aaron

Quick win (under 5 minutes): Ask an AI to generate a 60–90 second demo opener tailored to the prospect’s industry and top pain. Use the prompt I provide below and you’ll have a ready-to-read script in under five minutes.

Good point focusing on enterprise demos and discovery — that’s where deals either accelerate or stall. Here’s a practical, results-first plan to use AI to prepare for and win those meetings.

The problem: Demos that sound generic and discovery calls that miss the buyer’s real priorities lead to lost momentum and long sales cycles.

Why it matters: A 5–15% lift in demo-to-opportunity conversion or a 10–20% reduction in time-to-close materially improves quarterly revenue.

My experience / key lesson: Use AI to do the heavy prep — persona, value framing, tailored agenda, and objection playbooks — then rehearse and humanize. AI removes grunt work so you can focus on relationship and judgment.

1. What you’ll need
   • Basic account notes: company name, industry, role(s) attending, any public facts (ARR, use case)
   • Product value pillars (3–5 bullets)
   • Access to an AI assistant (chat interface or API)
2. Step-by-step prep (how to do it)
   1. Run the persona brief: ask AI to summarize the buyer’s priorities and likely objections based on role and industry.
   2. Create a 3-section demo script: 90s opener, 7–10 minute walkthrough tied to KPIs, 10-minute close with next steps.
   3. Generate 10 targeted discovery questions prioritized by impact (technical, operational, financial, decision-process).
   4. Build objection responses: map top 5 likely objections to concise ROI-based rebuttals and supporting data points.
   5. Rehearse with AI as the buyer to surface follow-up questions and time cues.

What to expect: A tailored meeting plan, cleaner messaging, reduced fumbling on objections, and a measurable improvement in conversion and meeting efficiency.

Copy-paste AI prompt (use this as-is):

“You are a senior enterprise sales coach. The prospect is [Company], in the [Industry] industry, with attendees: [Role 1], [Role 2]. Their likely top business goal is [e.g., reduce churn by X, scale operations, cut costs]. Produce: 1) a 90-second demo opener that ties to their KPI, 2) a 7-step demo flow with time allocations and what to show, 3) 12 discovery questions prioritized by impact, and 4) 5 concise objection rebuttals framed with ROI or risk-reduction metrics.”

Metrics to track

• Demo-to-opportunity conversion rate
• Average meeting length and on-time agenda completion
• Follow-up meeting rate (next-step acceptance)
• Time-to-close
• Objection closure success rate

Common mistakes and fixes

• Too generic opening — Fix: use the AI-generated opener tied to a KPI.
• Too many features, not enough outcomes — Fix: lead with ROI statements and outcome metrics.
• No plan for next steps — Fix: close every demo with a defined decision-date and owners.

1-week action plan

1. Day 1: Run the AI prompt for your top 3 deals; save scripts and questions.
2. Day 2: Rehearse one demo using the script; iterate language to sound natural.
3. Day 3: Run a mock discovery with AI playing buyer; capture new objections.
4. Day 4: Update collateral (slide highlights, demo checklist) based on rehearsal.
5. Day 5: Deliver a real demo using the new plan; capture metrics.
6. Weekend: Review results and adjust prompts/flows for next week.

Your move.

Hook: If a 10-year-old can explain your product at dinner, you’ll convert more adults on Monday. That’s the leverage of kid-friendly AI explanations.

Problem: Most AI answers default to textbook tone. Jargon, long sentences, fuzzy analogies. The result: confused readers, slow decisions, lost deals.

Why it matters: Clear explanations shorten sales cycles, boost training completion, and cut support tickets. Expect faster onboarding, higher page dwell time, and more “I get it now” replies.

Lesson from the field: You control quality by constraining reader profile, allowed vocabulary, analogy domain, and a teach-back test. Don’t ask for “simple.” Specify exactly what “simple” means and make the AI prove understanding.

What you’ll need:

• An AI chat tool
• 2–3 real topics (e.g., your product’s pricing model, an industry concept)
• 10 minutes per topic

The core, copy-paste prompt (use this as your default):

Explain {TOPIC} to a curious 10-year-old. Audience: {AGE 8–12}, no prior knowledge. Role: friendly librarian. Constraints: use words common to children; sentences 8–12 words; no jargon or acronyms; use one simple analogy from {ANALOGY DOMAIN}; include a 3-step example with numbers; end with a 3-question quiz and an answer key. Add a one-line moral that starts with “So,”. Reading level target: Grade 4–5. If any term is complex, add a one-sentence mini-glossary. Then list 3 ways the analogy could mislead.

Advanced prompt chain (for higher accuracy):

1. Calibrate the reader: Ask the AI to ask you three questions about the reader’s age, interests, and example domain. Then answer them. This ensures relevant analogies.
2. Produce the explanation: Use the core prompt.
3. Teach-back: Ask the AI to create a one-sentence summary the child could say. If it’s off, request “Explain again with a new analogy.”
4. Compression: “Now give me a 60-word version and a 15-second story version.”
5. Transfer: “Give two more examples from different everyday contexts.”

Insider trick: Use a blocklist and must-use words. You can do this:

Do not use: leverage, paradigm, stakeholder, algorithm, distributed ledger, throughput. Must use: because, for example, picture, step. Keep sentences 8–12 words. If you need a hard word, explain it in one short line.

Template with slots (premium version):

Explain {TOPIC} like I’m {AGE}. Use the analogy of {ANALOGY SOURCE, e.g., lemonade stand or school library}. Limit to {MAX WORDS}. Steps: 1) One-sentence idea; 2) Analogy in 3 steps; 3) Worked example with numbers; 4) Mini-glossary (max 4 items); 5) 3-question quiz with answers; 6) One sentence: “So, …”. Reading level: Grade {4–6}. Ask me one question to check understanding at the end.

Simple examples (copy, paste, run):

• Blockchain via a school notebook: Explain blockchain like I’m 10. Analogy: a class notebook that everyone can read. Show how a fake entry gets caught. Include a cookie-trading example with numbers. End with a 3-question quiz.
• Inflation via shopping: Explain inflation like I’m 9. Analogy: stickers getting pricier at the school store. Show a $5 allowance over 3 months. Include why saving and earning both matter.
• Photosynthesis via lunch-making: Explain photosynthesis like I’m 8. Analogy: the plant kitchen. Use sunlight, water, air as “ingredients.” Give a 3-step recipe and a one-line moral.

Step-by-step execution (10 minutes per topic):

1. Select topic that causes confusion (support logs or FAQs).
2. Choose analogy domain your audience knows (home, school, sports).
3. Run the core prompt with your topic and domain.
4. Scan for jargon. If any slips in, reply: “Replace all uncommon words. Keep grade 5.”
5. Teach-back. Ask for: “What one sentence could a 10-year-old say now?”
6. Package variants: 60-word version, story version, and a numbered steps version.
7. Store winning analogies in a shared doc for reuse.

What good output looks like: 120–180 words, one clear analogy, simple numbers, a mini-quiz with answers, and a final “So,” line that states the practical takeaway.

Metrics to track (weekly):

• Comprehension score: quiz accuracy from 5–10 internal testers (target: 80%+ correct on first try).
• Reading level: Flesch–Kincaid Grade 4–6 (ask the AI to report it).
• Time to clarity: time until a tester can explain back in one sentence (target: under 60 seconds).
• Support deflection: % drop in “what does X mean?” tickets after publishing simplified copy (target: 15–30% drop in 30 days).
• Engagement: average time-on-page for your explainer posts (target: +20%).

Common mistakes and fast fixes:

• Mistake: Analogy dominates and becomes inaccurate. Fix: Add the “3 ways this analogy could mislead” requirement and keep one analogy only.
• Mistake: Vague audience description. Fix: Force a 3-question audience calibration first.
• Mistake: Walls of text. Fix: Limit word count and require numbered steps.
• Mistake: Hidden jargon. Fix: Provide a blocklist and a mini-glossary.
• Mistake: No proof of learning. Fix: Always include a quiz and teach-back.

One-week rollout plan:

1. Day 1: Pick 5 topics from FAQs and sales objections. Define an analogy domain for each.
2. Day 2: Generate first drafts with the core prompt. Add blocklist and must-use words.
3. Day 3: Run teach-back and compression steps. Produce 60-word and story versions.
4. Day 4: Test with 5 internal readers. Collect quiz accuracy and time-to-clarity.
5. Day 5: Revise weak sections. Swap analogies where the “mislead” list is long.
6. Day 6: Publish to help center, onboarding emails, and sales decks.
7. Day 7: Review metrics. Set targets for support deflection and engagement.

Final, robust prompt (paste this when time is tight):

Explain {TOPIC} for a curious 10-year-old. Role: friendly librarian. Constraints: Grade 5 reading level; sentences 8–12 words; no acronyms; one clean analogy from {ANALOGY DOMAIN}; include a 3-step numeric example; add a 3-item mini-glossary; finish with a 3-question quiz and answers; end with “So,” + the practical takeaway. Then give a one-sentence teach-back the child could say. Report the estimated reading grade.

Make this your standard. You’ll get clearer customers, faster decisions, and better training outcomes.

Your move.

— Aaron

Hook: Use AI to turn every enterprise demo into a tailored, measurable step toward a win — not a shotgun show-and-tell.

Problem: Most demos are generic, feature-heavy, and fail to connect to the buyer’s priorities. That costs time and reduces conversion.

Why it matters: A focused demo that speaks to specific stakeholders increases demo-to-proposal conversion, shortens sales cycles, and raises perceived value — directly improving revenue per lead.

Checklist — Do / Do‑Not

• Do: Map stakeholder outcomes and quantify impact (time saved, cost avoided, revenue enabled).
• Do: Use AI to draft tailored talking points, discovery questions, objection responses, and follow-up assets.
• Do: Rehearse with AI as a role-playing buyer to refine timing and answers.
• Do‑Not: Lead with features — stop assuming everyone needs a full feature tour.
• Do‑Not: Skip pre-meeting research — never go into a demo blind to org priorities.

Experience / Lesson: In enterprise cycles the demos that close are the ones that: (1) show outcomes for each stakeholder, (2) have a 10–15 minute tailored core demo, and (3) end with a clear next step tied to an internal decision milestone.

Step-by-step: What you’ll need, how to do it, what to expect

1. Gather inputs — CRM entry, job titles, any public company intel, current customer metrics. Expect: 10–30 minutes.
2. Run stakeholder mapping with AI — ask AI to list likely priorities for each title and suggested KPIs to cite. Expect: a 1‑page persona brief.
3. Generate a 10–15 minute demo script — highlight 3 scenarios mapped to pain → solution → impact. Expect: exact speaking cues and screen sequence.
4. Prep discovery & objections — create tailored discovery questions and 6 live objection responses. Expect: confident, quick rebuttals and escalation triggers.
5. Create follow-up assets — one-page ROI snapshot, next-step checklist, calendar-ready proposal timeline. Expect: email + PDF to send within 2 hours after demo.
6. Rehearse — role-play with AI as buyer using the prepared script, adjust timing and language. Expect: tighter delivery and fewer surprises.

Copy‑paste AI prompt (use as-is):

“You are a senior procurement manager at a mid-market logistics company. The company struggles with route optimization, high fuel costs, and late deliveries. Create a 10‑15 minute product demo script that focuses on three scenarios: route optimization for cost savings, real-time exception handling for SLA adherence, and executive dashboard for KPIs. For each scenario provide: (1) 2-line context, (2) demo flow with exact screens to show, (3) 1 quantified outcome to cite, and (4) 1 anticipated objection and a concise response.”

Metrics to track

• Demo → Proposal conversion rate
• Meeting → Demo attendance rate
• Average demo length and % spent on tailored scenarios
• Follow-up email open/reply rate within 48 hours
• Time from demo to contract signed

Mistakes & fixes

• Mistake: Overwhelming with features. Fix: Limit to 3 scenarios mapped to stakeholder KPIs.
• Mistake: No follow-up assets. Fix: Send a one‑page ROI and next‑step checklist within 2 hours.
• Mistake: No stakeholder mapping. Fix: Use AI to create a persona brief and align your demo to each persona’s ROI.

Worked example (concise)

Company: National logistics firm. Stakeholders: VP Ops (reduce costs), Head of Dispatch (reduce exceptions), CFO (ROI). AI output used to create: 12‑minute demo script showing route optimizer -> exception dashboard -> executive KPIs. Outcome cited: 12% fuel cost reduction example from a similar case. Result: Demo-to-proposal increased from 18% to 38% in 6 weeks.

1‑Week action plan

1. Day 1: Pull CRM record + company notes (30m). Run stakeholder mapping prompt (30m).
2. Day 2: Generate demo script and discovery questions (1h). Create ROI snapshot template (1h).
3. Day 3: Rehearse with AI role-play twice (45m). Adjust script (30m).
4. Day 4: Finalize slides/screens + follow-up email (1h).
5. Day 5: Run a dry run with peer (30–60m), send calendar invite with confirmatory discovery questions.

Your move.

Noted there wasn’t an earlier point — I’ll pick up from zero and give a direct, usable path. Short answer: yes, AI can do market research and summarize trends for a GTM plan, but only if you run it through a tight, human-led process.

The reality: AI nails speed and synthesis (scanning reports, social sentiment, competitor mentions). It misses context unless you set the scope, validate outputs and convert insights into measurable experiments.

Why that matters: a fast, repeatable AI-driven research loop reduces time-to-decision, focuses your GTM experiments, and lowers the cost of hypothesis testing.

What I’ve learned: give AI a clear brief, multiple data sources, and a validation step. Treat AI output as a hypothesis generator, not a final plan.

1. Define goals & scope (what you’ll need)
   • Inputs: target segment, geography, product category, time horizon (3–12 months).
   • How: write a 2–4 sentence objective (e.g., “Identify top 3 channels and 5 buyer pain points for SMB accounting software in US, Q1-Q2”).
   • Expect: concise objective that guides every prompt.
2. Collect sources (what you’ll need)
   • Industry reports, Google News, LinkedIn posts, customer reviews, competitor websites, basic KPI data (benchmarks).
   • How: assemble into a single folder or sheet; note source and date.
   • Expect: 10–20 raw documents/URLs to feed AI.
3. Run targeted AI prompts (how to do it)
   • Use the prompt below (copy-paste). Run a first pass, then ask follow-ups for gaps.
4. Synthesize & validate (what to expect)
   • Turn AI output into a 1-page GTM: target audience, positioning, channels, 3 experiments, sample messaging.
   • Validate 5–10 customer/partner calls or quick surveys in 3 days.

Key metrics to track

• Time-to-first-insight (goal: <48 hours)
• Hypotheses generated and validated (target: 5 hypotheses, 60% validation)
• Experiment conversion lift vs baseline
• Cost-per-insight (hours & dollars saved vs agency research)

Common mistakes & fixes

• Relying on a single source — fix: triangulate (3+ sources per claim).
• Vague prompts — fix: give structure and desired outputs (bullets, tables).
• Skipping validation — fix: run micro-experiments or 5 customer interviews before scaling.

One-week action plan

1. Day 1: Define objective and gather 10–20 sources.
2. Day 2: Run primary AI prompt (below). Review results; note gaps.
3. Day 3: Deep-dive follow-ups for competitor and customer insights.
4. Day 4: Draft 1-page GTM and 3 experiments.
5. Day 5: Validate with 5 customer calls or quick surveys.
6. Day 6: Adjust plan and finalize KPIs for experiments.
7. Day 7: Launch first experiment and monitor.

Copy-paste AI prompt (primary)

“You are an expert market analyst. Using the following sources [paste URLs and notes], summarize the market trends for [product/category] targeting [segment] in [region] over the past 12 months. Deliver: 1) Top 5 industry trends with evidence and source links, 2) Top 6 buyer pain points (ranked by frequency), 3) 4 direct competitors with their strengths/weaknesses, 4) Recommended positioning statement (one sentence), 5) Top 3 customer acquisition channels with estimated CPA and rationale, 6) Three 30-day GTM experiments (hypothesis, metric to track, expected outcome). Format as bullet lists with sources for each claim.”

Variants

Competitor focus: “Summarize competitors: product offerings, pricing signals, messaging themes, recent moves (product launches, funding, partnerships).”

Customer focus: “Analyze customer reviews and social posts to list top 10 pain points and suggested value props to test in messaging.”

Your move. — Aaron

Hook: Teach AI to explain anything so a child can understand it—without sounding patronising or wrong.

Nice starting point: aiming for “kid‑friendly” explanations forces clarity and reveals gaps in your own thinking. Good call.

The problem: Generic AI outputs are either too technical, vague, or oversimplified to the point of being incorrect. That kills learning and trust.

Why it matters: Clear, accurate, age-appropriate explanations reduce support costs, increase adoption, and make your product or message accessible to families and non-experts.

Lesson from practice: When I refine prompts to require an age band, a relatable analogy, and a 3-question comprehension check, pass rates and follow-up questions fall sharply.

1. What you’ll need:
   • Topic or concept (one sentence)
   • Target age band (e.g., 6–8, 9–11)
   • Tone (playful, neutral, formal)
   • Length limit (e.g., 2–4 sentences)
   • 1 real-world example or constraint
2. How to do it (step-by-step):
   1. State the objective: what you want the child to walk away knowing.
   2. Specify age, tone, and length.
   3. Ask for a simple definition, one analogy, and one short example.
   4. Request 2–3 true/false or multiple-choice check questions plus answers.
   5. Ask for a single-sentence extension for curious kids (optional deeper link).
3. What to expect: Short, concrete explanations, memorable analogy, and a mini-quiz to confirm understanding.

Copy-paste prompt (use as-is):

Explain [TOPIC] to a [AGE] year-old in 2–4 simple sentences. Use a friendly, [TONE] tone. Include: (1) one one-sentence definition, (2) one analogy a child would relate to, (3) a one-sentence everyday example, and (4) two simple multiple-choice questions with correct answers. Keep language concrete and avoid technical words.

Variants: Replace [AGE] with 6–8, 9–11, or 12–14 and [TONE] with playful/neutral.

Metrics to track:

• Comprehension rate (quiz correct %)
• Number of clarifying follow-ups per piece
• Time to first/last interaction (engagement)
• User satisfaction / helpfulness score

Common mistakes & fixes:

• Too simplistic = loses accuracy. Fix: require a short extension that includes one precise fact.
• Condescending tone. Fix: specify tone and sample phrasing in prompt.
• Bad analogy. Fix: ask for two analogies and choose the better one.

1-week action plan:

1. Day 1: Create 10 prompts covering key topics and age bands.
2. Day 2–3: Run outputs with target audience (kids or proxies) and collect quiz results.
3. Day 4: Tally metrics and identify 3 weak areas (analogies, tone, accuracy).
4. Day 5–6: Iterate prompts, test top 3 fixes.
5. Day 7: Deploy best prompts and measure initial KPIs.

Your move.

— Aaron

Good question. Everyday writing—not just marketing copy—benefits most from AI when you give it clear constraints and keep control of meaning.

Here’s the fastest way I know to cut jargon and lift readability without dumbing anything down: build a simple Readability Spec + Control Dictionary, then run your draft through 3 tight AI passes (diagnose, rewrite with controls, risk-check). Expect a 30–50% jargon reduction and shorter sentences in under 10 minutes per piece.

Why this matters: Clearer writing gets faster approvals, higher reply rates, fewer clarifying emails, and less risk of misinterpretation. It’s a direct line to measurable business outcomes.

Lesson from the field: AI improves readability only when you tell it what to preserve. The trick is a Control Dictionary (terms to keep or define) and a Readability Spec (grade level, tone, sentence length, formatting). With those, you get precision, not mush.

Copy-paste prompts (start here)

• 1) Baseline analysis: “You are a readability editor. Analyze the text below. Return: a) Flesch Reading Ease, b) Grade Level, c) avg sentence length, d) list of jargon/acronyms, e) which sentences are hard and why, f) estimated reading time. Do not rewrite yet. Text: [PASTE TEXT]”
• 2) Rewrite with controls: “Rewrite the text for [AUDIENCE: e.g., busy managers over 40] using this Readability Spec: Grade [7–9 general / 10–12 technical], tone [professional, warm], sentences [12–18 words], keep bullets and headings, remove filler, prefer concrete verbs. Control Dictionary: keep these terms as-is unless defined once in plain English: [LIST TERMS]. Replace or define these terms: [LIST TERMS]. Preserve legal/compliance meaning. Output: a) revised version, b) list of changes with reasons.”
• 3) Risk check (meaning and tone): “Compare original vs revised. List any places meaning could have shifted, any numbers/dates altered, and any tone changes that may affect authority. If risk exists, suggest precise fixes (one sentence each).”
• Variant: Jargon-to-plain glossary: “From the text, build a glossary of confusing terms. For each: a) plain-English definition in one line, b) when to use/avoid, c) a simpler synonym. Keep accuracy.”
• Variant: Audience simulation: “Act as [ROLE: non-technical customer/procurement/board]. List 5–10 questions you still have after reading the revised text. Flag any sentences that feel unclear or too technical. Suggest one line to improve each.”

What you’ll need

• A capable AI assistant.
• Your draft (email, memo, proposal).
• Readability Spec (one-time setup).
• Control Dictionary (terms to keep vs replace).

How to do it (step-by-step)

1. Draft normally. Don’t self-edit for simplicity yet.
2. Run the Baseline analysis prompt. Capture metrics.
3. Create/update your Control Dictionary: must-keep terms (brand, legal, technical), terms to replace/define, preferred synonyms.
4. Run the Rewrite with controls prompt. Review the change list.
5. Run the Risk check. Accept or apply the suggested fixes.
6. Optional: Audience simulation to surface blind spots.
7. Paste the revised text into your email/doc. Do a quick human read for tone and any sensitive nuances.

What to expect

• Shorter sentences, fewer acronyms, clearer verbs.
• Meaning preserved where specified; definitions added once then shortened.
• Improved scannability with bullets and logical flow.

Metrics to track (weekly)

• Flesch Reading Ease: aim 60+ for general audiences; 50–60 for technical audiences.
• Grade Level: 7–9 general; 10–12 technical/internal.
• Average sentence length: target 12–18 words.
• Jargon count: total terms flagged vs remaining (aim 30–50% reduction).
• Response KPIs: reply rate on emails, time-to-approval on memos, fewer clarification replies.
• Reading time: aim under 2 minutes for most emails; under 6 minutes for one-pagers.

Common mistakes and quick fixes

• Mistake: AI over-simplifies and loses authority. Fix: Raise Grade Level one notch and add “preserve technical nuance; do not remove qualifiers.”
• Mistake: Key terms changed. Fix: Expand the Control Dictionary with “do-not-change” items and rerun the rewrite.
• Mistake: Corporate tone becomes generic. Fix: Feed 2–3 good past samples and say “match voice, vary sentence openings, keep confident cadence.”
• Mistake: Bloated rewrites. Fix: Add “maximum length [X words]; cut redundancy by merging similar points.”
• Mistake: Inconsistent formatting. Fix: Specify “keep bullets, bold key actions, one idea per paragraph.”

Insider trick: Build a reusable Readability Spec once, then prepend it to every prompt. Over time, your AI learns your house style. Keep the Control Dictionary alive—add terms after each project.

1-week action plan

1. Day 1: Draft your Readability Spec and initial Control Dictionary. Pick 10 recurring terms to keep/replace.
2. Day 2: Run the Baseline analysis on 5 recent emails. Record metrics in a simple log.
3. Day 3: Rewrite those 5 with the controls prompt. Track before/after metrics and reply rates.
4. Day 4: Apply the workflow to one longer document (one-pager or proposal). Use Audience simulation to refine.
5. Day 5: Add the top 15 new jargon terms to your Control Dictionary. Set default Grade Level targets by audience.
6. Day 6: Create a short internal checklist: “Run 3 passes: Baseline, Rewrite, Risk.” Share with your team.
7. Day 7: Review results. If Reading Ease <60 or replies flat, tighten the spec (shorter sentences, stricter length cap) and iterate.

Premium template you can reuse

“Editor, apply our Readability Spec: Grade [X], tone [Y], sentences [12–18 words], bullets on, bold key actions, no clichés, keep voice [confident, warm]. Control Dictionary: keep [A,B,C]; define once then shorten [D,E,F]; replace with [preferred synonyms]. Tasks: 1) return metrics (Flesch, Grade, avg sentence length, jargon count), 2) produce revised text under [N words], 3) list top 5 changes with reasons, 4) log any risk to meaning. Text: [PASTE].”

Your move.

Hook — People stop reading the moment they hit jargon. Use AI to delete friction, lift reply rates, and make decisions faster.

The problem — Everyday writing gets clogged with acronyms, internal shorthand, and long sentences. That slows readers down, invites confusion, and drives silence.

Why it matters — Readable messages get more replies, fewer back-and-forths, and faster approvals. If you improve clarity by one grade level, you’ll feel it in your inbox and your calendar.

Field lesson — The teams that win treat readability like a KPI. They set a grade target, run a repeatable AI edit loop, and measure response rates. No heroics—just a simple, consistent system.

What you’ll need

• An AI assistant (any reputable model is fine)
• Your audience profile (who they are, what they care about)
• A short “keep/kill” jargon list
• Two recent samples of your writing (email + doc)

How to do it (repeatable system)

1. Set targets. Choose a reading level (Grade 7–8 for general audiences, 9–10 for technical teams) and sentence length (12–16 words). Decide your tone (plain, direct, warm).
2. Build a jargon ledger. Make a two-column list: terms to keep (brand names, legal terms) and terms to translate (e.g., “utilize → use,” “leverage → use,” “synergy → working well together”).
3. Create your AI prompt template. Use the prompt below as your standard. Save it and reuse it for every draft.
4. Run a three-pass edit.
   • Trim: remove fluff, shorten sentences.
   • Translate: swap jargon with plain words; define any must-keep term once in parentheses.
   • Test: get a readability report and a shorter V2.
5. A/B check subject lines and first paragraphs. Ask AI for 3 options each; pick the clearest, not the cleverest.
6. Lock in your voice. Give AI two of your best emails and say “mirror this style” so it keeps your personality while simplifying.
7. Automate. Create a quick checklist: target grade, average sentence length, passive voice under 5%, jargon replaced, one-sentence summary added.

Copy‑paste prompt (robust template)

“You are my readability editor. Audience: [describe]. Goal: reduce jargon and improve clarity without losing accuracy. Targets: Grade 7–8; average sentence length 12–16 words; active voice; no idioms; no emojis. Keep: proper nouns, legal terms, numbers, dates, and the brand voice. If a technical term is required, define it once in parentheses on first use, then use the short term thereafter. Produce three sections: 1) V1 simplified (same meaning), 2) V2 concise (15% shorter), 3) Readability report (Flesch grade, avg sentence length, passive %), plus a 1‑sentence summary and three subject line options (if email). Ask up to 3 clarification questions if anything is ambiguous. Text: [paste your draft]”

Insider trick — Add a “keep/kill” list to the prompt each time. It forces consistency and stops AI from dumbing down terms you must keep.

What to expect

• Reading grade drops by 2–4 levels on first pass
• Shorter message length (10–25% reduction)
• Higher open/reply rates on emails and faster approvals on docs

Metrics to track (weekly)

• Flesch‑Kincaid Grade Level (target ≤ 8 for general audiences)
• Average sentence length (target 12–16 words)
• Passive voice rate (target ≤ 5%)
• Email reply rate or “yes” rate on approvals
• Time-to-approval or back‑and‑forth count (aim to reduce by 20–30%)
• Support or follow‑up questions after sending (aim to reduce by 25%)

Common mistakes and fast fixes

• Over-simplifying accuracy away. Fix: include a must‑keep terminology list and require a definition on first use.
• Letting AI change facts. Fix: ask for “language edits only—do not add new claims.”
• Generic tone. Fix: paste two strong samples of your voice and say “match this tone.”
• One-and-done editing. Fix: run the three‑pass loop and request a readability report every time.
• Ignoring mobile screens. Fix: require paragraphs under 3 lines and scannable bullets for key points.

One‑week action plan

1. Day 1: Define audience, set grade target, and draft your keep/kill jargon ledger (20 items).
2. Day 2: Save the prompt template. Run two existing emails through it. Compare V1 vs V2; choose the best.
3. Day 3: Build a 5‑point checklist (grade, sentence length, passive %, jargon replaced, 1‑sentence summary). Print it or pin it.
4. Day 4: Calibrate tone: feed AI two of your best messages and instruct “mirror this style” before editing a new draft.
5. Day 5: A/B test subject lines and first paragraphs on one outbound email campaign or internal update.
6. Day 6: Create a mini glossary page at the top of a policy or FAQ and have AI insert first‑mention definitions.
7. Day 7: Review metrics: grade level, sentence length, reply/approval rates. Keep what worked; add any new jargon to the ledger.

Refined micro‑templates you can reuse

• First‑line clarity: “In one sentence, what changes and what you need from the reader.”
• Three‑layer summary: 1 sentence, 3 bullets, then details.
• Definition rule: First mention: term (plain definition). Later mentions: short term only.

Make readability a system, not a hope. Set targets, use the prompt, measure the result. Your move.