AI in sales examples are most useful when they show what the AI actually does, what the rep still owns, and how the workflow gets built. According to HubSpot’s 2024 State of AI in Business and Sales, 64% of sales professionals save one to five hours per week using AI to automate manual tasks — but the value depends on how the workflow is structured, not on the model alone.
The workflows below require sales reps to make final approval decisions on AI-generated outputs before they affect customer interactions. Every example keeps the rep in the loop for approval and is honest about privacy, consent, permissions, and governance. Where HubSpot products are referenced, prebuilt agents are distinguished from custom agent builder workflows, and beta capabilities are labeled as such.
This guide covers six examples — prospecting, qualification, outreach personalization, meeting prep and call summaries, post-meeting follow-up, and pipeline review — each described through the same lens: input, AI action, rep action, approval point, setup, and outcome.
What is AI in Sales?
AI in sales — sometimes referred to as artificial intelligence in sales in more formal contexts — is the use of machine learning, natural language processing, and agentic systems to automate or assist tasks across the sales funnel.
AI in sales reduces repetitive work across the sales funnel — research, drafting, note-taking, CRM logging, and prioritization — enabling reps to spend more time on conversations and less on administrative overhead. HubSpot offers three AI capabilities for sales: Breeze Assistant answers questions within the workflow, prebuilt agents (like prospecting agent and deal agent) handle specific sales tasks, and custom agents can be built in agent builder to fit the team’s process.
The practical question of how to use AI in sales for a growing team usually comes down to two decisions: which workflows to automate first, and how to keep the rep in the loop on every output that touches a customer.
Examples matter more than category definitions for sales teams making adoption decisions. Broad overviews of AI in sales explain the landscape; this article works the other direction, showing what six specific workflows look like when built and run. Each example is described using the same six-part lens — input/trigger, AI action, rep action, approval point, setup, and outcome — so a reader can lift the pattern directly into their own team.
What AI Can and Cannot Do in a Sales Workflow
AI is well-suited to bounded, repeatable tasks with clean inputs:
- Research summaries
- First drafts of outreach
- Meeting transcription
- Deal-level pattern surfacing
- Structured CRM updates from unstructured conversation
In these areas, the AI produces a first pass that a rep reviews, adjusts, and approves.
AI is not well-suited to judgment calls that depend on relationships, context outside the CRM, or ethical trade-offs. Deciding whether to walk away from a deal, when to escalate to a customer’s leadership, how to handle a sensitive negotiation moment, or whether a prospect’s needs actually fit the product remain human calls. Every workflow in this article keeps those decisions with the rep and reserves AI for the surrounding preparation, drafting, and documentation.
Consent, privacy, and permissions apply throughout. Recording a call requires the participants’ consent under the applicable jurisdiction. AI features that process customer data must comply with the data-handling agreements the customer signed. Access to AI capabilities depends on the account’s seat, tier, plan, and credit configuration. These constraints are noted alongside each product reference in the sections that follow.
| Input | AI Action | Rep Action | Approval Point | Outcome | |
|---|---|---|---|---|---|
| Prospecting | ICP + target account list | Research accounts, source contacts, draft outreach | Review research, edit and send drafts | Rep approves each message before sending | Qualified conversations booked |
| Lead Qualification | New lead + CRM engagement | Score fit + engagement, route to owner | Review score, confirm or override routing | Rep confirms assignment before handoff | Right leads reach the right rep faster |
AI in Sales Examples for Prospecting and Research
Prospecting is the most time-intensive pre-conversation task in a sales workflow. A rep building a target account list manually — researching the company, identifying the right contact, finding a credible opening — can spend two to four hours before a single message is sent. AI compresses that preparation into minutes by handling the structured research and first-draft work while the rep focuses on what requires judgment: deciding which accounts belong on the list and whether the draft is worth sending.
How the Workflow Works
Input/Trigger: The rep defines an ideal customer profile (ICP) — industry, company size, revenue range, tech stack, geography, or other firmographic signals — and provides a seed list of target accounts or lets the agent source accounts against the ICP criteria.
AI Action: The prospecting agent researches each account using available data, identifies likely contacts by role and seniority, pulls relevant signals (recent funding, hiring patterns, product announcements, or news), and drafts an outreach sequence anchored to those signals.
Rep Action: The rep reviews the research summary for each account, checks the contact selections, reads the drafted messages, and edits for tone, accuracy, and relevance before approving.
Approval Point: No message is sent until the rep explicitly approves it. The agent queues drafts for review; the rep decides whether to send, revise, or discard each one.
Setup: In HubSpot, the prospecting agent is a prebuilt agent available on Sales Hub Professional and Enterprise. The rep configures the ICP by setting enrollment criteria in the agent settings. Contact sourcing from third-party data requires a connected data provider and compliance with that provider’s terms of use.
Outcome: Reps enter conversations with account context already assembled, and outreach reflects specific signals rather than generic templates, which improves reply rates on initial outreach.
AI in Sales Examples for Lead Qualification and Routing
Inbound lead volume creates a prioritization problem: not all leads are equal, and the time a rep spends on a low-fit lead is time not spent on a high-fit one. AI qualification uses fit signals (firmographic match to the ICP) and engagement signals (pages visited, content downloaded, emails opened, form fields submitted) to produce a score that helps reps and managers prioritize and route leads without manually reviewing every record.
How the Workflow Works
Input/Trigger: A new lead enters the CRM — through a form submission, an event scan, an inbound call log, or an import — and triggers the qualification workflow.
AI Action: The AI scores the lead on two dimensions: fit (how closely the lead’s company and role match the ICP) and engagement (how much and how recently the lead has interacted with the company’s content and communications). It then recommends a routing assignment based on territory, segment, or rep availability rules.
Rep Action: The assigned rep reviews the score and the signals that produced it, confirms the routing is correct, and decides whether to prioritize the lead for immediate outreach, add it to a nurture sequence, or disqualify it.
Approval Point: Routing is a recommendation, not an automatic assignment. A manager or the receiving rep confirms the assignment before the lead is worked.
Setup: In HubSpot, lead scoring uses the contact score property, configurable in Settings. AI-assisted scoring that weighs engagement and fit signals automatically is available on certain Sales Hub and Marketing Hub tiers. Routing rules are built in workflows. Custom qualification agents can be built in agent builder to apply more complex scoring logic.
Outcome: Reps spend their outreach time on leads most likely to convert, and high-fit inbound leads receive faster follow-up — both of which affect pipeline conversion rates.
AI in Sales Examples for Outreach Personalization
Generic outreach — a template with the first name swapped in — produces low reply rates because it signals to the prospect that the sender did not invest time in them. Personalized outreach performs better, but manual personalization at scale is not feasible. AI bridges the gap by generating drafts that incorporate account-specific signals, leaving the rep to review and refine rather than draft from scratch.
How the Workflow Works
Input/Trigger: The rep selects a contact or account for outreach. The trigger can be manual (rep initiates) or automated (a lead score threshold is crossed, or a deal stage changes).
AI Action: The AI pulls available signals — CRM history, recent news, the prospect’s role and seniority, the company’s industry and size, any prior interactions — and generates a draft message that opens with a relevant observation and ties it to a specific value proposition.
Rep Action: The rep reads the draft, verifies that the opening signal is accurate and appropriate, adjusts the tone or phrasing to match their voice, and either sends or schedules the message.
Approval Point: The rep sends every message. The AI produces a draft; the rep decides whether it is ready.
Setup: In HubSpot, Breeze Assistant can generate email drafts from within the contact or deal record. The rep highlights the context they want the AI to use or lets the assistant pull from the record automatically. Sequence emails can also be drafted with AI assistance from within the Sequences tool. Personalization that draws on third-party data (news, LinkedIn activity) requires a connected enrichment source.
Outcome: Reps send more personalized messages in less time, and outreach quality is consistent across the team rather than dependent on individual rep skill at drafting.
AI in Sales Examples for Meeting Preparation and Call Summaries
Reps who walk into discovery or demo calls without preparation lose time re-establishing context that is already in the CRM. After calls, manual note-taking competes with active listening during the conversation. AI solves both problems: it assembles pre-call briefs from existing data and transcribes and summarizes calls so the rep can focus on the conversation rather than documentation.
How the Workflow Works
Input/Trigger: A meeting is scheduled and appears on the rep’s calendar. After the meeting, the call recording triggers the post-call summary workflow.
AI Action (Pre-call): The AI pulls the contact’s CRM history, open deal stage, recent activity, and any available company news, and assembles a brief covering: who the rep is meeting, what the company does, where the deal stands, what was discussed last, and what open questions remain.
AI Action (Post-call): The AI transcribes the call, identifies key topics, extracts action items and next steps, notes any objections or questions raised by the prospect, and generates a summary structured for CRM logging.
Rep Action: Before the call, the rep reviews the brief and adds anything the AI missed. After the call, the rep reviews the summary, corrects any transcription errors, confirms the action items, and approves the summary for CRM logging.
Approval Point: The rep confirms the post-call summary before it is written to the CRM record.
Setup: In HubSpot, call transcription and summarization are available through the Calls tool on Sales Hub Professional and Enterprise. Call recording requires participant consent under the applicable jurisdiction; HubSpot displays a consent notice at the start of recorded calls made through the HubSpot dialer. Pre-call briefs can be generated using Breeze Assistant within the deal or contact record.
Outcome: Reps enter calls informed and exit calls with documentation already drafted, reducing post-call administrative time and improving CRM data quality.
AI in Sales Examples for Post-Meeting Follow-Up
The window between a call ending and a follow-up email arriving matters. Prospects who receive a timely, specific follow-up — one that reflects what was actually discussed — are more likely to remain engaged. AI accelerates follow-up drafting by generating a message from the call summary while the conversation is still recent.
How the Workflow Works
Input/Trigger: The call summary is approved and logged to the CRM. This triggers the follow-up draft workflow.
AI Action: The AI generates a follow-up email that references the specific topics discussed, confirms the agreed next steps, attaches or links any materials the rep promised to send, and proposes a time for the next meeting if one was discussed.
Rep Action: The rep reads the draft, verifies that the content accurately reflects the call, adds any context the AI missed, and sends or schedules the email.
Approval Point: The rep sends the email. The AI draft is a starting point, not an automatic send.
Setup: In HubSpot, follow-up email drafts can be generated from within the deal or contact record using Breeze Assistant, with the call summary as context. Reps can also trigger a follow-up sequence automatically when a deal moves to a new stage, with AI-assisted drafting applied to each sequence step.
Outcome: Follow-up emails go out faster, reflect the actual conversation, and reduce the chance that a committed next step is forgotten or delayed.
AI in Sales Examples for Pipeline Review and Forecasting
Pipeline reviews consume significant manager and rep time, often on deals where the status is already clear. AI surfaces the deals that need attention — stalled opportunities, misaligned close dates, deals with low engagement from the prospect — so review time focuses on decisions rather than status updates.
How the Workflow Works
Input/Trigger: The manager opens the pipeline view, or a scheduled review meeting triggers an automated pipeline digest.
AI Action: The AI scans the pipeline for signals: deals with no activity in a defined number of days, deals whose close dates have been pushed more than once, deals where the contact’s engagement has dropped, deals whose stage progression is slower than the team average, and deals where key fields (decision maker identified, next step logged) are missing. It surfaces these as a prioritized list with the signal that triggered each flag.
Rep Action: The rep or manager reviews the flagged deals, investigates the signals, and decides whether to intervene — by reaching out to a stalled prospect, adjusting the close date, changing the deal stage, or disqualifying the opportunity.
Approval Point: Every pipeline change — stage update, close date adjustment, disqualification — is made by the rep or manager. The AI flags; the human decides.
Setup: In HubSpot, deal pipeline health signals are available through the deal board and the forecast tool on Sales Hub Professional and Enterprise. The deal agent (a prebuilt agent) can surface deal-level insights and recommend next steps. Custom pipeline review agents can be built in agent builder to apply team-specific criteria for flagging deals at risk.
Outcome: Pipeline reviews become shorter and more decision-focused. Deals at risk are identified earlier, giving reps more time to intervene before a close date passes.
Your 30-Day Pilot Plan
A pilot that tries to automate everything at once produces noise, not signal. The plan below sequences the six workflows across 30 days so each one is validated before the next is added.
Week 1 — Prospecting: Configure the ICP criteria in the prospecting agent. Run the agent against 20 target accounts. Have reps review every draft before sending. Track reply rates and note which research signals produced the most accurate drafts.
Week 2 — Qualification and Routing: Enable AI lead scoring on inbound leads. Set routing rules in workflows. Have the receiving rep confirm every assignment. Track whether high-scored leads convert at a higher rate than low-scored ones over the following 30 days.
Week 3 — Outreach Personalization and Meeting Prep: Enable AI email drafting for outreach sequences. Enable pre-call briefs for all scheduled meetings. Have reps rate the quality of each brief and each draft on a simple 1–3 scale. Use the feedback to refine the prompts or context inputs.
Week 4 — Call Summaries, Follow-Up, and Pipeline Review: Enable call transcription and summarization. Enable follow-up email drafting from call summaries. Run one pipeline review using AI-generated deal flags. Collect rep feedback on accuracy and time saved.
At the end of 30 days, review time-saved data, rep feedback, and any measurable outcome changes (reply rates, lead conversion, deal velocity). Use that data to decide which workflows to scale, which to refine, and which to hold.
Frequently Asked Questions About AI in Sales
What is the difference between a prebuilt agent and a custom agent in HubSpot? Prebuilt agents — like the prospecting agent and deal agent — are configured by setting parameters within the agent’s settings panel. They handle defined task types and do not require custom development. Custom agents are built in agent builder, where a user defines the agent’s goal, the tools it can access, and the conditions under which it acts. Custom agents require more setup but can be tailored to workflows that prebuilt agents do not cover.
Does AI send emails automatically, or does the rep always approve? In every workflow described in this article, the rep approves outbound messages before they are sent. AI generates drafts; the rep decides whether to send, edit, or discard them. Fully automated sending — without rep review — is technically possible in some configurations but is not recommended for outreach that touches individual prospects, and it is not how any of the workflows in this article are structured.
How does call recording consent work? Call recording requires the consent of participants under the laws of the applicable jurisdiction. Some jurisdictions require all-party consent; others require only one-party consent. HubSpot’s dialer displays a consent notice at the start of recorded calls. Reps are responsible for understanding the consent requirements in the jurisdictions where they operate. Recording a call without required consent creates legal risk and is not covered by the product’s terms of use.
What data does the AI use to personalize outreach? The AI draws on data available in the CRM — contact properties, company properties, deal history, prior communications, and activity logs — and, where connected, third-party enrichment sources. It does not access data outside the connected sources. The quality of personalization depends directly on the quality and completeness of CRM data.
How do I measure whether AI is saving time? The most direct method is to measure the time reps spend on a task before and after AI is introduced. For drafting, track the time from opening a new email to sending it. For post-call documentation, track the time from call end to CRM update. For pipeline review, track meeting duration. Reps can self-report, or time-tracking tools can be used. HubSpot’s activity reports show task completion rates and volume, which provide indirect evidence of time savings.
Making AI in Sales Examples Repeatable
The six workflows in this article — prospecting, qualification, outreach personalization, meeting prep and call summaries, post-meeting follow-up, and pipeline review — share a common structure: AI handles the bounded, repeatable preparation and documentation work, and the rep makes every decision that touches a customer or affects a deal.
That structure is what makes the workflows repeatable. When the rep’s role at each approval point is clearly defined, the workflow can be refined over time — better ICP criteria, better scoring signals, better prompt inputs — without redesigning the whole process. The 30-day pilot plan sequences the workflows so each one is validated before the next is added, giving teams a practical path from individual examples to a consistent, team-wide AI-assisted sales motion.
