Jev + Claude / Sales & marketing / By Felix Wickholm
11 Jev Use Cases That Feel Like Cheating (Sales & Marketing)
Use Jev and Claude to review leads, find sales friction and turn customer questions into useful marketing. Eleven workflows with examples, prompts and an interactive canvas.
Pick one workflow below. Read the example, then copy its prompt into Claude or your preferred AI assistant. Jev helps review the inputs; Claude helps you build the workflow. The examples show what to try, rather than measured results.
Use material you are authorised to process. Review the output before taking action.
01Sales
Your best leads are hiding in plain sight
Give Jev your offer and your LinkedIn contacts. It scores how well each contact fits, so you can choose who to review first.
Illustrative examples
Input
Review or next step
Agency founder, 12-person team
Strong fit
Student seeking an internship
Low fit
Head of marketing at an agency
Strong fit
Rank contacts by how well they fit your offer.
How to try it
Supply a small, authorised sample and explain your business context.
Define the Score criteria with clear examples and an unclear option.
Review the output against the original material. Correct the rubric before expanding the batch.
Use an authorized contact export or supplied professional profile text. Explain the role, industry and company-fit criteria. Scores are illustrative, not buying intent. Review profiles before any outreach.
Copy this into your AI assistant
Use only the professional contact information I provide. Ask for my offer, ideal customer, and disqualifying criteria. Design a Jev Score workflow to review fit by role, company and need. Return a contact ID, supporting excerpt, fit assessment and uncertainty. Do not infer buying intent or send outreach. Ask me for the inputs and confirm the current Jev API setup before writing integration code.
Give Jev your website copy. It flags vague claims, repetition and empty buzzwords. Ask Claude to rewrite the weak passages using real details.
Illustrative examples
Input
Review or next step
Unlock limitless possibilities
High slop likelihood
We build websites for dentists
Low slop likelihood
Your success is our passion
High slop likelihood
Find vague copy. Make it specific.
How to try it
Supply a small, authorised sample and explain your business context.
Define the Noul criteria with clear examples and an unclear option.
Review the output against the original material. Correct the rubric before expanding the batch.
Define the condition explicitly: vague, repetitive or generic copy with little useful information. Noul estimates the probability that this condition holds. High/low labels here are illustrative, not measured outputs. This is writing-quality assessment, not reliable detection of AI authorship. Retain useful specific writing and verify factual claims.
Copy this into your AI assistant
Help me define a writing-quality rubric for vague claims, repetition and empty buzzwords. Design a Jev Noul check for these traits in each passage of my supplied copy. Return flagged passages and the reason for review. Then suggest specific rewrites using only facts I supply. Do not infer AI authorship. Ask me for the inputs and confirm the current Jev API setup before writing integration code.
Give Jev your sales-call transcripts. It groups the problems that keep coming up. Check the biggest pattern against your sales numbers before deciding what to fix.
Illustrative examples
Input
Review or next step
Wrong people keep booking calls
Targeting problem?
Buyers cannot explain your offer
Value unclear?
Deals stall at the price
Pricing or proof?
Find the repeated problem. Check it against your numbers.
How to try it
Supply a small, authorised sample and explain your business context.
Define the Choice criteria with clear examples and an unclear option.
Review the output against the original material. Correct the rubric before expanding the batch.
Use authorized transcripts with call IDs and timestamps. Inventory every available call, report omissions and sample ambiguity. Jev classifies passages; code counts patterns. Calls alone cannot establish the constraint across the whole business. Validate with funnel, capacity and delivery data.
Copy this into your AI assistant
Help me analyse the sales transcripts I provide. First inventory call IDs and missing coverage. Design a Jev Choice workflow to classify friction into targeting, value, price, timing, trust, delivery or unclear. Keep quotes and timestamps. Count categories with code, inspect ambiguous cases, and propose a bottleneck hypothesis plus the business metrics needed to test it. Ask me for the inputs and confirm the current Jev API setup before writing integration code.
Find the questions customers keep asking on sales calls. Jev groups them by topic. Claude helps turn them into useful pages for your website.
Illustrative examples
Input
Review or next step
How much does this cost?
Pricing page
How are you different from X?
Comparison page
Will this work with our CRM?
Integration guide
Turn customer questions into useful website pages.
How to try it
Supply a small, authorised sample and explain your business context.
Define the Choice criteria with clear examples and an unclear option.
Review the output against the original material. Correct the rubric before expanding the batch.
Transcribe recordings before Jev. Use the full authorized transcript inventory, preserve source timestamps, deduplicate recurring questions and check search intent. Draft with Claude; review factual claims and remove private details. Useful content can support SEO/GEO, but rankings and AI citations are not guaranteed. Measure search performance and observed citations separately.
Copy this into your AI assistant
Use my supplied sales transcripts to extract recurring buyer questions with call IDs and timestamps. Design a Jev Choice workflow to group pricing, comparisons, integrations and how-to questions. Deduplicate overlapping intent. Produce a prioritised content backlog and a brief for the first page. Remove private details and list claims needing verification. Do not promise rankings or AI citations. Ask me for the inputs and confirm the current Jev API setup before writing integration code.
Give Jev your offer and incoming enquiries. It scores which ones fit your business. Review the strongest matches first, then approve the follow-up.
Illustrative examples
Input
Review or next step
Needs a service website
Strong fit
Selling a contact list
Low fit
Wants a marketing partner
Strong fit
Review the enquiries that fit your offer first.
How to try it
Supply a small, authorised sample and explain your business context.
Define the Score criteria with clear examples and an unclear option.
Review the output against the original material. Correct the rubric before expanding the batch.
Explain the fit criteria using a fictional offer. A fit score does not prove buying intent. Show one ambiguous enquiry and why you would review it.
Copy this into your AI assistant
Ask for my offer and qualification rules. Design a Jev Score workflow for the enquiries I supply. Return enquiry ID, fit assessment, evidence and missing information. Keep ambiguous enquiries for human review. Draft a relevant follow-up for approval; do not send anything or treat fit as purchase intent. Ask me for the inputs and confirm the current Jev API setup before writing integration code.
Give Jev your sales notes. It groups objections about price, timing and trust. Use the repeated questions to improve your offer or explain it more clearly.
Illustrative examples
Input
Review or next step
It costs more than expected
Price
Maybe next quarter
Timing
Have you done this before?
Trust
Answer the objection you keep hearing.
How to try it
Supply a small, authorised sample and explain your business context.
Define the Choice criteria with clear examples and an unclear option.
Review the output against the original material. Correct the rubric before expanding the batch.
Use fictional or explicitly authorized notes. The grouping helps you inspect a pattern; counts should come from code, not invented examples.
Copy this into your AI assistant
Review the sales notes I provide. Design a Jev Choice workflow for price, timing, trust and other or unclear objections. Preserve source IDs, calculate actual counts with code and show ambiguous examples. Suggest one piece of useful content addressing a recurring objection. Distinguish stated objections from confirmed reasons a deal was lost. Ask me for the inputs and confirm the current Jev API setup before writing integration code.
Give Jev your audience brief and ad ideas. It helps shortlist the ideas that address a real customer problem. Test the shortlist to see what works.
Illustrative examples
Input
Review or next step
A specific customer problem
Review first
A list of generic features
Rework
An example they recognise
Review first
Choose the ad ideas worth testing.
How to try it
Supply a small, authorised sample and explain your business context.
Define the Score criteria with clear examples and an unclear option.
Review the output against the original material. Correct the rubric before expanding the batch.
This is text relevance, not a prediction of clicks or sales. Show the actual creative brief and keep any paid campaign off during the demonstration.
Copy this into your AI assistant
Ask for my audience, offer and creative brief. Design a Jev Score rubric for how directly each supplied ad concept addresses a specific customer problem. Return a shortlist with evidence, weaknesses and a proposed test. Do not predict clicks, revenue or conversion rates from these scores. Ask me for the inputs and confirm the current Jev API setup before writing integration code.
Give Jev your ad and landing-page copy. It flags promises that do not match. Fix the gap so visitors find the offer they clicked for.
Illustrative examples
Input
Review or next step
Ad: book a product demo
Ad promise
Page: download a report
Page promise
Do these promises match?
Check the gap
Make sure the ad and page promise the same thing.
How to try it
Supply a small, authorised sample and explain your business context.
Define the Noul criteria with clear examples and an unclear option.
Review the output against the original material. Correct the rubric before expanding the batch.
Show a fictional ad/page pair. Matching copy is a useful check, not proof of higher conversion.
Copy this into your AI assistant
Compare my supplied ad text and landing-page copy. Extract the audience, offer, promise and call to action from each. Design a Jev Noul check for material mismatches with explicit conditions. Show supporting quotes and propose edits using verified facts. This is a consistency check, not a conversion forecast. Ask me for the inputs and confirm the current Jev API setup before writing integration code.
Give Jev a few examples of your best writing and a new draft. It flags passages that sound different. Ask Claude to rewrite those parts in your style.
Illustrative examples
Input
Review or next step
Sounds like your examples
Keep reviewing
Too formal for the audience
Revise
Specific and easy to read
Keep reviewing
Keep your voice. Rewrite the parts that drift.
How to try it
Supply a small, authorised sample and explain your business context.
Define the Score criteria with clear examples and an unclear option.
Review the output against the original material. Correct the rubric before expanding the batch.
Use approved public examples or synthetic ones. A style score does not establish whether someone or an AI wrote the text.
Copy this into your AI assistant
Use the approved writing samples and new drafts I supply. Extract a concrete style rubric and design a Jev Score workflow for draft review. Show which passages drift from the rubric and why. Suggest revisions preserving meaning and verified claims. Do not make authorship judgments. Ask me for the inputs and confirm the current Jev API setup before writing integration code.
Give Jev your public comments. It groups questions, objections and experiences. Turn a question that keeps coming up into your next video.
Illustrative examples
Input
Review or next step
How do I connect my CRM?
Question
Will this work for my shop?
Objection
I tried it and got stuck
Experience
Turn a repeated question into your next video.
How to try it
Supply a small, authorised sample and explain your business context.
Define the Choice criteria with clear examples and an unclear option.
Review the output against the original material. Correct the rubric before expanding the batch.
Read retained comments in context. Categorization alone does not establish demand or future views.
Copy this into your AI assistant
Review the public comments I supply in context. Design a Jev Choice workflow to separate questions, objections, experiences and other or unclear comments. Group repeated questions and count them with code. Suggest three video ideas, each linked to supporting comments. Do not treat this as a view prediction. Ask me for the inputs and confirm the current Jev API setup before writing integration code.
Give Jev a call transcript. It flags vague answers and claims that need evidence. Use those flags to ask about deadlines, alternatives and proof. It cannot tell whether someone is lying.
Illustrative examples
Input
Review or next step
“We need this urgently.”
When do you need it, and why?
“Your price is too high.”
Compared with which option?
“Send me a proposal.”
What would make you say yes?
“We already have a solution.”
What is not working today?
“We guarantee twice as many leads.”
Can you show a case study?
Turn vague claims into specific questions.
How to try it
Supply a small, authorised sample and explain your business context.
Define the Choice criteria with clear examples and an unclear option.
Review the output against the original material. Correct the rubric before expanding the batch.
Fictional sales-call examples. These statements are not evidence of dishonesty. Read the full conversation before flagging them: the prospect may already have explained the deadline, alternative or decision criteria. Jev selects statements for review; a writing model can draft follow-up questions. Ask for specifics or supporting evidence, rather than inferring intent.
Copy this into your AI assistant
Review the call transcript I provide for contradictions, unsupported promises and missing commitments. Design a Jev Choice workflow to flag passages for review. Return exact quotes, timestamps, context, a possible benign explanation and a precise follow-up question. Do not claim to detect lies, assign truthfulness probabilities or infer intent. Ask for evidence before drawing conclusions. Ask me for the inputs and confirm the current Jev API setup before writing integration code.
