The Best Sellers in 2026 Will Win on Things That Have Nothing to Do with AI

Wrong question

First month of the year. Sellers keep asking me: "What do I need to know about AI to perform this year?"

Wrong question.

AI can absolutely help. But the best sellers in 2026 will still win on things that have nothing to do with AI. They'll win because they understand their customers deeply, because they know the difference between a job title and a persona, and because they're willing to adapt their workflow instead of clinging to what used to work.

If your sales process is broken, AI will just scale the broken process. If your outreach is generic, AI will generate more generic outreach, faster. If you don't understand your customers' world, agents can't invent that understanding for you.

Three things will matter more than any tool this year.

1. Confidence is a byproduct of prep

Confidence is not a personality trait. It's a byproduct of preparation.

When I was at Trustpilot, I closed nearly 600 deals in two years. Almost one deal a day. People assumed I was a natural on the phone. I wasn't. I was terrified of my debt from a failed e-commerce startup, and I was highly motivated because of Trustpilot's uncapped commission model.

But the real reason I was successful had nothing to do with motivation. Before every single call, I would research the company deeply. I would look at their checkout funnel. I would check their social media presence, their online reputation, how they handled customer reviews. By the time I picked up the phone, I wasn't guessing. I already knew their world. And that research is what made me sound like a peer instead of a seller.

That's confidence. Not some magical charisma. Just the quiet certainty that comes from knowing you've done the work.

Here's how it plays out on a cold call. If you've done enough research to connect what you found to a real pain the company has, you pick up the phone differently. Your first 20 seconds land. You get a longer conversation. A longer conversation means a higher chance of a meeting, assuming you called the right company and the right person.

If you did enough research to connect what you found to a real pain you can solve, you pick up the phone differently.

Yes, agents can do the research for you and make you faster. They can browse websites, read annual reports, check job ads, scan social media, and produce a briefing in minutes that would take you an hour. That's a real time saver.

But here's the thing: if you don't know what you need in order to feel confident, agents can't invent it. You need to know which signals matter for your sale. You need to know which pain points resonate with which persona. You need to understand your customer's world well enough to recognize when a piece of research is gold and when it's noise.

The research is the input. Confidence is the output. And the translation between the two happens in the rep's head, not in the agent.

2. "CMO" is a job title, not a persona

"CMO" is a job title. It tells you almost nothing about what that person actually cares about on a daily basis.

A CMO at a bank does not live in the same KPI world as a CMO in retail. A VP Sales at a 50-person startup doesn't have the same problems as a VP Sales at a 500-person enterprise company. A CRO who was promoted internally last month has completely different priorities than a CRO who was hired to turn around a team.

If you don't understand that, your "AI personalization" is just job title karaoke. You're singing the right words to the wrong melody.

Real persona knowledge means understanding the specific KPIs, frustrations, and strategic priorities that differ by vertical, company size, and seniority. It means knowing that a procurement leader in manufacturing cares about supplier risk and cost reduction, while a procurement leader in tech cares about vendor consolidation and compliance. Same title, completely different conversation.

When I was running consulting engagements, we would build ICP-driven playbooks with buyer personas for every vertical. Retail, healthcare, finance, manufacturing. Even if you're targeting the same job title across all of them, the messaging has to be different because the problems are different.

This is where agents become incredibly powerful. But only after you teach them what good looks like. If you train agents on the difference between a CMO in banking and a CMO in retail, they will produce outreach that reflects that difference every single time. They remember 100% of their training. They never forget. They never get lazy.

But if you skip that step and just point an agent at a job title, you'll get job title karaoke at scale. And that's worse than no personalization at all, because now the prospect knows you have AI and you still didn't bother to understand their world.

3. The real advantage is who can unlearn and relearn

The real advantage in 2026 is not who moves fastest. It's who can unlearn and relearn without ego.

Working with agents is not "another tool." It's a different workflow. The way you plan your week changes. The way you prepare for calls changes. The way you think about your pipeline changes. And the teams that treat this like a digital transformation will outperform the teams that treat it like a software rollout.

I see this constantly. A company buys an AI tool, gives it to the team, and expects results in 30 days. Nobody changes how they work. Nobody trains the agents. Nobody adjusts their process. And then they call it a "failed AI pilot."

It's not a failed AI pilot. It's a failed change management effort.

The best teams I've seen treat AI adoption like onboarding a new team member. They start with the training: what does this agent need to know about our market, our customers, our value proposition? Then they give it real work: research these 50 accounts, qualify this list, write talk tracks for these contacts. Then they review the output, adjust the training, and iterate.

Stop treating AI like software you install and forget. Start treating it like digital colleagues you onboard and manage.

The sellers who adapt their workflow around agents will have an unfair advantage. Not because the AI is smarter, but because the combination of a skilled rep and a well-trained agent produces output that neither can produce alone. The agent handles scale and consistency. The rep handles judgment and relationships. Together, that's a 3 to 5x productivity multiplier.

Start with what makes you win

So if you're thinking about 2026, here's the framework.

First, identify what makes you win without AI. Is it the depth of your research? Your understanding of specific verticals? Your ability to have a real conversation in the first 20 seconds of a cold call? Whatever it is, name it. That's your edge.

Then, document it. Write down what your best reps know about each persona, each vertical, each buying scenario. Turn tribal knowledge into something structured. That becomes your centralized intelligence, the foundation that every agent and every new hire can learn from.

Then, use agents to scale it. Deploy workflows that replicate the research, the qualification, the signal tracking, and the personalization that your top performers already do manually. Not to replace the craft. To remove the prep work that limits how many accounts a great rep can work in a week.

The tool doesn't matter. The question isn't which AI to use. The question is: are you confident on the phone? Do you understand your personas deeply enough to have a real conversation? And are you willing to change how you work?

Because sales is hard. Period. And no tool is going to change that.