Real enterprise deals rarely close with one person on the call. A rep might face a CFO focused on cost, a CTO focused on technical proof, and a champion who just wants the renewal signed, all in the same meeting. Until now, roleplay could only put one of them in the room at a time.
Why this matters
Single-persona roleplay is good at teaching a rep to handle one objection track from one buyer. It doesn't teach the harder skill: reading a room where two or three people disagree with each other, managing competing priorities in real time, and deciding who to address first. QBRs, committee-based buying decisions, and renewal conversations with cross-functional stakeholders all depend on that skill, and it wasn't trainable in a single session before.
How it works
Launched in v3.2, multi-persona agents let an admin build a roleplay with more than one AI participant in the same call, up to 3 personas per scenario. When creating the agent, the admin sets a group name, a roleplay type such as Discovery or QBR, and a scenario that gives the personas shared context. Existing buyer or agent bots are added as personas, at least 2 and no more than 3, and one can optionally be marked primary, for example the main decision-maker.
Instructions to the rep describe the setting so each persona's stance is clear going in, for example who is cost-focused and who wants technical proof.
During the call, the rep sees two or three participant cards and speaks to the group. The rep doesn't control who answers next: the AI decides which persona responds based on what was said and each persona's defined role, so addressing a persona by name is what actually signals who should speak. Personas respond individually and can build on or push back against each other, so the rep has to practice redirecting, summarizing, and addressing objections from more than one direction at once, including resisting the urge to ignore a quiet persona, since an unheard stakeholder derails real deals the same way it derails a roleplay. The session is graded on a single scorecard covering the whole conversation.
Multi-persona agents aren't available inside Call Blitz, Outdoo's rapid-fire live-calling format, so this kind of practice runs through standard roleplay or courses instead.
More on setup and persona behavior is in Create a Multi Persona Roleplay Agent.
What this unlocks
A rep preparing for an enterprise renewal can rehearse the exact combination of stakeholders they expect on the actual call, cost-focused buyer, technical evaluator, internal champion, instead of practicing each objection in isolation.
In financial services, where deals often require sign-off from both a decision-maker and a compliance or risk reviewer, a rep can practice satisfying both audiences in one conversation rather than assuming what works for one will work for the other.
The same setup works outside sales entirely. A candidate preparing for a final-round panel interview can rehearse against a hiring manager, a peer, and a senior leader in the same session, each asking from their own angle and following up independently, instead of walking in having only ever practiced questions one at a time.
The bigger picture
Deals are won or lost in the room, not one conversation at a time. Now practice can reflect that.
See it in Action
Tell us what you're working on, whether that's sales calls, support tickets, or leadership coaching, and we'll show you how it fits.
Talk to Sales