Most sales teams did not lose 2025 because their reps were bad at selling. They lost it because their reps walked into every conversation knowing less than the buyer expected them to know.
This is the story of two reps chasing the same deal in the same territory. One figured out the split between human AI sales skills and the machine. The other is still sending follow-ups into the void.
“AI is not going to replace salespeople. But salespeople that use AI will replace those that don’t.”
Scott Bliss, Sandler
|
TL;DR How do you combine human sales skills with AI? Does using AI make human skills less important? Where does AI fit in the deal cycle? What should only a human do? What is the biggest mistake sales teams make with AI? |
Two reps, one deal
Tyler opens his laptop at 8 a.m. with one deal on his mind. It has been quiet for ten days. He writes a follow-up, keeps it short, references their last call. He hits send and moves on to the next name on his list.
Julie has the same deal on her list. Before she types a word, she checks the signals feed on her account intelligence platform. The feed shows two updates from yesterday: the company’s Q2 earnings came in below expectations, with leadership flagging budget scrutiny heading into Q3, and a new VP of RevOps has joined the company.
She does not send a check-in.
She already knows what those two signals mean together. A Q2 miss with a new VP of RevOps means the deal’s approval path has changed. A VP of RevOps coming in mid-evaluation, with a Finance Business Partner now likely pulled into budget decisions, means two people are now in the room who were not there a week ago. She had not known either existed in this deal.
She runs both new contacts through her people intelligence tool, which builds personality profiles from public data.
The VP of RevOps reads as analytical: needs specifics, wants proof before he will move. The Finance Business Partner reads as outcome-focused: short on time, wants the bottom line in four lines or less.
She drafts a separate email for each, adjusts both before sending, then calls the Sales Ops Manager she actually knows to ask whether the timing has shifted.
He tells her the CFO put a hold on new approvals for the quarter, but the VP of RevOps has been pushing to get a few tools through in Q3 planning. He just received her email. He replies two days later. They get on a call.
Tyler’s reply came back: “Still evaluating, will circle back.” He marks the deal 60-day nurture and moves on.
Julie has a confirmed call, a contact carrying the deal internally, and a read on when the window opens.
One deal. Two reps. The difference was what each of them knew before they picked up the phone.
Why Tyler’s approach stopped working
Tyler is not a bad rep. He works hard, follows up, runs a clean process. The problem is that the market has moved and his preparation has not.
Pavilion’s 2025 GTM Benchmarks put the average win rate at 19%. 78% of sellers missed quota last year. The year before: 69%.
Gong’s 2025 research found that enterprise deals now average 17 stakeholders in the buying committee. One stakeholder in the room gives you a strong shot at yes. Six cuts your odds in half. At 17, you are not competing against another vendor. You are competing against the difficulty of getting 17 people to agree on anything.
Challenger’s research found that nearly 40% of losses end in no decision, not a competitor winning. Buyers are not choosing someone else. The deal just feels too risky to push through internally.
When Julie spotted two new stakeholders in the buying committee, that was not luck. That was account intelligence doing what Tyler had no way of doing manually. Most reps in his position would not even know to look.
The mistake most teams made when AI arrived
When AI tools arrived, most sales teams did what was easiest: they used them to do more of the same. More emails. More sequences. More outreach. Amarpreet Kalkat, founder of Humantic AI, applied a phrase that fits perfectly: the tyranny of more. When you scale activity without changing how you think about the conversation, you scale noise.
Every company has an AI mandate. They are buying random tools. Random tools add extra burden on salespeople because they make salespeople focus on non-sales stuff and create extra friction. Either salespeople do not end up using them, or if they use them, they just do the bare minimum and lean on strategies like spray and pray. Increase activity, but not increase it strategically. Do more but not do better.
Think of it like handing everyone in the stadium the same burger and calling it a personalized dining experience. Nobody asked for this burger. Not everyone wants it. But the budget got used.
It is not always the rep’s fault. Their manager is pushing AI adoption. Their manager’s manager tied it to the quarterly review. The result is more noise at exactly the moment buyers need less of it.
Now look at this from the prospect’s side.
A new VP of Sales joins a mid-market software company. His first week, 47 emails land from sales vendors. He cannot tell them apart: same subject lines, same pain point openers. He deletes most without reading past the first sentence. Tyler’s email is in that pile.
Julie’s is not. She spent fifteen minutes before reaching out: enough to know this VP had just stepped into a new role, was still mapping his team, and had three underperforming accounts he needed to fix fast. Her message referenced one of those accounts specifically.
The VP replied to Julie. Not because she had more time than the other 47 reps. Because she showed up knowing something about him.
Katya Tarapovskaya, an ABM specialist who has built buyer intelligence programs at companies including Snowflake and Twilio, put it plainly: the noise problem has nothing to do with volume. It has to do with whether the person on the other end feels understood.
Where human skills still win
The skills that close enterprise deals are human: reading the room, building trust, navigating a 17-person buying committee, handling a hard objection without losing the relationship.
People intelligence and account intelligence do not replace any of that. What they do is make sure you can bring those skills to every conversation, including the ones where you did not have two weeks to prep.
People intelligence tells you how your contact thinks and what they need to feel confident. Account intelligence tells you what is happening in their business right now. When you have both, you know the person and you know the company. You can sell from their point of view instead of your own.
What AI sales skills actually look like in practice
Julie did not get lucky with that deal. She worked a system. Here is the split.
| Stage | What AI handles | What the rep does |
|---|---|---|
| Account research | Surfaces signals: hiring trends, leadership changes, buyer intent data, financial news, tech stack | Decides which signals matter and what to do next |
| People research | Pulls background, communication style, stated priorities from public data. Surfaces likely stakeholders with contact info. | Builds the relationship and adapts in real time |
| Pre-call preparation | Generates a brief: account context, recent news, stakeholder map, talking points | Reviews, filters, decides what to lead with |
| Outreach and cold emails | Drafts based on research and personality data | Edits for tone, personalizes, makes the call to send |
| Follow-up sequences | Handles timing, consistency, and scheduling | Decides when to step off-script and pick up the phone |
| Call summaries and CRM updates | Auto-summarizes, suggests next steps, logs activity | Reviews accuracy, adds context AI cannot know |
| Account monitoring | Sends ongoing signals: leadership changes, hiring shifts, earnings updates. Vidar also flags when target accounts are actively visiting your website. | Decides when a signal is worth acting on and how |
| Discovery conversations | Surfaces facts about the buyer before the call: background, communication style, what they respond to. Generates discovery questions based on account context and personality type. | Executes discovery: asks the questions, listens, reads what is unsaid in real time |
| Objection handling | Predicts likely objections based on personality type. C-types need data and specificity. D-types need outcomes and control. | Delivers the response, reads the room, adjusts on the fly |
| Multi-stakeholder navigation | Maps the committee: Champions, Skeptics, Neutrals, Blockers. Surfaces who has influence and who has not been reached. | Builds the relationships, navigates the politics, earns the internal champion |
| Negotiation and closing | Tells the rep how this buyer type prefers to be approached: price-sensitive vs. risk-sensitive, needs consensus vs. moves fast | Has the conversation, makes the judgment calls, closes |
| Building long-term trust | Nothing | Fully human. This is what creates retention and expansion. |
Julie worked every row in that table. The two new stakeholders she found were not in anyone’s CRM. She found them because account intelligence flagged an org change. The personality profiles told her how to write to each of them. She still wrote the emails. She still made the call. She just went in knowing things Tyler had no way of knowing.
Why the advantage compounds
While Tyler’s deal sat in nurture, Julie was building a different kind of edge, one that does not reset between deals. Julie was using Humantic AI to get complete buyer intelligence across every deal she worked: people intelligence to understand how each stakeholder communicates and makes decisions, account intelligence to know what the company is actually dealing with right now, and visitor intelligence to see which accounts from her target list were already researching her category before she even reached out.
Every deal she ran with that preparation taught her something new: how this type of buyer responds, which signals matter in which accounts, where committees tend to fall apart. That knowledge compounds in a way that instinct slash generic playbooks never does.
What this looks like in the field
Dave Trapani, a Sandler franchise owner with fifteen years in the field, walked into a deal with the CEO of a technology company knowing two things from the people intelligence report: this buyer needed social proof and was a slow decision maker.
Dave prepared three third-party stories from competitors in that CEO’s space. He had them written down. He used all three. When it came time to schedule the decision meeting, he suggested pushing it out two and a half weeks. The CEO agreed.
Within 24 hours, Dave received an email: the CEO was ready to go and wanted to meet next Tuesday. The meeting moved up two full weeks.
The intel did not close the deal. Dave closed the deal. The intel made sure he walked in as exactly the right kind of seller for that specific buyer.
Scott Bliss, another Sandler franchise owner, used to block 3 hours a week for account research across about 10 midsize accounts a month. With account intelligence, that same research now takes 8 to 10 minutes per account. The hours he gets back go toward conversations, not research tabs.
Katya ran the same experiment at the campaign level. After rebuilding her ABM programs around people intelligence and account intelligence, her sales cycle shortened by 3x, conversion rate improved by 3x, and she now books first meetings at twice to three times the previous rate. “People want to talk to a person, not an AI agent. The advantage of this approach is that buyers want to talk to you because they feel understood.”
The numbers
Reps without these tools are running on experience and gut in a market that has become too complex for that to be enough.
Salesforce’s State of Sales research found that reps using AI tools are significantly more likely to hit quota than those who do not. Gartner’s research on buyer expectations points in the same direction: sellers who show up prepared consistently outperform those who rely on volume.
Teams using Humantic AI across buying committees have seen a 109% increase in pipeline, a 16.2% lift in closed revenue, and 36.5% faster deal velocity.
Tyler and Julie will both close deals this quarter. But Julie will close more of them. And because she walks into every conversation with a clear read on what the buyer actually needs, her clients get the right guidance, not just the right product. That is what drives the retention gap as much as the revenue gap.
Julie was promoted to sales manager because she was thoughtful about how she worked. Not because she worked more.
FAQs
How do you combine human sales skills with AI?
Split the job by stage. AI handles account research, people research, pre-call briefs, outreach drafts, follow-up sequences, and CRM updates. The rep handles discovery, rapport, objection handling, committee navigation, and closing. The mistake most teams make is using AI to spray and pray when the real leverage is showing up better to fewer, more targeted conversations.
Which human sales skills does AI make more valuable?
Executing discovery. Building genuine trust. Navigating the politics inside a buying committee. These are the skills AI helps you prepare for, but cannot replace. AI can tell you a buyer is a C-type who needs specificity and data. It cannot sit in the room and adjust when their tone shifts mid-call. Challenger research shows 53% of B2B customer loyalty comes from the sales experience itself, not the product or the company. AI is the preparation engine. The human is the execution engine.
What should AI handle in the sales process vs. what should the rep handle?
AI handles anything involving processing data and generating first drafts: account signals, stakeholder identification, pre-call context, outreach copy, follow-up timing, call summaries. The rep handles anything requiring real-time judgment: discovery conversations, objection handling, negotiation, and trust. The split is clean: if it happens before or after the conversation, AI can own it. If it happens in the room, the rep owns it.
How do top sales reps use AI without losing what makes them effective?
They use it to prepare, then they put it down. They walk into every call knowing the account, understanding each stakeholder, and with a clear read on what the buyer needs to hear. Then they sell. Reps who lose effectiveness are the ones who let AI do the talking: sending unedited drafts, following a sequence instead of their instincts. The tools are there to sharpen judgment, not replace it.
Julie already knows this. She does not outsource the conversation. She uses intelligence to show up ready for it.
Tyler has the same product and the same 24 hours in a day. The question is whether he figures out what Julie already knows before the gap between them gets too wide to close.
If your team is still running on volume, the gap is growing. Reps who prepare are also building out of slumps faster, because they are not flying blind when the quarter gets hard.
That 109% pipeline increase and 16.2% revenue lift did not come from doing more. It came from the same reps showing up better prepared to every conversation that already mattered. Find out more about what buyer intelligence actually means, or see how it applies when you need to get a sales performance improvement plan right.
Written by Humantic AI’s marketing team.
