀 𝗜𝗻𝘀𝗶𝗱𝗲 𝗧𝗼𝗼𝗹𝘀 𝗬𝗼𝘂 𝗔𝗹𝗿𝗲𝗮𝗱𝘆 𝗨𝘀𝗲 Adoption is everything. Modern AI integrates into your team’s daily tools, so insights appear exactly where reps work. 𝗪𝗶𝘁𝗵𝗼𝘂𝘁 𝗔𝗜 𝗮𝗴𝗲𝗻𝘁𝘀, 𝗿𝗲𝗽𝘀 𝗿𝗲𝗹𝘆 𝗼𝗻 𝗴𝘂𝘁 𝗳𝗲𝗲𝗹𝗶𝗻𝗴 𝗮𝗻𝗱 𝗿𝗲𝗮𝗰𝘁𝗶𝘃𝗲 𝗽𝗹𝗮𝘆𝗯𝗼𝗼𝗸𝘀. 𝗪𝗶𝘁𝗵 𝗔𝗜 𝗮𝗴𝗲𝗻𝘁𝘀, 𝗲𝘃𝗲𝗿𝘆 𝗺𝗼𝘃𝗲 𝗶𝘀 𝗴𝗿𝗼𝘂𝗻𝗱𝗲𝗱 𝗶𝗻 𝘀𝗶𝗴𝗻𝗮𝗹 𝗮𝗻𝗱 𝘁𝗶𝗺𝗶𝗻𝗴. 𝗜𝗳 𝘆𝗼𝘂 𝘄𝗮𝗻𝘁 𝗲𝗮𝗿𝗹𝘆 𝗮𝗰𝗰𝗲𝘀𝘀 to your own AI sales agent that does this for your team: 👉 https://tally.so/r/m6BA6P

Today’s revenue teams will look nothing like the best-run revenue teams of the next decade. The CRO role is being redesigned. For decades, revenue leadership meant managing pipelines, arguing over forecast math, judgment calls, and carrying a number into a board meeting. The CRO was the overall quota owner and enforcer. AI agents change that entirely. When agents absorb the invisible work of selling, the Orchestrator role emerges: designing an intelligent revenue system where humans and machines co-own outcomes. In the agentic era, the CRO becomes the orchestrator of the revenue system and owns these 4 roles:  1. Chief Growth Systems Designer 2. Chief Forecast Intelligence Officer 3. Chief Agent Governor 4. Chief Revenue Connector I sat down with Abhijit Mitra, CEO of Outreach, to dig into where AI transformation is heading for CROs and sales teams. His framing was direct: the best-orchestrated revenue system wins. With agents, the shift moves sales from activity-heavy execution to decision-driven selling. The real shift is from point AI solutions to end-to-end revenue orchestration, where AI coordinates inbound, outbound, and deal execution as a unified system. AI restructures today’s B2B sales work around strategy, orchestration, and trust. The meta-pattern: AI handles sense-making and analysis. New roles are emerging: - Sales AI Operator / Sales Ops AI Lead  - Buyer Signal Analyst  - Deal Strategy Orchestrator  - Trust & Compliance Sales Specialist The shift from traditional SaaS sales software to intelligent revenue systems is a big company-building opportunity. Here is the advice I am sharing with founders building in this space: 1. Build for decisions, not activity 2. Design for systems, not features 3. Build for the Revenue Orchestrator and the organization around them. 4. Price to outcomes. 5. Design trust from day one. The winners are not the companies adding AI features to existing workflows. They are the ones reimagining SaaS in the AI era and building an intelligent revenue system that compounds. This is part 2 of my series on the Future of CXOs. Watch the highlights from my conversation with Abhijit Mitra on the future of sales, and read my newsletter on The Future CRO: The Orchestrator. 

You don’t need to be a coder to lead in the age of AI. But you do need to ask the right questions. Many business leaders still treat AI adoption like a technical decision. Choose a model, plug it in, and let the tech team handle the rest. But AI isn’t just a tool. It’s a strategic lever. And leading its adoption means making calls that shape your business, not just your tech stack. That includes one of the biggest decisions you’ll face: Which large language model (LLM) should we integrate? It might sound like a technical question. But it's actually a leadership skill, knowing how to evaluate options based on what your business needs most. And here’s what you really need to consider: → Purpose-fit: Is the model designed for your use case? Some models excel at summarizing text, others at generating visuals or analyzing data. Choose based on the outcome you want. → Integration: How easily will it connect with your existing systems? Adding AI should feel like upgrading the engine, not rebuilding the car. → Output format: Do you need written content, images, or videos? Different models specialize in different outputs, know what matters to your operations. → Data control: Will your data stay in-house, or is it being sent to a third-party server? Open-source tools offer flexibility. Closed systems may provide simplicity, but at the cost of data exposure. → Cost structure: What’s the real investment? Beyond licensing, factor in training time, change management, and long-term scalability. → Training depth: How much data was used to train the model? More data can mean more accuracy, but only if it's relevant to your needs. Great AI choices aren’t about features. They’re about alignment with your goals, workflows, and team capacity. AI is no longer just an IT consideration. It’s a leadership conversation. If you're unsure how to navigate it, let’s chat. I help companies make practical, cost-effective AI choices that lead to real business impact.

Executives: Doomerism, FOMO, or wait-and-see aren’t AI strategies. Hype isn’t an AI product strategy. Businesses must be more constructive and lean into execution, or they’ll be left behind. The technical, business, and market paradigms are clear. It’s time to get off the sidelines: Build, Deliver, and Iteratively Improve. AI is evolving in 2 directions that businesses must operationalize now. Information & Knowledge Management Your AI is only as good as your information architecture. Here’s what knowledge graphs (KG) deliver. From data to decisions: KGs turn scattered data into connected, actionable facts (entities, relationships, lineage, dynamism). AI needs context to reason vs. simply retrieve. Governance: KGs provide provenance, policy enforcement, and auditability, which are critical for compliance, trust, and safe autonomy. Reuse > rebuild: Canonical ontologies break down silos so teams and systems share definitions (customer, product, risk, event, sale), reducing duplicated data engineering work. Lower cost, higher quality: Better grounding = fewer hallucinations, faster time-to-answer, smaller prompts, and lower inference spend. KGs enable smaller models to be used across use cases. Agentic Platforms & Agent-to-Agent Systems AI is moving from copilots to composable agents that plan, call tools, and collaborate with other agents. They collaborate with people to deliver outcomes instead of simply completing tasks. What this unlocks: New product surfaces: Quote-to-cash agents that negotiate terms; post-sale agents that drive expansion; compliance agents that watch activity and file evidence automatically. Operational leverage: Multi-agent workflows that coordinate across CRM, ERP, ticketing, data warehouses, and private APIs, running 24/7, with guardrails and humans-in-the-loop at every critical step in the decision chain. Partner ecosystems: Expose capabilities as tools that other agents can call, creating network effects and new revenue models. A pragmatic build order for leaders: 1️⃣ Start with relationships, not tables: Define your domain ontology (entities, relationships, policies, change functions, outcomes). 2️⃣ Stand up a KG and integrate it with your RAG stack and event bus. 3️⃣ Instrument trust by design: Observability, identity, policy, and human-in-the-loop from day 1. 4️⃣ Ship thin slices: Pick 1 high-ROI workflow (start internal and branch out to customer-facing once capabilities have improved), prove value, then scale. 5️⃣ Treat agents as products: Clear SLAs, sandboxes, evaluation harnesses, and a tool registry. 6️⃣ Executive scorecard to track outcomes: reinvent processes and workflows with an outcomes-centric view. Agents shouldn’t be bolted on, so track outcome metrics to see the bigger picture and adopt the new mindset. The mandate is clear: be constructive about AI. Build resilient information foundations and productize agentic capabilities. That’s how you create a durable advantage today.