Technology is moving quickly. Learn how revenue leaders choose to use it.

Earlier this year, I wrote about five practical ways AI can help revenue teams work smarter. Those included call coaching, account strategy, revenue analysis, CRM discipline, and improving the handoff between sales and customer success.

Since then, revenue leaders have been experimenting with AI agents that research accounts, prepare proposals, manage workflows, analyze pipeline, and perform work that previously required multiple people or systems. They're also confronting new challenges around costs, data quality, governance, employee development, and customer trust.

There's a lot to learn, and the pace of change can feel overwhelming. Every few weeks brings a new model or capability that seems capable of changing how we work.

In McKinsey's August 2026 State of AI survey, 80% of respondents report that AI has improved their individual productivity. However, only 37% attribute any EBITDA impact to their organization's AI use.

People are finding ways to work faster. Yet, it remains to be determined if we can impact the bottom line. Therefore, the question for leaders in 2027 is, how do we move from individual productivity gains to improving how the business operates?

Here are seven things to start with as we head into the new year.

‍

It's fun to build an impressive dashboard, automate prospecting, or create a beautiful presentation in minutes. And it’s easy to think that your team should be doing something similar.

However, before investing time in a task, be sure to ask: What problem are we trying to solve?

Perhaps the team struggles with inconsistent qualification, slow proposal development, incomplete account intelligence, or internal handoffs that leave out important context.

Choose one of those problems and map out how that process works today. Who initiates it? Where does the process stall? Which activities require human judgment?

That exercise will help you distinguish between tasks that need simple automation, work that benefits from AI assistance, and more complex workflows that justify autonomous agents.

Choose workflows according to their commercial impact and look at the whole process before automating any individual steps.

A study by Gartner in May 2026 found that AI tools enable sellers to save an average of 4.8 hours each week, but sales organizations report reinvesting only a small portion of those time savings into more valuable activities.

Think about what that means. A salesperson could save several hours preparing meeting notes, updating the CRM, conducting research, and writing follow-up emails. But if nothing changes about how that person uses the recovered time, the company may see little improvement in revenue.

If the AI provides a salesperson with five hours a week, then what is the best use for those hours? They should go toward more in-depth discovery calls, deeper discussions with decision-makers, strategic account planning, and extra coaching. Did their conversion rates improve? Has their account coverage increased? Was customer retention better?

Time savings are an input. Revenue growth, improved conversion, and stronger retention are the business outcomes we need to deliver.

Much of the AI discussion focuses on how sellers can use it. Revenue leaders also need to understand how AI changes the buying process.

According to Gartner research published in May 2026, 45% of B2B buyers surveyed reported using generative AI during a recent purchase. The research also found that 67% preferred a buying experience without sales representative involvement.

Yet 69% preferred to validate AI-generated insights with sales representatives.

I find this combination fascinating.

Buyers increasingly want to research solutions independently. They can ask AI to compare vendors, summarize product capabilities, investigate alternatives, and generate questions for potential partners.

However, buyers still need help evaluating whether information is accurate, determining how a solution applies to their organization, and navigating complex decisions.

This creates an exciting opportunity for excellent salespeople. A seller who understands the buyer's business, asks thoughtful questions, challenges assumptions, and helps the buyer make an informed decision provides something useful. For sales leaders, coaching must evolve accordingly.

Consider asking your team to discuss these questions during discovery:

1. What research has the buyer already completed?
2. What conclusions has the buyer reached, and what assumptions are behind them?
3. What information would help the buyer feel confident about making a decision?
4. What business outcome would make the investment worthwhile?

Our responsibility is to help the buyer make a better decision.

‍

The rise of AI agents is introducing a new leadership responsibility. Revenue teams are beginning to use agents to research accounts, prepare meeting briefs, and draft outreach. Some leaders are even organizing agents into specialized teams. But before building a team of agents, it's important to understand what each one does.

OpenAI's Practical Guide to Building Agents explains that agents can manage workflow execution, decide on the next step, and use tools to perform actions within defined boundaries.

Picture an agent who is in charge of account research. It could review publicly available information, summarize relevant company announcements, and recommend next steps for the account executive.

Now suppose the same agent is given the authority to alter CRM records, contact customers, change opportunity stages, or make commitments in the name of the business. The risks involved in those actions are quite different.

I deal with agent design in much the same way that I deal with assigning responsibility to an employee: I first determine the desired outcome, then identify the resources at our disposal, explain the kinds of decisions the agent is allowed to make, and also state when somebody will have to look over the work. For higher-risk activities, such as customer communication or contract changes, appropriate human approval is required.

Anthropic's research on building effective agents emphasizes that simple designs perform better than complex systems. You don't need an elaborate agent to begin realizing value.

Give agents defined responsibilities and appropriate authority, but don’t forget to hold a human accountable for the results.

Every revenue leader has experienced the frustration of incomplete CRM data. Contacts are missing important information. Deal stages are outdated. Customer priorities live in emails or meeting recordings. Critical relationship knowledge sits with individual salespeople. AI can help address these challenges, but it can also make poor information easier to distribute.

That point is important. Suppose you give an AI system access to your CRM and consider asking an AI agent to assess pipeline risk. The CRM shows that there is a strong relationship with the primary decision maker, even though that person has recently left the company; the most recent conversation indicates that the buyer is reconsidering the project, yet the opportunity is still in the same stage and has the same close date. Just as a human would, the AI could produce a well-polished analysis based on wrong assumptions.

Therefore, revenue leaders should work with RevOps and their technology teams to identify the information AI workflows need. That may include call transcripts, account history, product information, customer commitments, stakeholder relationships, and previous decisions. Teams should also establish which sources take priority when records conflict, who can access each type of information, and how the system identifies stale data. In its July 2026 guidance on AI agents in sales, Gartner recommends building a centralized context layer that connects company data, systems, and seller judgment.

Now more than ever, a data foundation and clear definitions are key to accurate analysis.

AI has made it easier to produce content at scale. Teams can generate articles, campaign concepts, presentations, advertising copy, and social media posts in minutes. However, producing content quickly does not guarantee it will influence buyers.

A study by the Pew Research Center found that users clicked on a traditional search result in 8% of cases when a Google AI summary was presented, compared with 15% when no such summary appeared. The implications for revenue are enormous. Companies need to understand how AI systems describe their brands, which sources they use, and whether customers receive accurate information about their products and capabilities. In its 2026 guidelines for generative AI search, Google stresses the importance of unique, valuable content, original expertise, and well-established search optimization practices rather than efforts to manipulate AI visibility.

This creates an important opportunity for businesses with proprietary information. Original research, customer interviews, industry data, expert perspectives, and documented case studies can give organizations something valuable to contribute to the market. AI can help analyze those materials and distribute the findings more efficiently. Leaders still need to determine which conclusions are defensible, and what the market can learn from the work.

‍

The more organizations grant AI authority, the more leaders must set expectations for safety, quality, and accountability. For example, an AI agent could prepare a response for a customer, suggest a change to pricing, update an opportunity, or carry out an action across multiple systems. A voluntary framework for organizations to identify and manage risks over the entire AI lifecycle is the Generative AI Risk Management Profile developed by the National Institute of Standards and Technology.

Organizations need to answer fundamental questions: Who is responsible for the agent? What information can it access? What actions can it carry out on its own? What decisions need to be approved? Can you determine what happened if anything goes wrong? What happens if the agent encounters a situation outside the scope of its instructions? Governance should help employees use AI confidently while protecting customers, the organization, and the people responsible for decisions.

‍

AI can support every one of our responsibilities as revenue architects. We decide how to allocate resources for the best company outcomes. We develop people, create systems, understand customers, and make decisions.

AI can assist in detecting patterns we may have overlooked, in organizing complicated information, and in helping employees gain access to expertise which used to take a great deal of time or knowledge. Nevertheless, the quality of the final outcome still relies on the questions we pose, the information we give, the standards we establish, and our readiness to assess the results.

As you plan for 2027, I would encourage you to ask one question of every AI initiative: How will this serve customers more effectively and therefore create measurable business value? If you can answer that question, you have a meaningful place to begin. And if you can't, spend more time understanding the problem before investing in the solution.

The opportunity ahead is substantial. Revenue leaders who combine sound business judgment with a willingness to experiment and learn will be well positioned to make the most of it.

‍

1. McKinsey & Company. (August 2026). The State of AI in 2026: On the Road to ROI.
2. McKinsey & Company. (September 2026). Cutting the Coordination Tax: How Agentic AI Can Reshape Workflows.
3. Gartner. (May 2026). AI Saves Sellers Nearly Five Hours Per Week, Yet 72% of Sales Organizations Fail to Reinvest Time.
4. Gartner. (May 2026). 69% of B2B Buyers Turn to Sales Reps to Validate AI Generated Insights.
5. Gartner. (July 2026). AI Agents and the Future of Sales Productivity.
6. OpenAI. (January 2026). A Practical Guide to Building Agents.
7. OpenAI. Evaluation Best Practices.
8. Anthropic. Building Effective Agents.
9. Pew Research Center. (July 2025). Google Users Are Less Likely to Click on Links When an AI Summary Appears.
10. Google Search Central. (2026). Optimizing Your Website for Generative AI Features.
11. Microsoft. (May 2026). 2026 Work Trend Index: Agents, Human Agency, and the Opportunity for Every Organization.
12. Dell'Acqua, F., et al. (2026). The Cybernetic Teammate: A Field Experiment on Generative AI and Teamwork. Organization Science.
13. National Institute of Standards and Technology. (2024). Generative AI Risk Management Profile.
14. OWASP. (2026). Top 10 for Agentic Applications.