Walk into every deal knowing how you win it

Who is on the deal, what sank similar ones, and what to ask first. One question in the AI tool you already use, answered from your own deals and the approved battle cards.

Start free trial

Works inand any MCP client

Question: Build my call brief for Helio Bank at 2:00. Northwind is in the deal.

Used Calven MCP

Biggest risk: hosting. 3 of your last 4 losses to Northwind needed on-premises, and 2 of those were banks. Nobody has asked Helio yet.

Ask first: Tom Whitfield, Director of IT Security, is on the deal but has not been on a call. Ask him: "Where does this have to run, and who signs off on data leaving your environment?"

Your proof: Kestrel Payments picked you over Northwind on time to value: "Live in six weeks, not five months."

Sources: Helio deal record · Northwind battle card · Win/loss, 31 deals vs Northwind · Kestrel interview

Example workspace. Names and numbers are illustrative.

What can sales teams use Calven for?

Call briefs, talk tracks, objection answers and recaps that start from the approved story and the deals you won. Calven is the approved context your AI workflows need to scale across the team and still be trusted.

Use cases

Example prompts for deal prep

Ask any of these in Claude, ChatGPT or any AI tool connected to Calven.

Build my call brief for [deal] at [time]. [Competitor] is in the deal.
Find what decides deals like [deal] against [competitor] in [segment].
Pick the proof point for a [persona] worried about [concern].
Check my call plan for [deal] against what wins: [paste your plan]
Start from our real base rate and update the win odds on [deal] fact by fact, Bayesian style, for the forecast call.
Replay three to five lost deals like [deal] and test which one change would have saved them.
Turn my next call on [deal] into an if-they-say, I-ask decision tree, three levels deep, as a clickable page.
Where does [deal] stand and who is on it?
Why did we lose the last three deals to [competitor]? Quote the buyers.
Which contact roles are missing from [deal] compared with the deals we won?

Open the full prompts on GitHub

Example prompts for objection handling

Ask any of these in Claude, ChatGPT or any AI tool connected to Calven.

The [persona] on [deal] said "[objection]". How did we answer that in deals we won?
Draft a written answer to "[objection]" that [contact] can forward to [stakeholder].
Find the objections from [last quarter] that our talk track does not answer.
Check my answer to "[objection]" before I send it: [paste your answer]
Have two answers to "[objection]" debate each other, let the [persona] judge, then write the answer that beats both.
Backtest how our reps answer "[objection]" against deal outcomes and show which approach actually wins.
Build an HTML flashcard drill of real buyer objections that resurfaces the ones I fumble until they stick.
How do we answer "we already use [competitor]"?
Show me a buyer who said "[objection]" and then bought. What changed their mind?
Is "[objection]" usually a brush-off or a real concern in our deals?

Open the full prompts on GitHub

Example prompts for competitive talk tracks

Ask any of these in Claude, ChatGPT or any AI tool connected to Calven.

Build my talk track against [competitor] for [deal]. The buyer liked [their demo].
What has [competitor] changed since [the last time I pitched against them]?
Play a [persona] who prefers [competitor] so I can rehearse.
Check the [competitor] battle card against our calls from [last quarter].
War-game [deal] against [competitor] in four move and counter-move rounds and find the move they can't answer.
Build a payoff matrix for [deal] from our real win rates against [competitor]: fight, reframe or walk away?
Role-play a [persona] leaning to [competitor], three rounds harder each time, and score every answer I give.
Where do we lose to [competitor], honestly?
What trap should I set against [competitor] on a first call with a [persona]?
A buyer says [competitor] has [feature] and we do not. What do we say?

Open the full prompts on GitHub

Example prompts for discovery questions

Ask any of these in Claude, ChatGPT or any AI tool connected to Calven.

Write my discovery questions for a first call with a [persona] at [account].
Give me the two questions that test the risk on [deal] against [competitor].
What did my discovery on [deal] miss? [paste your notes]
How well do we really know the [persona]? Show me the gaps.
Build a discovery guide on push, pull, anxiety and habit to learn whether a [persona] will actually switch.
Rank my discovery questions by how much each answer would move the forecast, and cut the rest to fit the call.
Write a rubric from our best reps' discovery and grade my own call transcripts against it, line by line.
What does a [persona] care about most, and what are they measured on?
What question exposes where [competitor] loses?
How do buyers describe the problem we solve? Give me their phrases.

Open the full prompts on GitHub

Example prompts for follow-up recaps

Ask any of these in Claude, ChatGPT or any AI tool connected to Calven.

Write the recap for my call with [contact] that they can forward: [paste your notes]
Check this recap for [account] before I send it: [paste the recap]
Which proof points do I keep lacking? [paste what buyers asked you to prove]
Build a scoring rubric for a forwardable recap and grade my last five against it, so I see the habit that weakens them.
Map the mutual action plan's critical path and estimate, PERT-style, the real odds we close by the date the buyer gave.
How do customers describe [pain] in their own words?
Do we do [thing they asked about]? Give me the honest answer.
What proof do we have on time to value?
Which phrase do [persona] buyers use for [pain] that our messaging never does?
What's the proof point closest to the metric [account] gave me?

Open the full prompts on GitHub

Example prompts for technical questions

Ask any of these in Claude, ChatGPT or any AI tool connected to Calven.

[Contact] at [account] asked "[question]". Answer from the approved product brief.
Does what I tell buyers about [topic] still hold? [paste your answer]
Find the buyer questions from [last quarter] the product brief can't answer.
Red-team what we've told [deal] as their architect and security lead, and find the answer that would stall it.
Turn the buyer's RFP into a risk-weighted fit spreadsheet against the product brief, with a bid or no-bid call.
Write 20 trick technical questions with golden answers from the product brief to test any draft before it goes out.
Does the product do [capability]? Say no if we do not.
What does the brief say about hosting, data residency and SSO?
Is this a question for the solutions engineer?
Which product feedback category shows up most on deals we lost?

Open the full prompts on GitHub

Example prompts for pricing objections

Ask any of these in Claude, ChatGPT or any AI tool connected to Calven.

The buyer on [deal] says [competitor] is cheaper. Answer with evidence, not a discount.
How often did price really decide our deals in [segment] over [the last two quarters]?
Check my pricing response to [account] before I send it: [paste your response]
Run a Van Westendorp price ladder with the buying committee's personas and find where our price reads as too expensive, and to whom.
Build a give-get matrix with both sides' walk-away options and the zone of agreement before procurement asks for a discount.
Find the break-even: how much does a discount have to lift my win probability before it's worth giving?
How often is price the reason we lose, and how often did it really decide the deal?
Which buyers said we were pricier and bought anyway? What did they say?
What is the approved answer to "it's too expensive"?
On deals lost to price, what did the buyer actually rank first in win/loss?

Open the full prompts on GitHub

Example prompts for account plans

Ask any of these in Claude, ChatGPT or any AI tool connected to Calven.

Build a one-page plan for [account] in [segment] for [Q1].
Which roles are we missing at [account] compared with the deals we win in [segment]?
Challenge my plan for [account] the way my manager would: [paste your plan]
Rank which of these accounts deserve a plan this quarter: [paste the list]
Rank the whitespace plays at [account] by expected value per hour, using our real win rates and deal sizes for each play.
Map the buying group at [account] on a power and interest grid and find the shortest path to the economic buyer.
Find the won accounts most like [account] and replay how we won them as a 90-day path to copy.
Is [account] in our ICP? Which tier, and why?
What deals have we had at [account] and what happened?
Which accounts in my territory have a trigger and no open deal?

Open the full prompts on GitHub

Example prompts for multi-threading

Ask any of these in Claude, ChatGPT or any AI tool connected to Calven.

Show me the threading gap on [deal] and how to close it through [champion].
Find my single-threaded deals in [segment].
Check my opener to the [persona] on [deal]: [paste the message]
Map the buying group as a network and find who really holds influence, not just who's on the org chart.
Backtest how many roles we threaded against how closed deals actually ended, and find the threading threshold that predicts a win.
Build a decision tree for which missing contact to open first, weighing the win-rate gain against the risk of upsetting my champion.
Which personas are on the deals we win in [segment]?
Which role, when missing, costs us deals most often?
Give me a three-line note to a [persona] from their champion's point of view.
Which persona appears on our won deals but almost never on lost ones?

Open the full prompts on GitHub

Example prompts for exec briefings

Ask any of these in Claude, ChatGPT or any AI tool connected to Calven.

Build my briefing for [contact], [title] at [account], and a version my champion can forward.
Play the [CFO] on [deal] so I can rehearse the meeting.
Read my briefing as a [CFO] and tell me what they skip: [paste the briefing]
Put my business case through a Delphi panel of executive personas and let them converge, over rounds, on what would make them sign.
Have the CFO stress-test my business case: which assumption breaks it, and how far can it fall before the deal doesn't pay back?
What does a [executive persona] care about and distrust?
What objections do economic buyers raise that kill our deals?
Where does our win rate drop when the economic buyer is not engaged?
Which outcome do executives in our won-deal quotes cite most?
What does the economic buyer persona's canvas say counts as proof?

Open the full prompts on GitHub

Example prompts for demo storylines

Ask any of these in Claude, ChatGPT or any AI tool connected to Calven.

Build the demo storyline for [deal], with [attendees] in the room and [competitor] in play.
Debrief the [deal] demo against the record: [paste your notes]
Read this demo storyline as a [persona]: [paste the storyline]
Run my demo past a synthetic panel of the buying group and chart each persona's attention beat by beat.
Make the trade-off: which features earn a place in my demo's time, given what each buyer values and what the competitor shows?
What does a [persona] need to see in a demo?
What should I show that [competitor]'s demo cannot?
Which customer quote fits the [capability] moment?
Which capability do buyers in [segment] mention most in won-deal drivers?
What did lost buyers say they wanted to see that we didn't show?

Open the full prompts on GitHub

Example prompts for proposal drafts

Ask any of these in Claude, ChatGPT or any AI tool connected to Calven.

Draft the proposal for [deal] around the problem they stated: [problem].
Fact-check every claim in this proposal: [paste the proposal]
Read this proposal as the [CFO] and tell me what they push back on: [paste]
Have their procurement lead red-team my proposal line by line and tell me what they'll strike, question or use against me.
Build a working ROI spreadsheet for my proposal with low, expected and high ranges the buyer can edit.
Frame my proposal as good-better-best options and test with the personas which one they pick, and whether a decoy shifts them.
What are our unique attributes against [competitor]?
Which quote shows a measurable outcome for [segment]?
Does the brief say we integrate with [system]?
Which proof point do won deals against [competitor] cite most?

Open the full prompts on GitHub

Example prompts for deal reviews

Ask any of these in Claude, ChatGPT or any AI tool connected to Calven.

Build the deal card for [deal] for tomorrow's pipeline review.
Which of my open deals in [segment] look weakest, and why?
Challenge my commit on [deal] for [date] the way my manager would: [paste reasons]
Score my past commit calls against what actually closed with a Brier score, and show where my forecast bias lives.
Simulate my quarter a thousand times from my open deals and tell me the real odds of hitting my number.
Grade every MEDDICC letter on my deal by the strength of its evidence, with a confidence score, not a colour.
Do we have an economic buyer on [deal]?
Why do deals like [deal] slip past their close date?
Which of my open deals is out of ICP profile?
Which stage do deals in [segment] most often stall at, and for how long?

Open the full prompts on GitHub

Example prompts for CS handoff

Ask any of these in Claude, ChatGPT or any AI tool connected to Calven.

Write the handoff note for [account] so the CSM never has to re-ask me.
List every promise we made to [account] and check each against the product brief.
Check my handoff note for [account] against the record: [paste your note]
Score the first 90 days at [account] on probability and impact, from what nearly stopped the deal.
Trace every promise made to [account] on calls and paper to the product brief, and flag the ones we can't keep.
Why did [account] buy, in their own words?
What did our reps promise on calls with [account]?
What does [account]'s champion say success looks like?
What did customers in [segment] complain about most during onboarding?
Which competitor did [account] evaluate, and what did they like about them?

Open the full prompts on GitHub

Example prompts for renewal prep

Ask any of these in Claude, ChatGPT or any AI tool connected to Calven.

List the renewals in the next [90] days and the risk signals on each.
Build the renewal brief for [account], renewing on [date], for [sponsor].
[Competitor] is in the [account] renewal. Prepare me for it.
We lost the [account] renewal. Tell me why and what it says about the rest.
Start from your base renewal rate and update [account]'s odds one piece of evidence at a time, so the forecast is defensible.
Model the renewal negotiation as a game: their walk-away, yours, and the price uplift both sides can live with.
Play a price ladder to [account]'s buying group and find the uplift where the economic buyer starts looking elsewhere.
Why did [account] buy from us originally? Quote them.
Has anyone at [account] mentioned [competitor]?
What should I propose as an expansion to [account], based on their asks?

Open the full prompts on GitHub

No new tool to learn

Calven works where your team already does: in Slack, and in Claude, ChatGPT or any AI tool through Calven MCP. Nobody opens another app to get the approved story.

In Slack

Mention @Calven in a thread or send it a DM. It answers from the approved version, with sources.

Maya Chen9:12 AM

The buyer on Corvid Capital says Northwind is cheaper. Answer with evidence, not a discount.

CalvenAgent9:12 AM

Price came up in 7 of your 11 deals against Northwind this year. You won 5 of those 7 without a discount.

What won them: time to value. Kestrel Payments: "Live in six weeks, not five months."

Say: "Northwind costs less to buy. Ask them how long it takes to go live, and what those months cost your team."

Sources: Win/loss, your 11 deals vs Northwind · Kestrel interview · Northwind battle card

Example workspace. Names and numbers are illustrative.

In your AI tool

Add Calven to the AI workflows you already run. Every step reads from the approved version, with sources.

Question: Check this recap for Helio Bank before I send it.

Helio Bank call recapGoogle Doc

Used Calven MCP

Cut: "on-premises on every plan". Your approved product profile has it on Enterprise only.

Use his words: Tom Whitfield said "nothing leaves our environment without my sign-off." Your draft says "flexible deployment".

Add proof: he asked how fast banks go live. Kestrel Payments approved a quote: "Live in six weeks, not five months."

Sources: Approved product profile · Helio call transcript · Kestrel interview

Example workspace. Names and numbers are illustrative.

What changes when sales teams' AI tools read from Calven?

Sales teams using AI without a shared source get confident answers built on old battle cards and guessed product facts. Connected to Calven, every brief, talk track and recap starts from the approved context and the deals you won.

Pain point	Without Calven	With Calven
Prep time	Thirty minutes in the CRM, Slack and old decks before every call	A one-page call brief from the deal record, battle card and win/loss in one prompt
Competitors	A comparison slide the competitor has already answered	Talk tracks from the deals you won against that rival, with the trap and the proof
Objections	Answered from a script, or from what the rep remembers	Answered with what changed the mind of a buyer who said the same thing
Accuracy	Ping product marketing before every technical call, or promise what does not ship	Answers from the approved product profile, including what you don't do
Follow-up	Recaps written from memory that repeat the pitch	Recaps in the buyer's words with the proof they asked for
Deal reviews	Pipeline review as opinion versus opinion	Deal cards with buyer evidence behind every criterion, and the risk named first

Two weeks trial. You'll only need a day.

A day is all it takes to see what the agents can do for you.
You won't go back.

Start free trialNo credit card required

Questions, answered

How does Calven help account executives prepare for a deal?

Ask your AI tool about the deal and Calven returns where it stands, who is on it, what sank similar deals against the same competitor, and the first question to ask, each point linked to the deal, interview or battle card behind it.

Where do the objection responses come from?

From deals you won and the objection handling product marketing approved. Calven quotes the buyers who raised the same objection and how the deal was won, so reps answer with evidence instead of a script.

Can reps see CRM deals through Calven?

Yes, where your workspace allows it. Calven mirrors CRM accounts, contacts and deals, and each rep sees only what their Calven role permits. Admins decide whether pipeline is readable over MCP.

Do sales teams need to log into Calven?

No. Product marketing runs Calven. Everyone else reaches the same approved context from the AI tool they already use over Calven MCP, or in Slack. Professional and Enterprise include unlimited seats, so each person signs in as themselves.

Which AI tools work with Calven?

Any client that supports remote MCP: Claude, ChatGPT, Cursor, VS Code with GitHub Copilot, Claude Code, Codex and others. Paste https://api.calven.ai/mcp, choose OAuth and sign in.