Territories built on who actually buys
Every CRM account scored against the ICP the agents keep current, with the reason for every rank. Win rates you can cut by segment, competitor and persona.
Works in
and any MCP client
Question: Split the strong ICP fits in fintech into never worked and lost 6+ months ago, with who to contact.
Used Calven MCP
Used Calven MCP| Account | Status | Who to contact |
|---|---|---|
| Ledgerline | Never worked | Marcus Lee, Head of Data |
| Parcelwise | Lost in February, on budget | Jonas Albrecht, the champion, still listed |
| Orbit Pay | Never worked | No data lead on record. Start with the CFO |
Sources: CRM accounts, contacts and loss reasons · 412 CRM deals · ICP, last approved Sep 12
What can revenue operations teams use Calven for?
Ranked target lists, territory plans, win-rate cuts and forecast risk reads that start from the approved ICP and why deals are won or lost. Calven is the approved context your AI workflows need to scale across the team and still be trusted.
Use cases
Example prompts for target account ranking
Ask any of these in Claude, ChatGPT or any AI tool connected to Calven.
- Rank our [segment] accounts by ICP fit and give the reason for each rank.
- Split the strong fits in [segment] into never worked, open and lost [6]+ months ago, with who to contact.
- Score this account list against our ICP: [paste the list]
- Check this target list against our ICP and the CRM before we work it: [paste the list]
- Backtest our ICP fit score on last year's closed deals and show whether high scores actually win more.
- Rank my [segment] accounts by expected pipeline: fit times odds of winning times deal size, in a live spreadsheet.
- Have a veteran AE tear apart my top 20 target accounts, then keep, caveat or swap each one.
- Which Tier 1 accounts have no deal recorded?
- What is our win rate in profile versus out of profile this year?
- Why is [account] Tier 2?
Example prompts for pipeline health
Ask any of these in Claude, ChatGPT or any AI tool connected to Calven.
- Give me the pipeline health read for this week's review.
- Match deals closing [this quarter] to the patterns we lost on most.
- List the open deals missing an amount, close date or competitor, by owner.
- Simulate this quarter's pipeline 10,000 times and tell me the odds of hitting [target] and which deals swing it.
- Run a survival analysis on closed deals and flag every open deal that's already outlived its odds of winning.
- Put every open deal on a fit versus momentum two-by-two and give each quadrant its own play.
- How much of the open pipeline is Tier 1?
- Which late-stage deals have only one contact?
- Which open deals face [competitor], and what is our win rate against them?
- Which open deals have been open longer than our average won sales cycle?
Example prompts for win-backs
Ask any of these in Claude, ChatGPT or any AI tool connected to Calven.
- Find the deals we lost in the [last 12 months] whose loss reason no longer applies.
- Rank these win-backs by fit and whether the champion is still there: [paste the deals]
- Write the re-engagement angle for each account on this list: [paste the ranked list]
- Replay last year's lost deals against what we've shipped since and estimate which ones would flip.
- Rank closed-lost accounts by win-back odds, starting from a base rate and updating on each signal, Bayesian style.
- Design my win-back campaign as an experiment with a control group and a sample size that can detect real lift.
- Which deals did we lose on "Missing feature" in the last year, and which feature?
- What did we ship since [date] that answers a recorded loss reason?
- Is the champion from [lost deal] still listed at [account]?
- Which lost deals had a champion who's still listed and a loss reason we've since addressed?
Example prompts for forecast inputs
Ask any of these in Claude, ChatGPT or any AI tool connected to Calven.
- Give me the risk read on deals over [amount] closing [this quarter], for [Monday]'s forecast call.
- What could sink [deal]? Give me the risks and what past wins had that it has.
- List the deals closing [this quarter] that are missing forecast data.
- Forecast each commit deal from its closest closed lookalikes, then compare that outside view with the rep's call.
- Have a sceptical CFO interrogate every deal in this week's commit, then defend, haircut or cut each one.
- Which deals closing this quarter are single-threaded?
- Which committed deals face a competitor we lose to more than we win?
- Is [deal] in profile, and what is the win rate for its tier?
- Which commit deals have a contact marked as Blocker?
- How long do won deals in [segment] usually sit at the negotiation stage?
Example prompts for segment win rates
Ask any of these in Claude, ChatGPT or any AI tool connected to Calven.
- Cut win rates for [last quarter] by segment, competitor and persona for the [board] pack.
- Explain why our win rate against [competitor] in [segment] moved from [48%] to [31%].
- Reconcile this win-rate table with Calven and tell me why the numbers differ: [paste the table]
- Rank segments by win rate without being fooled by tiny samples, using empirical Bayes shrinkage and credible intervals.
- Tell me whether our win rate moved because we got better or just sold to easier segments.
- Stage a debate between doubling down on our best segment and fixing our worst, judged by the CRO.
- What is our competitive win rate this quarter versus last, with n?
- What is our win rate against [competitor] this year?
- Do we win more when a technical buyer is on the deal?
- Which segment's win rate rose while its deal count fell?
Example prompts for lost deal analysis
Ask any of these in Claude, ChatGPT or any AI tool connected to Calven.
- Build the lost deal readout for [last quarter], by reason, competitor and segment.
- Compare the CRM loss reasons for [last quarter] with what the buyers said.
- Turn this loss readout into actions for each team: [paste the readout]
- Build a causal map of why we lose, separating root causes from the symptoms reps put in the CRM.
- Rank the fixes for our top loss drivers by pipeline recovered per dollar, with break-even effectiveness for each.
- Grow a small decision tree from closed deals that predicts losses, and turn it into a rep checklist.
- s words?",
- , "How often does the rep
- Which losses had a champion but no economic buyer on the deal?
- What do buyers say about us in deals we lost at the final stage?
Example prompts for territory planning
Ask any of these in Claude, ChatGPT or any AI tool connected to Calven.
- Score and segment the account pool for [next year] territory planning.
- Show how Tier 1 potential splits across territories: [paste the mapping]
- Explain the [East] territory to its rep: the ten accounts to work first and why.
- Solve territories as an optimisation problem so every rep gets roughly equal expected bookings, not equal account counts.
- Stress my territory plan by shifting ICP weights and win rates, and show which territories break.
- War-game my territory rules against reps chasing commission, and close the loopholes before launch.
- How many Tier 1 accounts are in [region]?
- Which territory has the fewest strong-fit accounts?
- Which accounts in [territory] are disqualified by the ICP?
- Which region has the highest share of Tier 1 accounts with a recent trigger?
Example prompts for ICP refreshes
Ask any of these in Claude, ChatGPT or any AI tool connected to Calven.
- Build the evidence pack for the ICP review of [last quarter].
- Show me why [segment] buys from us, in their own words.
- Check this proposed ICP edit against our win data: [add payments as a priority vertical]
- Fit a regression on closed deals and show where our ICP scorecard weights disagree with what actually predicts wins.
- Run a three-round Delphi panel of sales, PMM, CS and finance on my proposed [ICP change].
- Map push, pull, anxiety and habit for [segment] from buyer quotes, and tell me if they'll actually switch.
- Does the ICP still predict wins? Give me the four KPIs with the prior period.
- Which attribute predicts a win most?
- Which out-of-profile segment wins most often?
- Which out-of-profile wins share an attribute the ICP doesn't mention?
Example prompts for lead routing rules
Ask any of these in Claude, ChatGPT or any AI tool connected to Calven.
- Check our lead scoring against the ICP and what predicts a win: [paste the scoring sheet]
- Write the inbound routing spec by persona, with the titles that map to each.
- Audit the leads we routed in [last quarter] against the ICP and the win predictors.
- Replay my proposed routing rules on last year's leads and show which won deals they'd have misrouted.
- Model the queue my new routing rules create and find the volume where the response SLA breaks.
- Which ICP attributes should carry the most points?
- Which persona converts best, and what titles map to it?
- Does company size predict a win in our data?
- Which job titles show up on won deals but aren't mapped to any persona?
- How many contacts at Tier 1 accounts are still Lead after 90 days?
Example prompts for persona coverage
Ask any of these in Claude, ChatGPT or any AI tool connected to Calven.
- Give me the persona coverage read for deals over [amount] in [last quarter].
- Brief the rep on [deal] on who to add, given these contacts: [paste the contacts]
- Which personas show up on our wins but never reach the pipeline in [last quarter]?
- Estimate the pipeline each missing persona is worth on every open deal, controlling for deal size.
- Let a rep role-play the missing [persona] on their deal, played from our canvas, and score the conversation.
- What is the win rate for multi-threaded versus single-threaded deals?
- Which open deals over [amount] have no economic buyer?
- Which deals have a blocker and no champion?
- Which persona appears on won deals but almost never on lost ones?
- Which open deals have an end user but no buyer persona?
Example prompts for QBR metrics pack
Ask any of these in Claude, ChatGPT or any AI tool connected to Calven.
- Build the GTM metrics pack for [Q3], with the prior quarter and n.
- Explain the five biggest movers this quarter, with the evidence: [paste the movers]
- Write the one-page QBR narrative from this pack and [our bookings numbers].
- Red-team my QBR narrative for cherry-picked windows, missing sample sizes and moves that are only noise.
- Build a driver tree for pipeline won and show which lever explains the change from last quarter.
- s headline KPIs for last quarter, with the prior period?",
- ,
- Which KPI moved against the trend of the last two quarters?
- Which dashboard's sample is too small to report this quarter?
- How did the ICP share of wins change while pipeline won changed?
Example prompts for CRM data quality
Ask any of these in Claude, ChatGPT or any AI tool connected to Calven.
- List this week's CRM gaps by owner, with the analysis each one breaks.
- Check which of last week's CRM gaps closed: [paste last week's list]
- Map each CRM field to the analyses it feeds for [product], with today's gap count.
- Show how wrong our win rates could be because of missing CRM fields, with best and worst case bounds.
- Turn our stage definitions into exit rules a script can run, then show which deals fail which rule.
- How many open deals have no close date?
- Which late-stage deals have no competitor recorded?
- Which lost deals this quarter have no loss reason?
- Which lost deals have no loss reason and no survey response, so we'll never know why?
- Which accounts are missing a size, so they drop out of segment win rates?
Example prompts for win/loss program health
Ask any of these in Claude, ChatGPT or any AI tool connected to Calven.
- Give me the win/loss program health read for [last month].
- List the deals decided in [last month] that have no survey.
- Check how far to trust the numbers in this win/loss readout: [paste the readout]
- Check whether survey respondents look like our decided deals, and reweight the loss readout if they don't.
- Tell me how many responses each segment and competitor cut needs before its findings mean anything, and how long that takes.
- What share of decided deals had a completed survey this quarter?
- What is the survey completion rate, and how did it change?
- Which segment has the lowest survey coverage?
- Do lost deals or won deals answer the survey more often?
- Which competitor's losses have the fewest completed responses?
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.
Calven
Sam Okafor9:12 AM
CalvenAgent9:12 AM
Sources: ICP dashboard · 412 CRM deals · ICP, approved Sep 12
In your AI tool
Add Calven to the AI workflows you already run. Every step reads from the approved version, with sources.
Question: Reconcile the win-rate table in this board deck with Calven and tell me why the numbers differ.
Used Calven MCP
Sources: 412 CRM deals · Competitive dashboard, Q3 · Win/loss dashboard, Q3
What changes when RevOps' AI tools read from Calven?
RevOps teams using AI without a shared source get fast answers from a CRM export that cannot say why deals are won. Connected to Calven, every ranking, win-rate cut and forecast read starts from the same approved ICP and loss evidence.
| Pain point | Without Calven | With Calven |
|---|---|---|
| Account fit | Target lists sorted by headcount and region | Every account scored against the approved ICP, with the reason for its rank |
| Win rates | A pivot on a CRM export that changes every time someone reruns it | Win rates by segment, competitor and persona, with n and the window |
| Loss reasons | The rep's pick-list reason, taken at face value | The buyer's own words, and where they disagree with the CRM |
| Forecast risk | A commit built on rep confidence | Each committed deal checked against past loss patterns, threading and competitor |
| Win-backs | Every closed-lost deal older than six months | The losses whose reason no longer applies, and whether the champion is still there |
| QBR numbers | Rebuilt from exports every quarter, and they never reconcile | One source, the same window, the prior period and the movers explained |
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.
Questions, answered
How does Calven help RevOps teams?
It adds the context the CRM lacks: ICP fit with the reason, segment tiers, loss reasons in the buyer's words, and win rates by competitor and persona with sample sizes. RevOps uses it to plan territories, rank target accounts and find win-backs.
Does Calven write to our CRM?
Calven MCP is read-only, so nothing you ask in an AI tool changes the CRM. The Calven app can write ICP scores and competitive intel back to the records reps open. Routing, forecasts and other CRM edits stay with RevOps.
How does Calven score accounts against the ICP?
The ICP agent learns which traits predict a win from the deals you won and lost: industry, size, tech stack and use case. Every account is scored against that profile with the reason for its rank.
Do revenue operations 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.