Jeff Bullas

Nice start — it’s great you’re focused on tone. A polite follow-up is often just a small tweak away from sounding persistent rather than pushy.

Why this matters: A short, value-led follow-up gets results. The goal is to remind, add value, and make it easy to respond.

What you’ll need

• Original message or thread summary
• A clear, single purpose for the follow-up (ask, info, meeting?)
• One added value item to include (helpful link, quick idea, times)
• Recipient’s name and context

Step-by-step: write a polite follow-up

1. Subject: keep it gentle and specific. Example: “Quick follow-up on [topic]”
2. Open with warmth: one-line acknowledgement. Example: “Hope you’re well.”
3. Reference the prior message briefly: “I wanted to follow up on my note about X sent last week.”
4. Add one tiny value point: a quick tip, link, or time options for a call.
5. Make the ask simple and low-effort: give two reply options (“Yes, I’m interested” / “Not now”).
6. Timeframe nudge, not demand: “If I don’t hear back, I’ll check in once more in two weeks.”
7. Sign off warmly and simply: name and a brief courtesy line.

Copy-paste prompt for ChatGPT

Prompt: “Rewrite this short follow-up so it sounds polite and non-pushy. Keep it under 80 words. Use a warm, professional tone. Reference the original email about [TOPIC], add one small value point (a quick tip or resource), and include a simple choice for reply (Yes / Not now). Original message: ‘[PASTE ORIGINAL MESSAGE HERE]’.”

Worked example

Original: “Did you see my last email about our services? Let me know.”

Rewritten (using prompt): “Hi [Name], hope you’re well. I’m following up on my note about [topic]. One quick idea that might help: [one-line tip]. If you’re interested, we could schedule 15 minutes — or simply reply ‘Not now’ and I won’t follow up. Thanks, [Your Name]”

Do / Do not — quick checklist

• Do: Keep it short, add value, offer clear choices.
• Do not: Repeat the same ask with pressure, use all caps or multiple exclamation marks.
• Do: Use a gentle timeframe for follow-up.
• Do not: Beg or apologize excessively.

Action plan — 3 quick wins

1. Today: pick one old outreach and rewrite using the prompt above.
2. This week: A/B test two subject lines on similar recipients.
3. Next week: Track responses — tweak the value item based on replies.

Keep it human, short, and useful. Follow-ups that respect the recipient’s time get replies — not resentment.

Yes — the 5‑minute export + AI extract gets you a tidy list fast. Here’s the next level: turn that JSON into an accountable loop that assigns owners, sets realistic dates and nudges follow‑through automatically.

Why this works

Chats are great at generating motion, not completion. A simple two-step upgrade — structured extraction and a tight publish‑and‑nudge loop — turns fuzzy threads into finished tasks without adding new tools or meetings.

What you’ll need

• Chat export or copy (24–72 hours is ideal).
• Any text‑capable AI.
• A one‑page “People Dictionary” (nicknames → real names, time zones, roles).
• One reviewer, with a 24‑hour validation SLA.
• A place to publish the action list (channel post, email, or shared doc).

The 3‑pass method (small effort, big clarity)

1. Pass 1 — Extract: Pull actions, owners, due dates, confidence, quotes.
2. Pass 2 — Normalize: Resolve nicknames, deduplicate, turn “ASAP/this week” into real dates, tag Decision vs Action vs Question.
3. Pass 3 — Assign: Apply default owner rules, set priorities, and prep a clean publishable list plus nudges.

Copy‑paste prompts

• Pass 1 — Extraction (JSON only) “Read the chat transcript. Extract every explicit or implied action item. Return a JSON array with fields: action (max 12 words), suggested_owner (name or ‘Unassigned’), suggested_due_date (verbatim from chat if present), confidence (High/Medium/Low), supporting_quote (exact message), type (‘Action’ or ‘Decision’ or ‘Question’). Also return a separate list of open questions. Do not add commentary.”
• Pass 2 — Normalization “You will normalize the previous JSON. Use this People Dictionary: [paste name → nickname → timezone → role]. Do the following: (1) Resolve nicknames to canonical names. (2) Convert vague dates into concrete dates using the owner’s timezone and these rules: ‘today’ = today 5pm; ‘tomorrow’ = next business day 5pm; ‘this week’ = Friday 5pm; ‘next week’ = next Wednesday 5pm; ‘ASAP’ = two business days 5pm. (3) Deduplicate near‑identical actions (Levenshtein or obvious wording overlap) and merge supporting quotes. (4) Keep action text ≤12 words. (5) Keep type tags. Output JSON only with fields: action, owner, due_date (YYYY‑MM‑DD), confidence, type, quotes (array), source.”
• Pass 3 — Assignment and publish pack “From the normalized JSON, apply a default‑owner rule: if owner is Unassigned, set owner to the thread starter or meeting host and mark ‘assumed_owner: true’. Add ‘priority’ (High if decision deadline mentioned or external dependency; else Medium; Low for nice‑to‑haves). Return two objects: (a) ‘publish_table’ as CSV with columns: Action, Owner, Due Date, Priority; (b) ‘nudges’ as a list of short DM messages for each owner with overdue or High‑priority items. Output only the CSV and the nudge messages.”

Step‑by‑step workflow (15–25 minutes end‑to‑end)

1. Export 24–72 hours of chat and remove obvious noise. Leave short context lines.
2. Run Pass 1 prompt. Skim the JSON for anything obviously wrong.
3. Paste your People Dictionary and run Pass 2 prompt. You’ll get clean names, real dates, and deduped tasks.
4. Run Pass 3 prompt. You’ll get a clean CSV for publishing and ready‑to‑send nudges.
5. Reviewer validates in under 24 hours: confirm owners, tweak dates, and pin the publish table to the channel.
6. Send nudges for unclaimed or High‑priority items. Reassign if no response in 24–48 hours.

Insider tricks that raise completion rates

• People Dictionary: Tiny doc that maps @handles and nicknames to real names and time zones. This alone halves mis‑assignments.
• Decision vs Action: Tag decisions separately so they don’t clog your task list; link actions back to the decision quote.
• Priority by consequence: If a task unblocks others or a client, it’s High by default.
• Pin and freeze: Pin the weekly action table; create a new one each week to avoid endless edits.

What good looks like

• First week: 15–30% edits by the reviewer; completion rate climbs as owners get clear nudges.
• Week two: Under 20% corrections; 80%+ on‑time completion; fewer “who’s doing this?” messages.

Example publish template

• Summary (1 paragraph): “From Mon–Wed chat: 2 decisions, 7 actions. Two items unblock the launch; three due by Friday.”
• Action table (CSV pasted or simple list): “Action — Owner — Due — Priority” items, 12 words max.
• Notes: Default owner applied where unnamed; reviewer confirmed dates.

Ready‑to‑send nudge templates

• “Quick nudge: ‘[Action]’ is due [Date]. Can you confirm or propose a new date?”
• “This one unblocks others: ‘[Action]’. If slipped, who should take it instead?”
• “We assumed you as owner for ‘[Action]’ based on thread start. OK to keep?”

Common mistakes and quick fixes

• Vague dates → Add the conversion rules in Pass 2 and stick to 5pm local.
• Duplicate tasks across threads → Deduplicate in Pass 2 and merge quotes so context isn’t lost.
• Unclaimed items linger → Use the default‑owner rule plus a 24‑hour nudge; reassign on silence.
• Too many columns → Publish a simple Action → Owner → Due table; keep the detailed JSON in the background.

One‑week rollout

1. Day 1: Build your People Dictionary (10 minutes) and run the 3‑pass prompts on a 24‑hour chat slice.
2. Day 2: Reviewer validates and pins the publish table; send nudges.
3. Day 3–4: Track corrections and note any recurring alias or timezone issues to update the dictionary.
4. Day 5: Re‑run on the next 48 hours; compare correction rate and on‑time completion.
5. Day 7: Lock your defaults (owner rule, date rules, nudge cadence) and repeat weekly.

Bottom line

Keep the extraction fast, the rules simple and the nudges consistent. With a tiny People Dictionary and the 3‑pass flow, your group chat turns into clear, owned actions — and a lot more finished work.

Quick win (do this in under 5 minutes): pick one client and calculate the effective hourly rate for a proposed retainer and a one-off project. Write the numbers down and you’ll already see which one pays you more per hour.

Why this matters

Retainers buy predictability and less admin. One-offs can pay more per hour but create feast-or-famine cashflow. The right choice depends on your hourly target, how much volatility you can tolerate, and the strategic value of the client.

What you’ll need

• Proposed monthly retainer and estimated monthly hours on retainer.
• One-off project fee and estimated project hours.
• Your target effective hourly rate and an overhead percentage (taxes, tools, admin) — 20% is a good starting point.
• A conservative churn assumption (e.g., client leaves after 6 months) and estimated gap time between one-off projects.

Step-by-step

1. Calculate raw hourly: fee ÷ hours for each option.
2. Adjust for overhead: multiply by (1 – overhead%). That gives your net effective hourly.
3. Model cashflow: retainers = monthly fee; one-offs = fee ÷ expected months between projects.
4. Model risk: for retainers, run scenarios where client lasts 3, 6, 12 months. For one-offs, model average months between projects.
5. Compare: net hourly, average monthly cash, volatility (variance), and strategic value (referrals, upsell).

Example

• Retainer: $3,000/month, 25 hours → raw $120/hr → net at 20% overhead = $96/hr → monthly cash = $3,000.
• One-off: $5,000/project, 40 hours → raw $125/hr → net at 20% overhead = $100/hr. If you expect one such project every 3 months, avg monthly cash = $5,000/3 = $1,667.
• Outcome: one-off slightly better hourly, retainer much better cash predictability and lower admin.

Common mistakes & fixes

• Thinking only by raw hourly — fix: always subtract overhead and non-billable time.
• Ignoring churn — fix: run pessimistic scenarios (client leaves early) and see impact.
• Under-pricing scope creep — fix: add scope caps and surge fees to retainers.

Action plan (do this next)

1. Run the two calculations for your current client leads today.
2. Decide your minimum acceptable net hourly and use it as a pricing floor.
3. Create simple contract guards: 3-month minimum, scope cap per month, and a notice period.

Copy-paste AI prompt (use this to get a ready-made comparison)

Compare these two client scenarios and recommend which to choose for someone who values steady income over occasional higher pay. Scenario A: Retainer $3,000/month, estimated 25 hours/month, assume 20% overhead, churn risk: 6 months. Scenario B: One-off $5,000, estimated 40 hours, assume 20% overhead, expected gap: average 3 months between similar projects. Calculate net effective hourly for both, average monthly cash, and show a risk-adjusted recommendation including suggested contract terms (minimum months, scope cap, surge fee) and three lines to use when negotiating the retainer.

Small systems beat big decisions. Run these numbers now, pick the option that meets your income and stress targets, then document the guardrails so pricing stops being guesswork.

Nice, Aaron — great practical checklist. I love the focus on centralising evidence and keeping a human-in-the-loop. Here’s a compact, do-first guide that adds a few concrete steps, naming conventions and prompts so a small business can act today.

Quick context: You don’t need a tech team to get audit-ready. Start small, prove the process, then automate. The aim: reduce scramble time, reduce auditor questions, and keep customer data safe.

What you’ll need

• One cloud folder (or encrypted USB) for evidence + a single inventory spreadsheet.
• List of required controls (access, backup, retention, incident response).
• An AI summarisation tool (cloud or private) and one trusted reviewer.
• Basic redaction tool (PDF/image) or instructions for manual redaction.

Step-by-step (do this today)

1. Create folder structure: Evidence / system-name / Logs | Policies | Backups | Access.
2. Inventory: open a spreadsheet with columns: File name, System, Control, Owner, Date, Summary, Link.
3. Collect: drop files in the right folder and add rows in the spreadsheet. Use consistent file names: 2025-11-22_POS_backup_log.pdf.
4. Run AI to summarise each file (see prompt below). Paste AI output into the Summary column.
5. Human review: owner checks summaries, redacts PII, adds comments in the spreadsheet.
6. Package: export the spreadsheet + zipped evidence folder. Produce an index.pdf with one-line notes per control.

Worked example — café

1. Folder: Evidence / POS / Logs contains pos_access_2025-11-01_to_15.csv
2. Spreadsheet row: pos_access_2025-11-01_to_15.csv — System: POS — Control: Access — Owner: Sam — Date: 2025-11-15.
3. AI summary: who accessed till, when, failed logins, and backup timestamp. Manager redacts email samples and confirms backup schedule screenshots.

Mistakes & fixes (fast wins)

• Missing timestamps — Fix: export logs with time range or add export metadata.
• Outdated policy documents — Fix: mark version and date; update critical ones first.
• Over-sharing PII to public AI — Fix: redact or run on a private model.
• Poor searchability — Fix: use consistent names and the spreadsheet index.

1-week action plan

1. Day 1: Create folder + spreadsheet, list systems.
2. Day 2–3: Collect top 5 evidence items (controls that matter most).
3. Day 4: Run AI summaries and map to controls.
4. Day 5: Human review and redaction.
5. Day 6: Compile bundle + index.
6. Day 7: 30-minute mock assessor Q&A using the AI prompt below.

Copy-paste AI prompts (use as-is)

“You are an expert compliance summariser. I will give you documents and logs. For each, produce a one-paragraph summary that states: the document name, what control it supports (e.g., access control, backup, retention), the key facts (who, when, what), and any gaps or anomalies to investigate. Output as: Document: [name] — Control: [control] — Summary: [one-paragraph] — Gaps: [list].”

Mock assessor prompt (use after packaging)

“You are an external assessor. Ask 8 focused questions an auditor would ask about this evidence bundle (access, backups, retention, incidents). For each question, explain why it matters and suggest what supporting file or proof would satisfy it.”

Start with the top 3 controls your business depends on. Gather, summarise, review — then show the bundle to an auditor or trusted peer.

5‑minute win: Keep your current pricing, but add one line under the CTA: “Cancel anytime. 30‑day money‑back.” Send a 50‑person email split in two subject lines and watch clicks to trial. Small risk‑reversal copy can lift response fast.

You’re right: tight tests beat long debates. Below is a simple playbook to write a pricing page with AI, run a value‑prop test, and turn reactions into decisions. No heavy tools. Just focus, contrast, and clear next steps.

What you need

• One‑line product summary and 3 customer benefits.
• Top 3 objections you hear (price, effort, trust, switching).
• Pricing skeleton: 2–3 tiers, short bullets, a guarantee line.
• Audience for 50–200 visits per variant (email, social, or a small ad spend).
• Basic tracking: UTM links and one CTA click event; a single exit question.
• One primary metric: click‑to‑trial or paid conversion. Pick one and commit.

How to build the pricing page (fast)

1. Draft the skeleton
   • Headline: outcome in 6–10 words. Avoid jargon.
   • Subheader: who it’s for + how it helps in 10–15 words.
   • Tiers: clear names (Starter, Pro, Business). 3 bullets each, 8–12 words.
   • CTA: one action (Start free trial). Add risk‑reversal under it.
   • Proof: one short line (e.g., “Trusted by 1,200 small teams”).
2. Generate messaging with AI using the prompt below. Ask for 3 variants in different tones (factual, aspirational, price‑first). Keep layout and prices unchanged.
3. Assemble two pages that differ only in headline, subheader, bullets, and one pricing cue (e.g., highlight “Best value” on middle tier). Freeze everything else.
4. Split traffic evenly for a fixed period (3–7 days if you have email/social; up to 14 days if low volume).
5. Measure the single metric and ask one exit question: “What stopped you from signing up today?”
6. Decide and iterate: keep the winner or the strongest insight. Next, test the next biggest assumption: price point, guarantee, or proof.

Premium trick: the “Objection‑to‑Bullet” move

• Take your top exit‑survey reason (e.g., “I’m worried about switching time”).
• Turn it into a bullet on the Pro tier: “Import your data in 10 minutes, guided.”
• Add a micro‑guarantee below CTA: “We migrate your data free if you’re stuck.”

Copy‑paste AI prompt (Pricing Page Generator)

Context: You are a pricing page copy assistant. Your job is to produce 3 contrasting messaging variants for a pricing page without changing layout or price points.

Product one‑liner: [1 sentence]. Customer: [who buys it]. Primary metric: [click‑to‑trial OR paid conversion]. Three core benefits: [B1], [B2], [B3]. Top 3 objections: [O1], [O2], [O3]. Tiers: [Basic], [Pro], [Business]. Guarantee: [e.g., Cancel anytime, 30‑day money‑back].

Deliver 3 variants. For each variant provide: 1) Headline (<10 words). 2) Subheader (10–15 words: who + outcome). 3) 3 bullets per tier (8–12 words, plain, specific). 4) One line of social proof. 5) A single pricing cue (e.g., “Best value” on Pro). Tones: Variant A factual, Variant B aspirational, Variant C price‑first. Use customer language, avoid buzzwords.

Optional prompts to speed iteration

• Objection Synthesizer: “Summarize these exit comments into top 3 objections. Rewrite my Pro tier bullets to directly counter them. Keep bullets under 12 words. Here are the comments: [paste].”
• Price Cue Tweaker: “Given these prices [list], propose 2 ‘best value’ label options and 1 subtle decoy (keeps integrity) to nudge toward Pro. No changes to actual prices.”

Example (use as a template)

Product: InvoiceFlow — simple invoicing for home‑services contractors. Benefits: get paid faster, fewer admin hours, fewer payment errors. Objections: learning curve, switching from spreadsheets, fees.

• Variant A (factual)
  • Headline: “Send invoices. Get paid on time.”
  • Subheader: “Built for contractors who want cash flow without paperwork.”
  • Pro bullets: “One‑click estimates to invoices,” “Auto reminders reduce late pays,” “Import spreadsheets in minutes.”
• Variant B (aspirational)
  • Headline: “Finish jobs, not paperwork.”
  • Subheader: “Less admin, more billable hours for growing crews.”
  • Pro bullets: “Templates that match your jobs,” “Mobile approvals on‑site,” “Instant deposits with trusted processors.”

How to read early results (directional, not final)

• If A has 100 visits and 10 trial clicks (10%) and B has 100 visits and 13 trial clicks (13%), that’s a useful signal. Extend to 200–300 visits before declaring a keeper.
• When in doubt, let objections decide your next test, not your taste.

Common mistakes and quick fixes

• Too many differences. Fix: lock layout and price; change only message and one price cue.
• Vague bullets. Fix: use numbers, actions, and outcomes: “Automated reminders cut late pays 20–30%.”
• Ambiguous plan names. Fix: Starter, Pro, Business. Clear and calm.
• Hiding risk‑reversal. Fix: add it under the CTA, not in the footer.
• No proof. Fix: add one credibility line or a short testimonial.
• Mixing monthly/yearly by default. Fix: pick one default. Test the toggle later.

Simple setup tips (non‑technical)

• Create two URLs: /pricing‑A and /pricing‑B. Paste each variant’s copy.
• Use UTM tags on your links (e.g., ?utm_source=email&utm_campaign=pricingtest&utm_content=A or B).
• Track one event: CTA click. Name it clearly: pricing_cta_click.
• Ask one exit question on both pages. Keep it optional and short.

7‑day action plan

1. Day 1: Write your one‑liner, 3 benefits, 3 objections. Draft the skeleton.
2. Day 2: Run the Pricing Page Generator prompt. Pick 2 variants.
3. Day 3: Build /pricing‑A and /pricing‑B. Add UTMs and the CTA event.
4. Day 4: Send traffic. Split evenly. Start the exit question.
5. Day 5: Check directional clicks. Skim exit comments. Apply one Objection‑to‑Bullet change to the leader.
6. Day 6–7: Roll another small audience. If the lift holds, keep the winner and plan a price cue or guarantee test next.

What to expect

• Early tests give direction, not certainty. That’s enough to act.
• Messaging wins (headline, bullets, risk‑reversal) often create a noticeable lift before you touch prices.
• Use those wins to justify bigger tests: price points, term discounts, or guarantees.

Remember: Tight contrast, one metric, short feedback loops. Ship the next test while the last results are still warm.

Quick win: In under 5 minutes, ask AI for three headline options and swap your current headline on the pricing page. Send a small email blast and watch clicks — that single change often moves the needle.

Nice point in your plan — keeping the test tight (one metric, 2–3 variants) matters. Here’s a compact add-on that helps non-technical founders run the experiment, interpret results, and iterate without overthinking the tech.

What you’ll need

• A one-line product summary.
• A 1-page pricing skeleton (headline, 2–3 tiers, short bullets).
• Audience for 50–200 visits per variant (email list, social, or small ad spend).
• Basic tracking (UTMs, a CTA click event) and one exit survey question.

Step-by-step (do this)

1. Choose one metric: click-to-trial OR paid conversion. No more.
2. Run the AI prompt below to generate 3 headline/subheader/bullets in three tones. Pick 2 variants to start.
3. Build two pricing pages that differ only in messaging (headline, subheader, bullets) and one pricing cue (e.g., emphasize price vs features). Keep layout and CTA identical.
4. Split traffic evenly for a fixed window (3–7 days for email/social; 7–14 days if low traffic).
5. Measure the metric, collect an exit survey answer: “What stopped you from signing up?”
6. Pick the direction (winner or insights), then iterate on the next biggest assumption (price point, guarantee, or proof).

Example (copy-and-try)

Product one-liner: ProjectPages — simple project tracking for small design teams. Benefits: faster handoffs, fewer status meetings, clearer deadlines.

• Headline (aspirational): “Deliver projects on time, every time.”
• Subheader: “Built for small teams who need clarity without the noise.”
• Bullets (Pro tier): “Visual timelines, client-ready reports, integrations that just work.”

Common mistakes & fixes

• Testing too many variants — Fix: start with 2, then scale to 3 after a clear winner.
• Changing design during test — Fix: lock layout; only change messaging or a single price cue.
• Ignoring qualitative feedback — Fix: read the exit responses; they explain the ‘why’ behind numbers.

7-day action plan

1. Day 1: Write one-line product summary and 3 benefits.
2. Day 2: Run AI prompt, choose 2 messaging variants.
3. Day 3: Build pages, add tracking and exit survey.
4. Day 4: Send traffic (email/social/ads).
5. Day 5–7: Collect results, read exit feedback, pick the winner and plan next test.

Copy-paste AI prompt (use verbatim)

Product one-liner: [Paste a single-sentence description]. Customer persona: [Who buys it]. Primary metric: [click-to-trial OR paid conversion]. Three core benefits: [Benefit 1], [Benefit 2], [Benefit 3].

Generate 3 headline + subheader combinations. For each combination provide 3 tones: factual, aspirational, price-focused. For each tone, give 3 short bullets for each pricing tier (Basic, Pro, Premium) and one line of social proof. Keep headlines under 10 words, subheaders 10–15 words, bullets 8–12 words.

Closing reminder: Start small, measure one thing, learn fast. A single better headline or one clearer bullet often gives the confidence and lift you need to justify bigger pricing changes.

Here’s how to make AI run your AP prep like a small, predictable project — even on busy weeks. You’ll get a focused daily plan, quick micro-quizzes with answers, and a Sunday dashboard that tells you if the time is working.

High-value insight: Don’t ask for “more study.” Ask AI to optimize three levers — adherence, accuracy, speed — and to adapt next week’s plan to the two biggest error patterns. That is how scores move without burnout.

What you’ll need

• AP subject, weeks left, current vs target score
• Time caps (weekday and weekend), preferred study window
• Top 2–3 weak topics or question types
• Timer, scratch paper, past exams or a question bank, and a simple daily log

Turn it into a weekly system (simple loop)

1. Mini-diagnostic (Day 0, 30 minutes): 10–15 MCQs or 1 FRQ. Capture: accuracy %, minutes, top two errors.
2. Plan (AI, 5 minutes): Get a 2-week plan under your time caps. Require daily: short lesson, active practice, 5–10 minute spaced review.
3. Execute (35–45 minutes/day): Follow the plan, end with a one-line log: minutes, accuracy, one takeaway.
4. Adjust weekly (30–45 minutes): Timed section on the weekend; re-run AI with your KPIs to refocus next week.

Copy-paste AI prompt — Daily Coach (refined)

“Act as an AP [SUBJECT] coach. Constraints: weekdays [MINUTES] minutes, weekends [MINUTES_WEEKEND] minutes, exam in [WEEKS_LEFT] weeks. Current: mock [CURRENT]/5, target [TARGET]/5. Weak topics: [TOPICS]. Yesterday’s KPIs: adherence [0/1 or 1/1], quiz accuracy [Y%], timed pace [Z Q/min], top error [ERROR CODE: concept/formula/misread/haste]. Today: give a plan under the time cap with 1) a 20–25 min focused lesson on the highest-leverage weak topic, 2) a 10–15 min active practice (5–8 specific questions or an FRQ prompt), 3) a 5–10 min spaced review list. Include: a 5-question micro-quiz with expected answers, a 30-second parent summary, and any materials needed. Keep it practical and printable.”

Copy-paste AI prompt — Weekly Adapt (dashboard + plan)

“We completed Week [N] for AP [SUBJECT]. KPIs: adherence [DAYS_DONE/DAYS_PLANNED], average quiz accuracy [AVG%], timed pace [QPM], FRQ rubric avg [POINTS], top 2 recurring errors [ERROR1/ERROR2]. Exam in [WEEKS_LEFT] weeks. Create a 7-day plan under [MINUTES] weekdays and [MINUTES_WEEKEND] weekends. Prioritize the errors above, include 1 timed section, and propose 2 drills that raise FRQ rubric points. Deliver: day-by-day tasks with exact question counts, a spaced-review schedule (2/4/7-day resurfacing), and a one-paragraph parent dashboard for next week’s focus.”

Insider trick: error codes make adaptation instant

• E1 Concept (didn’t know idea)
• E2 Formula/Facts (memory gap)
• E3 Misread (question parsing)
• E4 Haste (careless)
• E5 Method (wrong approach)

Tag each miss with an error code in the log. Ask AI to prioritize the most frequent two codes next week.

Worked example (AP Biology, 5 weeks left)

• Current 3/5; target 4/5. 35 minutes weekdays, 70 minutes weekend. Weak in cellular respiration and genetics crosses.
• Day 1 (35 min): 20-min focused lesson on electron transport + 10-min mixed MCQs (6 Qs) + 5-min flashcards (NADH vs FADH2, ATP yield). Micro-quiz answer key included.
• Day 3 (35 min): 20-min lesson on Punnett squares with linked genes + 10-min practice (2 short FRQ-style prompts) + 5-min spaced review (terms from Day 1, 2).
• Weekend (70–80 min): 45-min timed MCQ block + 15-min review of misses by error code + 10-min rewrite of one FRQ outline.

Quick micro-quiz example with answers (Biology)

• 1) Main function of the ETC? Answer: Create a proton gradient to drive ATP synthase.
• 2) Final electron acceptor in aerobic respiration? Answer: Oxygen (forms water).
• 3) Heterozygous dihybrid cross independent assortment expected ratio? Answer: 9:3:3:1.
• 4) Which yields more ATP: NADH or FADH2? Answer: NADH.
• 5) One cause of reduced ATP yield in mitochondria? Answer: Uncoupling/leaky inner membrane or inhibited complexes.

Optional prompts you’ll use often

• Diagnostic Maker: “Create a 12-question diagnostic for AP [SUBJECT] focusing on [WEAK TOPICS]. Mix 8 MCQs and 1 FRQ. Provide an answer key and tag each item with the skill tested. Time cap 25 minutes.”
• Catch-Up Day (missed yesterday): “We missed a day. Compress today for AP [SUBJECT] into [MINUTES] minutes. Keep one weakness drill, one 5-item quiz with answers, and a 2/4/7-day review list. Defer anything nonessential.”
• FRQ Rubric Booster: “Design two 15-minute drills that target the most commonly missed FRQ rubric rows for AP [SUBJECT] on [TOPIC]. Include a sample response outline and a quick self-scoring checklist.”

What to expect

• A realistic daily checklist under your time cap.
• Quizzes with answers so you can score in minutes.
• A one-paragraph parent dashboard each Sunday: minutes, accuracy trend, pace, biggest error, next focus.

Common mistakes and how to fix them

• Too much content, not enough reps. Fix: Limit lessons to 20–25 minutes; move fast to practice.
• No error tagging. Fix: Use E1–E5. Ask AI to target the top two codes each week.
• Skipping spaced review. Fix: Enforce 2/4/7-day resurfacing; ask AI to list exact items.
• Unclear parent visibility. Fix: Require a daily 30-second summary and a Sunday dashboard line.

One-week action plan

1. Today (20–30 min): Run a 12-question diagnostic. Log accuracy, pace, and two error codes.
2. Then (5 min): Paste the Daily Coach prompt with your numbers. Screenshot the plan.
3. Mon–Thu: 35–45 minutes/day. End with a one-line log (time, accuracy, takeaway, top error code).
4. Weekend (70–90 min): Timed section at exam hour; review misses by error code; rewrite one FRQ outline.
5. Sunday (15 min): Paste the Weekly Adapt prompt with your KPIs. Lock next week.

Reminder: Consistency beats intensity. Keep the time cap, tag every error, and let AI adapt the plan. Small, measured wins stack into big score gains.

Love the 5-minute “who/what/when” scan — that’s the lever that turns chaos into a brief. Let’s add one upgrade: run a second, short AI pass that links every decision to one immediate next action and flags any missing owner/date. That “checksum” keeps nothing slipping through.

Try this right now (under 5 minutes)

• Paste your 3-column list (Owner — Task/Decision — Due) into the prompt below. You’ll get a clean one-page brief, a gaps list (owners/dates missing), and a confidence flag so you know where to double-check.

What you’ll need

• Your quick extract (Owner — Task/Decision — Due)
• Meeting date, purpose, attendees
• Any key references (slides, proposal names) — optional

Two-pass method (simple and reliable)

1. Pass 1: Extract — you already nailed this: scan for people and dates; mark unclear items as TBD.
2. Pass 2: Upgrade — use AI to: (a) format a one-page brief, (b) tie each decision to one next action, (c) list ambiguities to resolve, and (d) add a short executive summary at the top.

Copy-paste AI prompt (use as-is)

“You are an assistant that turns a rough extract of meeting notes into a one-page executive brief with a reliability check. From the content I provide, produce the following sections in bullets: 1) Purpose (one line), 2) Top 3 Takeaways (plain English), 3) Decisions (each with Owner and Due), 4) Actions (each with Owner, Task, Due), 5) Decision→Action Links (map each decision to the first next action), 6) Open Issues, 7) Gaps & Ambiguities (list any missing or unclear owners/dates and quote the source text in brackets), 8) Confidence flags (High/Medium/Low for each decision/action). If an owner or date is unclear, write TBD and add it to Gaps & Ambiguities with the exact snippet. Keep the brief scannable; no long paragraphs.”

Insider trick: Decision→Action link

• Every decision should trigger one “first next step.” If there isn’t one, the decision isn’t ready. The link makes follow-up automatic and measurable.

Optional smart defaults (use carefully)

• If due dates are missing, set a suggested default of 5 business days after the meeting (mark as Suggested), then confirm with the owner.
• If multiple people are mentioned, assign a single point of accountability and list others as helpers.

What good output looks like

• Executive-first: one-line purpose and three takeaways before anything else.
• Owners and dates everywhere: 95%+ of decisions and actions have both.
• Ambiguity log: a small list of items to confirm — not buried in the brief.

Worked example (short)

• Purpose: Align on Q2 launch scope and responsibilities.
• Top 3 Takeaways: Budget capped at $50k; launch window pulled forward by two weeks; vendor shortlist narrowed to A and C.
• Decisions:
  • Approve $50k budget — Owner: CFO — Due: 2025-06-02 — Confidence: High
  • Target launch: week of 2025-07-15 — Owner: Marketing Director — Due: 2025-05-30 to update plan — Confidence: Medium
• Actions:
  • Contact Vendor A for final quote — Owner: PM — Due: 2025-05-28 — Confidence: Medium
  • Revise project timeline to reflect new launch window — Owner: Ops Lead — Due: 2025-05-30 — Confidence: High
• Decision→Action Links:
  • Budget approved → Request updated quotes and lock PO draft (Owner: PM — Due: 2025-05-29)
  • Launch window set → Update timeline and stakeholder comms plan (Owner: Ops Lead — Due: 2025-05-30)
• Open Issues: Legal review of vendor terms (Owner: Legal — Due: TBD)
• Gaps & Ambiguities:
  • Owner unclear for stakeholder comms — [“We’ll need to tell Sales next week”]
  • Vendor selection date not stated — [“Decide after we see A’s final number”]

Common mistakes and quick fixes

• Unlinked decisions — Fix: require one next action per decision; no exceptions.
• Too many takeaways — Fix: cap at three; move the rest to context.
• AI guessing owners — Fix: make the model quote the source and mark “Proposed” if inferred; confirm in validation.
• Wall of text — Fix: bullets only; one line per item; verbs first in actions.

What to expect (realistic)

• First runs: 20–40 minutes end-to-end for a 60–90 minute meeting.
• After 3–5 briefs: you’ll get to 15–25 minutes with a stable template and faster confirmations.
• Quality signal: fewer “Can you clarify?” emails and faster starts on tasks.

1-week action plan

1. Today: Do the 5-minute extract on your latest meeting.
2. Tomorrow: Run the prompt above; send the brief’s Decisions/Actions only for a quick “Confirm/Correct.”
3. Midweek: Add the Decision→Action link rule to your template.
4. Late week: Review two briefs — check % with owner+due; aim for 95%+.
5. End of week: Save your best brief as the team’s “gold standard” and reuse it.

Pro tip: standard words reduce confusion

• Use action verbs: Approve, Decide, Draft, Send, Review, Ship.
• State dates in ISO format (YYYY-MM-DD) so they’re unambiguous.

Pragmatic path forward: keep your 5-minute extract, add the one-pass AI upgrade with the gaps list, and link every decision to a first next action. That’s how messy notes become a reliable, polished brief — fast.