Skip to content
JevHub

[ Paid resource · v1.0 · September 2026 ]

Jev Marketing Recipe Pack

Ten ready-to-run decision recipes for marketing and growth teams: the exact Jev questions, the thresholds to act on, and what each one costs to run.

[ What's a recipe? ]

A complete decision setup for one job: the input to send, the exact questions, and the rules that turn Jev's answers into actions. For example: strong fit and high intent → assign to a rep now.

[ Who it's for ]

Teams with someone who can run a script or wire the questions into an existing tool. The recipes themselves are written in plain Markdown, so anyone can review the logic.

What you get

  • 10 recipes in readable Markdown: what each decides, the input to send, and the decision rules
  • Ready-to-send JSON question files for every recipe
  • A TypeScript helper and batch runner: batching, concurrency and retries handled
  • Cost per recipe at 1,000 and 100,000 items
  • A calibration checklist for tuning thresholds on your own data

Not included

  • Jev access or credits. You use your own API key from TypeSafe AI or a gateway
  • A hosted app, dashboard or ongoing service
For developers: the files in the pack
jev-marketing-recipe-pack-v1/
├── README.md
├── CALIBRATION.md
├── recipes/
│   ├── 01-lead-fit-intent.md + .json
│   ├── 02-inbound-email-triage.md + .json
│   ├── 03-search-intent.md + .json
│   ├── 04-ad-angle-hook.md + .json
│   ├── 05-ad-claims-precheck.md + .json
│   ├── 06-content-draft-scoring.md + .json
│   ├── 07-social-post-preflight.md + .json
│   ├── 08-competitor-mentions.md + .json
│   ├── 09-review-testimonial-mining.md + .json
│   └── 10-churn-risk-signals.md + .json
├── jev.ts        batching · retries
├── rules.ts      thresholds as code
├── run.ts        JSONL in → decisions out
└── examples/

The ten recipes

Each recipe decides one thing. Four have free write-ups on this site you can read first.

RecipeDecidesQuestion types
Lead fit & intent scoringFree write-upWhich inbound leads get a sales conversation today, which go to nurture, and which are noise.ScoreNoul (yes/no)Choice
Inbound email triageFree write-upWhich emails need a reply, how soon, and which ones a person must see regardless of category.ChoiceScoreNoul (yes/no)
Search-intent classificationFree write-upWhich keywords deserve a page on your site, and what kind of page.ChoiceScore
Ad angle & hook classificationFree write-upWhich persuasion angles you and your competitors use, and which hooks are strong.ChoiceScoreNoul (yes/no)
Ad claims pre-checkWhich ads need a compliance or legal look before launch.Noul (yes/no)
Content draft scoringWhether a blog post, landing page section or newsletter is ready to publish, and what to fix first.ScoreNoul (yes/no)
Social post pre-flightWhich drafts to post, which to rework, and which to drop.ScoreNoul (yes/no)
Competitor mention monitoringWhich public mentions of your competitors are sales or content opportunities.Noul (yes/no)ScoreChoice
Review & testimonial miningWhich reviews can become testimonials, and which themes customers repeat.ChoiceScoreNoul (yes/no)
Churn-risk signalsWhich customer messages suggest they're about to leave, and why.ScoreNoul (yes/no)Choice

Preview: Lead fit & intent scoring

The decision logic from recipe 01, as it ships. The pack adds the input template, cost table, calibration steps and variations.

Decision rules

  • Sales now: route = sales AND intent ≥ 0.7 AND fit ≥ 2.5 → assign to a rep and alert them.
  • Sales queue: route = sales AND (intent ≥ 0.5 OR fit ≥ 3) → add to the rep's queue for today.
  • Nurture: route = nurture, or route = sales below the thresholds above.
  • Hand off: route = partner → partnerships inbox; route = support → support queue.
  • Drop: route = other with confidence ≥ 0.7.
  • Review: any route with confidence < 0.55 → a human looks at it.

Questions

  • Score

    How well does this lead match our ideal customer? Our ideal customer is: [REPLACE: e.g. B2B software companies with 50–1,000 employees that have a dedicated marketing or RevOps team].

  • Noul (yes/no)

    Is this person actively looking to buy or implement a solution soon?

  • Choice

    Who should handle this lead?

For developers: the questions file
recipes/01-lead-fit-intent.json · questions
{
  "fit": {
    "type": "score",
    "instructions": "How well does this lead match our ideal customer? Our ideal customer is: [REPLACE: e.g. B2B software companies with 50–1,000 employees that have a dedicated marketing or RevOps team].",
    "criteria": [
      "No fit",
      "Weak fit",
      "Plausible fit",
      "Strong fit",
      "Ideal customer"
    ]
  },
  "intent": {
    "type": "noul",
    "instructions": "Is this person actively looking to buy or implement a solution soon?",
    "criteria": {
      "true": "Mentions a timeline, budget, active evaluation, a concrete project, or asks for pricing or a demo.",
      "false": "Browsing, researching, a student, a job seeker, or a vendor pitching us."
    }
  },
  "route": {
    "type": "choice",
    "instructions": "Who should handle this lead?",
    "criteria": {
      "sales": "A potential customer worth a sales conversation now",
      "nurture": "A potential customer who is early or small; add to nurture",
      "partner": "An agency, consultant, reseller or integration partner",
      "support": "An existing customer with a support question",
      "other": "Job seeker, vendor, spam or otherwise not a buyer"
    }
  }
}

Questions before you buy

What's in the Marketing Recipe Pack?
10 decision recipes for marketing and growth work, each with the exact Jev questions, the input to send, the thresholds to act on and the cost to run. It ships as a ZIP of Markdown, JSON and a small TypeScript helper.
What else do I need?
A Jev API key from TypeSafe AI or a gateway such as Vercel AI Gateway or OpenRouter; usage is billed by them, typically cents. Running the recipes means running a small script or wiring the questions into a tool you already use.
How do I get the files after paying?
You're sent straight to a download page after checkout. If you lose the link, reply to your receipt email or contact us with your payment ID and we'll resend it.
Can I get a refund?
Yes. If the pack isn't useful to you, email us within 14 days of purchase for a full refund. See the refund policy for details.
Do I need an account on JevHub?
No. There are no accounts on JevHub. Checkout is handled by Dodo Payments.