Skip to main content

Salesforce Piper and Hunter: The New AI SDR Agents Explained

ยท 9 min read
Sunder Iyer
Founder, marketbetter.ai
Share this article

Salesforce now employs SDRs with first names. On September 11, 2026 โ€” four days before Dreamforce โ€” Salesforce announced seven "job-ready" Agentforce agents, each with a name and a job description. Two of them sit directly on the B2B pipeline: Piper, the inbound pipeline agent, and Hunter, the outbound sales agent.

This is the moment the AI SDR category stopped being a startup niche. The largest CRM vendor on the planet is now shipping named, pre-built AI sales reps as products. If you run a sales team, here's what Piper and Hunter actually do, what they cost, where they're genuinely strong, and where the fine print bites.

The short versionโ€‹

  • Piper (inbound): engages website visitors and inbound leads, qualifies them against CRM data, and books meetings. GA now. This is the Qualified product Salesforce bought for $1.2 billion, rebadged as a job-ready agent.
  • Hunter (outbound): researches accounts, prioritizes targets, writes outreach, and works a pipeline "over weeks and months." Pilot now, GA November 2026. It's the first agent on Salesforce's new long-horizon runtime.
  • Both require you to live inside the Salesforce (and increasingly Slack) ecosystem. On HubSpot or Pipedrive? These agents don't exist for you.
  • Pricing was conspicuously absent from the announcement. The Agentforce commercial units underneath โ€” Flex Credits, per-conversation pricing, user licensing โ€” are where the real math lives.

Where Piper came from: the $1.2B Qualified dealโ€‹

Piper isn't new. It's the flagship AI SDR from Qualified, the company Salesforce agreed to acquire in December 2025 and closed on April 1, 2026 โ€” for a consideration of $1.2 billion, $1.1 billion of it in cash, per Salesforce's own 10-Q filing.

Qualified built Piper as an anthropomorphized AI agent that greets inbound website visitors over chat, voice, and video, qualifies them using Salesforce CRM data, and books meetings for human reps. We covered the product in depth before the acquisition in our Qualified review and broke down what customers actually paid in our Qualified pricing analysis โ€” list price around $68K/year for 25 users, with most teams negotiating to $40โ€“50K, and total cost of ownership reaching six figures once you add the required Salesforce stack and services.

What the Dreamforce announcement changed: Piper is no longer a bolt-on you buy from a partner. It's a first-party Salesforce product in the Agentforce catalog, generally available today.

What Piper doesโ€‹

  • Engages inbound leads across your website and inboxes โ€” chat first, with voice and video inherited from the Qualified platform
  • Qualifies visitors against live CRM data (firmographics, account history, open opportunities)
  • Books meetings directly onto rep calendars
  • Routes work to other Agentforce agents when a job crosses roles or systems

Salesforce's proof points: its own internal Piper deployment (live since April 2026) reports a 6% conversion rate and 60+ meetings booked per week, and Asana reported 4x conversation volume from website deployments. Those are vendor-published numbers โ€” treat them as a ceiling, not a floor.

One number worth sitting with: Salesforce says a typical Piper deployment takes 45 days. That's forty-five days of configuration, routing rules, and CRM hygiene before the inbound agent earns its keep. Teams evaluating AI SDRs on "time to first meeting" should price that in.

Hunter: Salesforce's first real outbound agentโ€‹

Hunter is the more interesting announcement, because outbound is the harder problem. Inbound agents respond to demand that already exists. Outbound agents have to create it โ€” research an account, decide it's worth pursuing, write something a buyer will actually open, and keep working the account over weeks.

Salesforce's description: Hunter "works a sales pipeline from research to outreach, collaborating with sellers over weeks and months." Concretely, it:

  • Researches and prioritizes accounts
  • Drafts and sends outreach
  • Learns from how your top sellers win
  • Pursues quarter-long goals with persistent memory, rather than resetting every chat session

That last point is the technical story. Hunter is the first agent built on Salesforce's long-horizon runtime โ€” the architecture that keeps a plan running over weeks, preserves memory across sessions, and accepts mid-flight steering from human reps. At Dreamforce, Salesforce said other agents will move to this runtime "over time," but at launch, Hunter is the only one.

Status: pilot now, general availability November 2026. Early proof point: Perk, a pilot customer, reported that 60% of its sales pipeline is now built by Hunter. Again โ€” vendor-selected customer, vendor-published number.

If you want the broader context on how Salesforce's agent stack fits together โ€” the Koa reasoning model, Agentforce Coworker, and the AIforce surfaces โ€” we covered that in Salesforce Koa and AIforce: what sales teams need to know and Claudeforce, Salesforce inside Claude.

The pricing question nobody answered on stageโ€‹

The September announcement contained no pricing for Piper or Hunter. What exists instead is the Agentforce commercial model underneath:

UnitPublished rateWhat it covers
Flex Credits$500 per 100K credits (~$0.10 per action)Metered agent actions
Agentforce conversations~$2 per conversationSession-based pricing for conversational agents like Piper
User licensingVaries by editionSeat-based access for employee-facing agents

The break-even math matters: at $0.10 per action, a conversation that takes more than 20 agent actions costs more under per-action pricing than the $2 conversation rate. Inbound qualification conversations โ€” multi-turn, CRM lookups, calendar checks โ€” chew through actions fast.

And that's before the platform underneath. Piper and Hunter require Salesforce โ€” see our Sales Cloud pricing breakdown for what that base layer runs. The pre-acquisition Qualified pattern (list ~$68K, TCO $95Kโ€“165K+ with stack and services) is the most honest anchor for what a serious Piper deployment will cost. We'd expect Hunter, as the premium long-horizon product, to land above inbound pricing, not below it.

The lock-in fine printโ€‹

Three constraints that didn't make the keynote slides:

1. Salesforce-only, permanently. Qualified always required Salesforce CRM; the acquisition made that dependency permanent and extended it to Hunter. If your CRM is HubSpot, Pipedrive, or Attio, these agents are not an option โ€” full stop. Our Qualified alternatives guide exists precisely for those teams.

2. The roadmap serves Agentforce, not you. Piper's development priorities are now set by Salesforce's platform strategy. Features that made sense for Qualified as an independent company โ€” broader integrations, CRM flexibility โ€” are off the table.

3. Hunter's runtime is a walled garden at launch. The long-horizon runtime is Hunter-exclusive. Custom agents you build on Agentforce don't get persistent multi-week execution yet, and Salesforce gave no delivery date for when they will. You're buying the packaged rep, not the capability.

Named agents are a packaging move โ€” and a smart oneโ€‹

Give the agent a name and a job title, and buyers stop evaluating a platform and start evaluating a hire. "Should we deploy Agentforce with custom topic configuration for outbound prospecting?" is a six-month committee decision. "Should we hire Hunter?" is a line-item.

Every vendor is converging here. OpenAI shipped always-on Dots agents, Gong launched autonomous pipeline agents, and the entire AI SDR tool market is racing from copilots to named digital workers. Salesforce's edge is distribution: 150,000+ customers who can switch Piper on inside a stack they already pay for.

The open question is quality. An inbound agent with a 6% conversion rate is useful. An outbound agent that sprays mediocre outreach at your best accounts under your domain's reputation is a liability. Hunter "learning from top sellers" is the right idea โ€” whether it works outside pilot conditions is a November question.

What sales teams should actually doโ€‹

If you're a Salesforce shop with real inbound volume: Piper is worth a pilot. The Qualified product underneath is mature, and first-party packaging removes integration risk. Budget the 45-day deployment and model costs on conversations, not the sticker.

If you're evaluating Hunter for outbound: wait for GA in November and ask pilot references hard questions โ€” reply rates, meetings sourced, and how much human steering it needed. "60% of pipeline built by Hunter" tells you nothing about pipeline quality.

If you're not on Salesforce โ€” or don't want your outbound brain locked to one vendor: this launch doesn't change your options, it just raises the stakes. The pattern we see across teams comparing these tools (see our MarketBetter vs Qualified comparison) is that the hard part of outbound isn't sending messages โ€” it's knowing who is in-market right now and what to do about them. That's the gap most agent platforms still leave to the human: a dashboard of signals you have to interpret yourself.

MarketBetter takes the opposite approach: it identifies who's actually showing buying intent โ€” down to visitors on your own website โ€” and hands your team a specific playbook of next actions, whether or not Salesforce is your CRM. If you're building the kind of signal-to-meeting motion Hunter is promising, our SDR playbook guide shows the workflow, and you can see the live product in action in 30 minutes.

Book a demo โ†’


Sources: Salesforce Dreamforce 2026 announcements and job-ready agents launch (September 11, 2026); Salesforce FY2026 10-Q (Qualified acquisition consideration); Salesforce and Qualified acquisition announcements; Agentforce published pricing units. Vendor performance claims (Piper 6% conversion, Asana 4x conversations, Perk 60% pipeline) are Salesforce-published figures.

Share this article