How Salespeople Use AI Every Day: What Top Sales Teams Do Differently

Salesforce's 2026 State of Sales report found that 87% of sales organizations already use some form of AI, yet the average seller still spends only 40% of their time actually selling. Having AI and getting selling time back are two very different things, and the gap between them is where the best sales teams are pulling ahead.

What the data shows

• 87% of sales organizations use AI for tasks like prospecting, forecasting, lead scoring or drafting emails (Salesforce State of Sales 2026, 4,050 sales professionals in 22 countries).
• Sellers expect AI agents to cut prospect research time by 34% and email drafting time by 36% (Salesforce).
• Sellers who effectively partner with AI tools are 3.7 times more likely to meet quota (Gartner survey of 1,026 B2B sellers).
• Teams that use AI deeply generate 77% more revenue per rep (Gong, analysis of 7.1 million sales opportunities).
• Only 8% of sales reps are not using AI at all (HubSpot 2026 State of Sales).

The pattern behind these numbers is consistent. Almost every salesperson now touches AI, but the results depend on which tasks they hand over, how much of the sales process it covers, and whether the company builds habits around it.

Salespeople use AI first for prospect research and call prep

Research is the task most sellers hand to AI first, because it is slow, repetitive and easy to check. Gartner predicts that by 2027, 95% of sellers' research workflows will begin with AI, up from less than 20% in 2024.

The clearest named example comes from Lumen Technologies, the US telecom company. Microsoft reported in July 2024 that Lumen estimates Copilot saves its sellers an average of four hours a week, which Lumen values at roughly $50 million a year. Lumen's Delvin Holman described what changed in account prep: going through a prospect's financial filings used to take "a day or two." Now sellers get through the same material, validate it and share it internally in about an hour before customer calls.

In practice, a rep preparing for a first meeting asks the AI to summarize the prospect's annual report, recent news, open job postings and leadership changes, then turns that into three likely priorities and a short list of questions. The rep still decides what matters. The AI just removes the hours of reading.

Salesforce's data shows why this matters for results. High performers, meaning sellers who substantially grew revenue year on year, are 1.7 times more likely than underperformers to use AI agents to help with prospecting. And 48% of sellers say they lack the bandwidth to do adequate cold outreach, which is exactly the time research automation gives back.

Email drafting and follow ups: faster, but only if they stay personal

Writing is the second big use. Gong's research puts the time reps spend writing emails at an average of 12.1 hours each week, and Gong estimated that generative AI could offer roughly 50% productivity gains on that work, freeing six hours or more per rep each week.

Salesforce uses its own tools on its own sales team, and published the results in May 2025: more than 43,000 personalized emails generated to drive targeted outreach, and more than 800,000 leads and contacts added to sales outreach sequences through AI.

There is a catch. HubSpot's 2026 State of Sales research found 29% of sellers lost prospects because the sales process was not personalized enough. Generic AI emails make that worse, not better. The teams that get value use AI for the first draft and the follow up reminder, then add the detail only a human knows: what the client said last week, the objection they raised, the property or product they actually liked.

HubSpot's data also shows that 74% of sellers see AI making it easier for buyers to research products on their own. Buyers arrive better informed, so a vague email is easier to ignore than ever.

Meeting notes and CRM updates are where the hours come back

The least glamorous use of AI in sales is the one that returns the most time: capturing what happened in a call and getting it into the CRM without a rep typing it.

Bain's Technology Report 2025 estimates that sellers may spend only about 25% of their time actually selling to customers, and argues AI "could double that" by taking on the surrounding work. Much of that surrounding work is notes, updates and internal reporting.

Microsoft published what happened when its own sellers adopted Copilot for Sales. In the first few months, per seller revenue went up 9.4%, opportunities per seller grew 5%, and individual win rates rose 20%. One Microsoft seller put the day to day change simply: "I don't feel like I have to spend 90 percent of my call taking notes."

Salesforce reported similar operational numbers from inside its own sales team: more than 440,000 sales activities logged each month without human intervention, more than 72,000 AI generated summaries to speed up sales planning, and 50,000 hours saved through automated call and conversation summaries.

For a manager, this has a second benefit. When every call is summarized and logged the same way, the CRM becomes something you can trust, and handovers between reps stop losing context.

What top performing sales teams do differently with AI

Access alone does not create results. The research points to five habits that separate the teams getting real gains.

They use AI across the whole sales process, not for one task. Gallup's July 2026 workplace research found that among employees using AI for one or two purposes, 45% report a positive productivity impact. Among those using it for seven or more purposes, that rises to 90%. Bain makes the same point for sales specifically: AI helps teams improve conversion at every step of the selling funnel, and Bain says those gains "add up to more than a 30% increase in win rates."

They tell reps what to do next. Gartner's survey of 227 chief sales officers, released in May 2026, found that sales organizations giving sellers AI generated next best actions are 2.6 times more likely to achieve commercial growth.

They act on the suggestions. Gong's analysis of more than 1 million sales opportunities found average win rates were 50% higher in deals where reps completed all of their AI recommended tasks, and 35% higher when sellers used AI deal guidance.

They train people, not just buy licences. The same Gartner survey found organizations that prioritize upskilling sellers on AI are 2.4 times more likely to achieve strong revenue growth. This matters because Gartner's earlier research found 72% of sellers already feel overwhelmed by the number of skills their job requires, and 50% feel overwhelmed by the amount of technology.

They keep humans on the judgment calls. Gartner's view is that AI is well suited to account research, personalized messaging, signal monitoring and next best actions, while sellers remain differentiated in empathy, judgment, contextual understanding and value framing. HubSpot found 87% of sales managers say in person meetings lead to stronger business relationships. AI buys time for those meetings. It does not replace them.

Gong's December 2025 research sums up the gap: organizations that make AI a core part of how they go to market are 65% more likely to increase win rates than their peers.

What this means for a 25 to 150 person sales team

Most of the named examples above are large companies with dedicated teams. The lessons still apply to a 30 person brokerage or a 100 person distributor, and in some ways they apply more, because each rep carries a bigger share of the revenue.

Three practical starting points:

1. Pick the three most repeated tasks. For most teams that is account research before calls, follow up emails after meetings, and logging notes. Build a shared set of prompts for each so every rep gets the same quality.
2. Give the AI your company's knowledge, not just the internet's. A rep's AI is far more useful when it knows your price lists, past proposals, product details and client history. It also needs to respect who can see what, so a junior rep does not see commission structures or another team's clients.
3. Measure one number before and after. Hours spent on admin, response time to new leads, or meetings booked per week. Salesforce's figures show expectations are high; your own numbers will show whether it is working.

Questions people ask

Will AI replace salespeople?

The current evidence points to AI taking over tasks, not roles. Gartner separates the work AI does well, such as research, drafting and monitoring signals, from the work sellers still own: empathy, judgment and framing value for a specific buyer. Salesforce found 87% of sellers say AI makes their job less stressful.

Which sales tasks should we automate first?

Start with account research, meeting notes and follow up drafts. These are frequent, easy to check, and they return time quickly. Lumen's example, where reviewing a prospect's filings went from a day or two to about an hour, is typical of where the early time savings come from.

How much time does AI actually save a salesperson?

It depends on how widely it is used. Lumen reports an average of four hours a week per seller, and Salesforce's survey shows sellers expect research time to fall 34% and email drafting time 36% with AI agents. Teams using AI for only one or two tasks see much smaller gains than teams using it across the week.

Is it safe for reps to put client information into AI tools?

Not in free personal accounts. Client names, deal terms and pricing should only go into a company controlled business version of ChatGPT, Claude, Copilot or Gemini, set up so data is not used for training and each person only reaches the information they are allowed to see.

CroVIN trains sales teams in person and sets up private company AI so every rep can use it on real client work safely.