The HubSpot Prospecting Agent Playbook

The always-on prospecting teammate

HubSpot's Prospecting Agent is the part of Breeze that does the work an SDR does before a conversation ever starts. It watches your target accounts for signs they are ready to buy, finds the right people to talk to, researches them, and drafts personalized outreach, all inside the same Sales workspace your reps already live in. It is currently in Beta, and HubSpot describes it as an always-on AI sales teammate rather than a tool you open and close.

The distinction that matters is where it sits. This is not a chatbot bolted onto your CRM, and it is not a sequence with a new name. It is an agent that carries a multi-step job end to end: monitor, source, research, draft, and, if you let it, send. Because it runs on the records you already own, its research, your reps' edits, and the deal history all live in one system. Nothing gets stitched together across tools, and your prospect data does not leave your stack.

This playbook is the complete picture: what the agent actually does, how it works step by step, every use case it fits, the setup and guardrails that keep it safe, an honest read on pricing and limits, and how INSIDEA deploys it for revenue teams. Every product detail here is checked against HubSpot's own documentation, because the feature is moving quickly and Beta capabilities change.

The problem it removes: the research tax

Ask any SDR where their day goes and the answer is rarely selling. It goes to research. Finding the accounts worth touching, working out who the real buyers are, digging up a reason to reach out that is not generic, and only then writing the email. By the time the message is sent, most of the hour is gone, and the next contact starts the cycle again.

That research tax is invisible on a dashboard but it is the single biggest drain on prospecting capacity. It is also the reason cold outreach so often reads as cold: when research is expensive, reps skip it and fall back on templates. The Prospecting Agent is built to remove exactly this tax. It does the finding, the researching, and the first draft, so the human time goes to the part that actually needs a human: the conversation.

The point is not to replace the rep. It is to change what the rep spends the day on, from assembling context to using it. One HubSpot customer, Sandler, reported that work which used to take two months to produce a discovery call happened in two weeks with the agent, and that their BDRs moved from token personalization to real personalization because the research time was handed back to them.

How it works, from signal to conversation

The agent runs a loop, not a single action. It is worth understanding the whole loop before you configure any of it, because each setting you touch later maps to one of these steps.

It starts by monitoring your enrolled accounts for buying signals. When a signal fires, it uses connected data providers to surface the right contacts, the buying group, at that company. It researches each contact against the account context and recent engagement, then drafts a personalized email in your voice. From there you have a choice: the agent can hold the draft for a rep to review and send, or send automatically under rules you set. It keeps working a contact, spacing sends and adjusting, until the contact replies, books a meeting, or hits the limit you defined, at which point it unenrolls them and hands the conversation back to a person.

The buying signals it watches

Signals are the trigger for everything the agent does. In the current buying-signals capability, it looks outward at real-world events that suggest an account is entering a buying window, and inward at how contacts have engaged with you. Reaching out on a signal is the difference between interrupting someone and arriving at the right moment.

The inward signals are easy to underrate. The agent considers recent engagement associated with a contact record from the past year, including form submissions, page views, calls, meetings, notes, and email opens. That means a marketing-sourced contact who quietly returned to your pricing page is a stronger prospect than a cold name, and the agent knows the difference. This is the payoff of a CRM-native agent: it reads the engagement your marketing already earned.

Contact sourcing and the buying group

Knowing an account is ready is only half the job. You still have to reach the right people, and in most B2B deals that is a group, not a person. The agent's contact sourcing is built for that. When a company signals intent, the agent uses connected data providers, including ZoomInfo, Apollo, and Surfe, to find and enrich the contacts who make up the buying group, complete with the role detail you need to write to each one differently.

This is where the agent quietly replaces a stack of manual steps: the list-building, the enrichment, the cross-referencing of who reports to whom. It assembles the committee and hands your rep a set of real, enriched contacts tied to the signal that surfaced them. The rep decides the angle; the agent removes the hours it used to take to get there.

Selling profiles, plays, and guardrails

You do not configure the agent once and set it loose on everyone. You configure it through selling profiles, and this is the single most important thing to get right, because the profile is where your judgment enters the system.

A selling profile carries the value proposition, tone, and positioning for a specific product, audience, or motion. HubSpot lets you create up to 85 of them, which is the mechanism for making outreach feel specific: a profile for your core product to mid-market, a different one for an expansion motion into an existing base, another for a new segment. Breeze can draft the selling information from your website, and you review and refine it. Each profile also sets the identity the agent sends as (the contact owner, or a single sender), the call to action it offers (a meeting link, a document, a URL), and the tone, either a preset or your brand voice.

Guardrails are where you keep control. Per profile you set the autonomy mode, the send window and business-day rules, the minimum days between emails, the maximum emails per enrollment, and exclusion lists of contacts the agent must never touch. Advanced custom instructions let you give the agent specific guidance in plain language, and it will tell you what it will do, what it is unclear about, and what it cannot do before you save.

Semi-autonomous or fully autonomous

The agent runs in one of two modes, and choosing between them is really a question of how much trust it has earned. In semi-autonomous mode, review before sending, the agent drafts every email and waits for a person to approve it. In fully autonomous mode, send automatically, it sends under the rules you set, with no per-email review.

The right path is almost always to start in review mode and graduate. Watch the drafts, correct what the agent gets wrong, and let the quality prove itself on volume you have actually seen. HubSpot's own customers describe this exact ramp: Stax Payments flipped on auto-send only after they saw their BDR needed to edit just 3% of the agent's drafts. That is the signal to widen the line, not a launch-day setting.

Trust the agent the way you trust a new hire. Start it on drafts you review, watch the edits fall, then widen what it can do on its own.

Setup and what you need

Getting the agent live is not a heavy lift, but it has real prerequisites. It is available on Sales Hub Starter, Professional, and Enterprise, and it runs on HubSpot Credits, which are consumed when a lead is enrolled. If you are setting it up for the first time, HubSpot offers a 14-day free-access window so you can run it without drawing down credits while you learn it.

One practical note that saves teams a bad first week: the agent only supports your default brand voice, not additional voices from the Brands add-on, and credit-consuming actions cannot run in a sandbox. Plan to build and test your selling profiles in a real account, with a tight exclusion list, during the free window.

Enrollment, limits, and how contacts move

You decide which contacts the agent works, and there are four ways in: enroll them manually, set up automated rulesets, drive enrollment from a workflow, or enroll the buying group from a target account. Manual enrollment handles ten contacts at a time and the agent researches ten per minute; the rest queue. Automated rulesets enroll contacts as they meet criteria, with the same research throughput.

The limits are worth knowing before you scale. The agent researches up to 1,000 contacts per account per day and sends up to 1,000 emails per day; beyond that, contacts sit in a queued status until the next day. Automated enrollment is deliberately careful: it will not enroll a contact who is already in a sequence, already in the agent, or associated with an open or closed-won deal, so it does not step on active plays or live deals.

Each enrollment also has an outreach type. Adaptive adjusts timing and content to how the prospect behaves and ends after 30 days by default; scheduled sends on a set cadence. As contacts move, they pass through clear statuses, and they unenroll automatically the moment the work is done.

Deliverability, protected by design

The fastest way to turn an AI prospecting tool into a liability is to let it burn your sending reputation. HubSpot built guardrails into the agent specifically to prevent that, and you should treat them as features, not friction.

The agent caps cold outreach at three emails to a contact within a 90-day period when there is no signal or engagement, and it will not exceed that even if you re-enroll the contact. It prioritizes the first three sends, then monitors weekly and adjusts. It takes a deliberately conservative approach to contacts who have received one-to-one emails from you before without replying. And it checks that an address has not previously bounced before enrolling. You can override some of this with custom instructions, but doing so trades your domain reputation for volume, which is rarely a good trade.

The use-case library

The agent is not a single-purpose tool, and the teams who get the most from it run it across several motions at once. HubSpot's customers describe using it for cold outbound, inbound follow-up, conference campaigns, and re-engagement, and reporting that it held up across all of them. Here is the practical library.

Pricing, without the fog

HubSpot moved the Prospecting Agent to outcome-based pricing, and it is refreshingly simple to reason about: you pay for a lead the agent recommends outreach for. Per HubSpot, that is one dollar for each lead the agent recommends reaching out to. It runs on HubSpot Credits, some subscriptions include a monthly allowance, and you can buy more as you need them.

The practical implication is that cost tracks value rather than seat count. You are not paying a flat monthly fee to monitor a large list; you pay when the agent surfaces a lead worth pursuing. That makes the 14-day free window the right way to start, because it lets you see the shape of your own volume and spend before you commit. Always confirm the current numbers on HubSpot's pricing page, since Beta pricing can change.

Where it falls short

A comprehensive picture has to include the edges, because knowing them is how you deploy well. The agent is strong inside HubSpot and weaker at its boundaries.

None of these are reasons to avoid the agent. They are reasons to build the foundation first, define the ICP and personas clearly, keep the data clean, and start in review mode. Get those right and the limitations rarely bite. Skip them and the agent will faithfully automate the gaps.

How INSIDEA runs it

The agent is powerful, but it rewards setup and punishes shortcuts. The teams who see the reported results are not the ones who turned it on fastest; they are the ones who prepared the ground. This is the sequence we use with clients.

Clean the CRM first, because every draft the agent writes inherits your data quality. Define the ICP and the personas precisely, so the signals and the sourcing point at the right companies and the right roles. Build a selling profile per real motion rather than one generic profile, so outreach actually sounds different for different buyers. Start every motion in review mode, watch the edit rate, and only move to auto-send when the drafts are consistently landing. Govern the exclusion lists and the credit usage from day one, not after the first surprise. And measure the thing that matters, meetings booked and pipeline created, not emails sent.

The agent removes the research tax. Your job is to make sure it is researching the right accounts, in your voice, with a human owning the conversation.

This is exactly the work INSIDEA does every day. As an Elite HubSpot Partner, we have built HubSpot prospecting and RevOps foundations for more than 1,500 businesses across 25+ countries, and the pattern that holds up is always the same: clean foundation, sharp ICP, well-built plays, a clear human line, honest measurement. If you want the agent stood up properly rather than switched on and hoped for, our Prospecting Agent Kickstart gets your Sales Hub foundation set up and the agent activated in days, not weeks. Book it there, or start with a strategy call and we will map where it fits in your motion first.