Campaigns in the buyer's words that land
Briefs from the approved story, copy in the words buyers use on calls, and every variant tested on your personas before budget goes behind it.
Works in
and any MCP client
Question: Write LinkedIn ad variants for Heads of Data at mid-market fintechs, in our customers' words, and rank the ones they would click.
Used Calven MCP
Used Calven MCP- Find the broken pipeline before your CFO does.
- Data your board can trust.
- Real-time lineage for every fintech pipeline.
Sources: Head of Data persona review · 11 customer calls · Approved product profile
What can marketing teams use Calven for?
Briefs, emails, ads and landing pages that start from the approved story and the words your buyers use on calls. Calven is the approved context your AI workflows need to scale across the team and still be trusted.
Use cases
Example prompts for campaign briefs
Ask any of these in Claude, ChatGPT or any AI tool connected to Calven.
- Write the campaign brief for [campaign], aimed at [personas] in [segment].
- Find the strongest campaign angle for [persona] in [segment], from last quarter's calls.
- Set the competitive frame for a campaign aimed at [competitor]'s customers.
- Check this agency brief against our approved story: [paste the brief]
- Simulate my [campaign options] and score each on reach, buyer reaction and likely pipeline before I fund one.
- Assume my campaign failed ninety days after launch and tell me the five most likely reasons, with the fix for each.
- Put my campaign in front of a synthetic buying committee and show me where the deal stalls internally.
- What is the top pain for [persona] in [segment] this quarter, with a quote?
- Which objections should a [segment] campaign pre-empt?
- Give me three proof points for [pillar].
Example prompts for email nurtures
Ask any of these in Claude, ChatGPT or any AI tool connected to Calven.
- Map the buyer journey for a nurture aimed at [persona] who [downloaded a guide].
- Design a [5]-email nurture for [persona] in [segment] that ends in [a demo request].
- Write email [3] of the nurture for [persona], pre-empting [objection].
- Fact-check this nurture against the approved product brief: [paste the sequence]
- Read this nurture as [persona] and tell me where they stop reading: [paste the sequence]
- Email [4] has a [1.2%] click rate against a [3%] average. Tell me why and rewrite it.
- Design a nurture A/B test my list can actually power, with variants our persona really splits on.
- Run a survival analysis on my nurture export to find the email where people quietly drop off, and why.
- Score my nurture on push, pull, anxiety and habit, and write the email for the force it ignores.
- What does [persona] care about at the evaluation stage, and what do they object to?
Example prompts for landing pages
Ask any of these in Claude, ChatGPT or any AI tool connected to Calven.
- Draft a landing page for [persona] arriving from [the ad line they clicked].
- List the questions [persona] needs answered before they [request a demo], in order.
- Fact-check this landing page and review it as [persona]: [paste the page]
- This page converts at [1.5%] for [persona] against a [4%] benchmark. Tell me why.
- Run a five-second test on my landing page with synthetic visitors built from our personas.
- Have procurement, a skeptic and [competitor]'s AE tear my landing page apart, then fix what they find.
- Turn why my page doesn't convert into a hypothesis tree and rank the tests by expected lift per week.
- What headline would [persona] stop on, in their words, about [pain]?
- What does [persona] need to believe before requesting a demo?
- Which objection should the FAQ answer first for [persona]?
Example prompts for paid ads
Ask any of these in Claude, ChatGPT or any AI tool connected to Calven.
- Write [LinkedIn] ad variants for [persona] in [segment] and rank the ones they would click.
- Write search ads for buyers searching for [competitor].
- Tell me why [losing headline] lost to [winning headline] with [persona].
- Run a conjoint test with our personas to see whether the pain, the proof or the CTA drives the click.
- What's the most I can pay per click in each segment and still pay back? Build the model.
- Can I call my ad test early? Give me a Bayesian read of which ad is best and what being wrong costs.
- What pain would make [persona] click, in their words?
- Would [persona] trust "[headline]"?
- Which words in our messaging do customers never use?
- Which pain has the most mentions this quarter but no messaging pillar behind it?
Example prompts for ABM plays
Ask any of these in Claude, ChatGPT or any AI tool connected to Calven.
- Build a target list of [50] accounts in [segment] for an ABM play, ranked by fit.
- Write the account brief for [account]: who we know, what they care about, where to start.
- Write messages for each of [personas] in the [segment] buying group.
- Set the angle for accounts that run [competitor] today.
- Re-rank the accounts that engaged with the play: [paste accounts and signals]
- Backtest how I pick ABM accounts on last year's closed deals: which wins would my rule have missed?
- Split my ABM budget across 1:1, 1:few and 1:many by expected value, as a spreadsheet I can tweak.
- Map the buying committee at [account] and plan who to reach next as a decision tree with odds.
- Which Tier 1 accounts in [segment] has nobody worked?
- Who do we know at [account], and what roles are missing?
Example prompts for launch campaigns
Ask any of these in Claude, ChatGPT or any AI tool connected to Calven.
- Define who should hear about [feature] first, by persona and segment.
- Build the [tier 2] launch campaign for [feature] on [launch date] across [channels].
- Write the sales follow-up kit for the [feature] launch: emails, talk track, objections.
- Fact-check every [feature] launch asset at once and review each as [persona].
- War-game how [competitor] answers our launch over three rounds, and pre-build our counter-move.
- Size launch demand bottom up from named accounts and lost deals, and show what the number hinges on.
- Write the scale, fix or kill rules for my launch campaign up front, then update them each week Bayesian-style.
- Which lost deals named [capability] as the gap?
- Which persona should hear about [feature] first?
- What does [feature] not do, per the brief?
Example prompts for webinars
Ask any of these in Claude, ChatGPT or any AI tool connected to Calven.
- Pick a webinar topic [persona] asked about most last quarter and write the abstract.
- Outline the webinar [title] for [persona], with the questions they would ask.
- Write the promotion and follow-up emails for [webinar title], aimed at [persona].
- Run a Delphi panel of our personas to pick the webinar topic they'd really give up an hour for.
- Run a cohort analysis of past webinar attendees to find which topics actually led to deals.
- What topic does [persona] ask about most on calls this quarter?
- s words about [pain].",
- Which high-severity trends for [segment] are still marked new, not acted on?
- What question do buyers ask on calls that our messaging never answers?
- Which customers have Quantified outcome quotes about [topic]?
Example prompts for event messaging
Ask any of these in Claude, ChatGPT or any AI tool connected to Calven.
- Build the messaging kit for [event]: booth lines, talk angle and answers about [competitor].
- Find the Tier 1 accounts on the [event] attendee list and write the outreach.
- Write the [event] follow-ups for [personas], based on [the topics that came up].
- Sponsor, speak or skip [event]? Build an expected-value decision tree from who's attending and what they're worth.
- Role-play six booth visitors built from our personas and score my answers before the event.
- Give me a 30-second pitch for [persona].
- How do we answer "we already use [competitor]" on the floor?
- Which attending companies are Tier 1 accounts?
- Which signals from [competitor] in the last 30 days will come up in booth conversations?
- Which open deals stalled at a stage an in-person meeting could unblock?
Example prompts for social posts
Ask any of these in Claude, ChatGPT or any AI tool connected to Calven.
- Plan next week's posts for [personas] from [the pieces we are promoting].
- Write this week's five posts from the plan, in the words [persona] uses.
- Write the launch post for [feature], aimed at [persona].
- Check this week's posts against approved claims before they go out: [paste the batch]
- Draft a reply to this comment from a [VP of Marketing]: [paste the comment]
- Tag my post export against our pillars and buyer language, then find which tags really drive engagement.
- Design a social post A/B test with a power calculation, so it gives a real answer in a set number of weeks.
- Throw the eight hostile comments my post is likely to draw at me one by one and score each reply.
- Which messaging pillar has had no post this month?
- Rewrite this post in the words [persona] uses: [paste]
Example prompts for SEO briefs
Ask any of these in Claude, ChatGPT or any AI tool connected to Calven.
- Define the reader and the angle for a piece on [topic], targeting [keyword].
- Collect the proof and the phrases customers use about [topic] for [persona].
- Assemble the content brief for [working title], aimed at [persona] at the [consideration] stage.
- Review this draft against the brief and our approved story: [paste the draft]
- Score three candidate angles on buyer evidence, deals touched, timeliness and our right to win, then show which weights would flip the choice.
- Let our most sceptical persona attack the brief before a writer touches it: every claim they'd dismiss, every vendor tell, every missing proof.
- Walk the outline section by section as the persona and predict where they stop reading, before anyone writes a word.
- Give me the five phrases customers use most for [pain], with the quotes.
- Which themes from customer calls have the most mentions this quarter and no content yet?
- Does our product do [capability]? What exactly does the brief say?
Example prompts for lead qualification criteria
Ask any of these in Claude, ChatGPT or any AI tool connected to Calven.
- Draft our lead qualification criteria for [product] from the ICP and the deals we won.
- Show me how last quarter's leads measure against our qualification criteria.
- Review our current lead qualification criteria against what actually wins: [paste the definition]
- Backtest my lead score on last year's closed deals: does it actually separate wins from losses?
- Set the MQL threshold with a queueing model of SDR capacity and the real cost of each wrong call.
- Stage a judged debate between sales and marketing on our MQL definition, argued only from our win data.
- Which attributes predict a win for us, with n?
- Which triggers were present on accounts we won?
- What share of open pipeline is in profile?
- What's our win rate by lead source over [window], with n?
Example prompts for campaign retros
Ask any of these in Claude, ChatGPT or any AI tool connected to Calven.
- Were the leads from [campaign] worth having? Check them against the ICP: [paste the accounts]
- Diagnose why [campaign] performed as it did with [persona], asset by asset.
- Turn the [campaign] retro into the brief for the next campaign aimed at [persona].
- Replay [campaign] as if we'd targeted differently and estimate the pipeline we'd have got instead.
- Run a driver analysis on my campaign's leads to find what really predicted pipeline, not just which channel looked good.
- Why would [persona] ignore "[headline]"?
- Which objection did the campaign fail to pre-empt?
- What should the next [segment] campaign open on?
- Which titles on the [campaign] lead list map to none of our personas?
- What share of [campaign] accounts already had an open deal before the campaign started?
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.
Priya Shah9:12 AM
CalvenAgent9:12 AM
Sources: Northwind dossier · Positioning · ICP
In your AI tool
Add Calven to the AI workflows you already run. Every step reads from the approved version, with sources.
Question: Fact-check this landing page and review it as our Head of Data persona.
Used Calven MCP
Sources: Head of Data persona review · Approved product profile · Kestrel quote
What changes when marketing's AI tools read from Calven?
Marketing teams using AI without a shared source get fast drafts of the wrong story. Connected to Calven, every brief, email, ad and page starts from the same approved go-to-market context.
| Pain point | Without Calven | With Calven |
|---|---|---|
| Prep time | Every brief rebuilt from the last deck, a persona slide and a call with product marketing | The brief drafted in one prompt from the approved ICP, messaging and proof |
| Consistency | Ads, emails, landing pages and social each tell a slightly different story | Every channel pulls the same approved positioning and messaging |
| Buyer language | Copy written in vendor language | Copy in the phrases buyers use on calls, quoted with the source |
| Differentiation | Lines any competitor could run on their own homepage | Lines checked against the approved competitive positioning, so you claim what only you can |
| Accuracy | Claims checked by hand, usually after launch | Every product claim checked against the approved product profile |
| Measurement | Campaign results explained after the fact with a dashboard and opinions | Retros that read results against personas, buyer language and objections, then feed the next brief |
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 demand generation teams?
It gives every campaign the same starting point: the approved ICP, persona pains by stage, the messaging matrix and proof points. Marketers pull them into briefs and copy from their AI tool and test variants on the personas before launch.
Can Calven test ad copy against our buyer personas?
Yes. Paste an ad, email or landing page and Calven reviews it as each persona, built from your real buyers: what lands, what reads as vendor talk, and which line they would not believe.
Does Calven stop unsupported claims in marketing copy?
It flags them. Every product, pricing and competitor claim in a draft is checked against the approved product brief and battle cards, and anything without support is marked before it ships.
Do marketing teams need to be advanced with AI to use Calven?
No. Teams new to AI start with one-line questions in the AI tool they already have, such as "What phrases do customers use for [topic]?" Teams further along save the workflow prompts from the public use-case library as recipes they run every week. Both read from the same approved context.
Do marketing 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.