aaron

Nice focus on rotations — that’s the hardest part to get right. You want fairness and fewer reminders. Here’s a practical, step-by-step way to use AI to run chore assignments, reminders, and simple accountability without needing to be technical.

Problem: Household chores fall through the cracks, distribution feels unfair, and reminders become nagging.

Why it matters: Clean, predictable routines reduce conflict, save time, and free mental energy for higher-value activities.

Lesson from practice: Start with a simple rotation, track completion, and let AI handle scheduling and nudges. Don’t automate everything at once — iterate weekly.

1. What you’ll need
   • List of chores + estimated time per chore
   • Names of household members and constraints (availability, preferences)
   • A shared calendar (Google/Apple) or simple task app (Reminders, Todoist)
   • Access to an AI assistant (chatbox on phone or email-integrated assistant)
2. How to set up (step-by-step)
   1. Create a single chores spreadsheet or document: chore, frequency, duration, priority.
   2. Decide rotation rules: equal time, alternating weekly, or skills-based.
   3. Use the AI prompt below to generate an initial 4-week rotation and role assignments.
   4. Import assignments to your shared calendar with reminders (one notification on assignment day, one the morning it’s due).
   5. Set a weekly 10-minute check-in where the household marks completed items and flags issues.
   6. Feed completion results back to the AI weekly to refine the schedule.

Copy-paste AI prompt (use as-is)

“You are a household assistant. Create a 4-week rotating chore schedule for 4 adults: Alice, Ben, Carla, Dan. Chores: dishes (daily, 15 min), vacuum (twice weekly, 30 min), laundry (twice weekly, 45 min), trash (weekly, 10 min), groceries (weekly, 60 min). Prioritize equal total time per person each week, avoid assigning heavy tasks back-to-back, and note any preferences: Alice dislikes trash, Ben prefers groceries. Output a readable weekly assignment table and a one-paragraph explanation of rules.”

What to expect & KPIs

• Initial setup: 1–2 hours. Ongoing maintenance: 10 minutes/week.
• Metrics to track: completion rate (% tasks done on time), average time per person/week, complaints per week, and satisfaction score (1–5) after check-ins.

Common mistakes & fixes

• Overcomplicating the system — fix: reduce to top 10 chores and core rotation.
• No accountability — fix: calendar invites + a quick confirmation reply required.
• Not iterating — fix: review metrics weekly and let AI rebalance.

1-week action plan

1. Day 1: List chores and people (30 min).
2. Day 2: Run the AI prompt and generate a 4-week schedule (15 min).
3. Day 3: Import week 1 to shared calendar, set reminders (15–30 min).
4. Day 4–7: Follow schedule; track completion in a simple note or app.
5. End of week: 10-minute review, record metrics, ask AI to rebalance if needed.

Your move.

Good question. You want photo‑real lifestyle scenes with your actual product, fast enough for marketing cycles. Yes—doable. The win is speed-to-creative and volume, if you follow a tight workflow.

The challenge: AI can make gorgeous scenes, but it often distorts labels, proportions, and lighting. Unchecked, that hurts trust and wastes ad spend.

Why it matters: Fresh, on-brand lifestyle creative typically lifts thumb‑stop rate and CTR, reduces CPC, and slows creative fatigue. Expect to ship 20–50 variants in a day once the pipeline is set.

What you’ll need:

• 3–5 clean product photos on plain background (multiple angles; neutral lighting)
• Your brand style guide (colors, tone, do/don’t props)
• One of: Midjourney, SDXL (Stable Diffusion) with ControlNet/IP-Adapter, or Adobe Firefly; plus Photoshop (or similar) for label cleanup
• A simple QA checklist (proportions, label accuracy, color match, legal)
• Ad testing setup (3–5 variants per audience, UTM tracking)

How to do it (step-by-step):

1. Prep the product: Shoot front/45°/side/top. Use soft, even light. Include a color card once; use that to correct all shots. Export PNG with transparent background.
2. Choose workflow: Fastest: Firefly or Photoshop Generative Fill for comping the real product. Most control: SDXL + ControlNet + IP‑Adapter to lock the product in-place. Midjourney: great look, but use “image prompt + inpainting” to preserve your product.
3. Generate the scene (no product yet): Prompt a lifestyle background that fits your audience, lens, and lighting. Keep it clean—leave space where the product will sit.
4. Insert the real product: Use inpainting/Generative Fill or ControlNet’s reference to place the cutout product. Match angle and scale; nudge until shadows align.
5. Match light/shadow: Add a soft drop shadow (1–3% opacity, Gaussian blur) and a subtle ground reflection if glossy. Ask AI to “harmonize lighting” around the product.
6. Fix labels and color: If text warps, paste the original label as a top layer and warp it manually. Verify brand color values. Remove any extra logos AI added.
7. Upscale and polish: Run 2–4× upscale. Check edges, glass reflections, and hand/fabric details. Export master at 3000–4000 px on the long side.
8. Variant quickly: Swap background mood, camera, and props. Generate 10–20 variants per concept; shortlist 3–5 per audience.
9. QA + compliance: No false claims, no trademarked backgrounds, no fake endorsements. Add “AI‑generated scene” tag for internal tracking even if you don’t disclose externally.
10. Launch and measure: Test 3 variants head‑to‑head against your current best creative. Kill losers in 48–72 hours; roll learnings into next batch.

Copy‑paste prompt templates (use with your tool’s image + prompt feature and your product photo as reference):

• Background scene: “Create a realistic [SETTING] lifestyle scene for [TARGET AUDIENCE], natural textures, minimal clutter, soft morning window light, shot on a 50mm lens at f/2.8, true‑to‑life colors, commercial photography, space in the foreground to place a [PRODUCT], no logos, no people cropped awkwardly.”
• Place my product: “Using the attached product photo, place the [PRODUCT] naturally on the surface, keep exact proportions and label details, match perspective and lighting, add a soft grounded shadow, no extra logos, no distortion, photo‑real, advertising quality.”
• Variant generator: “Create 6 subtle variations of this scene changing only props and lighting temperature. Maintain camera angle, keep [PRODUCT] identical, ensure brand palette [COLORS], avoid busy backgrounds.”

Insider tricks:

• Prompt like a photographer: lens, aperture, time of day, surface material. It stabilizes realism.
• Lock the product: Use ControlNet/IP‑Adapter or Photoshop’s Generative Fill with your product layer on top to avoid warped labels.
• Shadow plate: Generate a clean surface with only shadows, then multiply‑blend under your product for instant realism.
• Reflections sell it: For glossy items, duplicate the product, flip vertically, blur 2–4 px, drop opacity to 10–20% for a believable table reflection.

What to expect:

• Per scene: 5–15 minutes from prompt to final, once you’ve built your template
• Pass rate: 60–80% usable after quick fixes; labels and hands (if any) will need attention
• Cost: cents to low dollars per variant depending on tool

Metrics to track:

• Thumb‑stop rate (3‑second view) and CTR vs. your current best
• CPC and CPA/ROAS trend over first 3–5 days
• Creative fatigue: performance decay by day; retire when CTR drops >30% from day‑1
• Production efficiency: variants/hour and cost/asset

Common mistakes and quick fixes:

• Wrong scale or angle: Overlay a perspective grid; resize until edges align with the grid.
• Color off-brand: Apply a LUT or manual HSL to match your style guide.
• Busy scenes: Remove props; add negative prompt terms like “minimal, no clutter, no text, no watermark.”
• Soft labels: Paste original label at 95–98% opacity; add a tiny grain to blend.
• Legal risk: Avoid recognizable private property, artwork, or brand marks in the scene.

1‑week action plan:

1. Day 1: Shoot product angles on white; color‑correct and export PNGs.
2. Day 2: Define 3 lifestyle scenarios tied to your top audiences; write prompts using the templates above.
3. Day 3: Generate 30–50 scenes; shortlist 12; fix labels/colors; upscale.
4. Day 4: Produce 3 ad variants per scenario (headline/body/call‑to‑action constant; image only changes). Tag with UTMs.
5. Day 5–6: Launch A/B/C tests. Monitor CTR, CPC, and early CPA. Pause bottom third.
6. Day 7: Roll best performer into 10 more variants; archive learnings (lens, light, props) as your “creative recipe.”

Bottom line: Yes, AI can deliver photo‑real lifestyle scenes that move the numbers—as long as you anchor the product in reality, control light/scale, and enforce QA. Start with one product, three scenes, and disciplined testing.

Your move.

Make a poster that sells your short film — not just a pretty image.

Problem: You want cinematic, festival-grade poster art but you’re not a designer and don’t want to waste time or money on trial-and-error.

Why it matters: A strong poster increases festival invites, social traction, and viewer curiosity. It’s the first sales asset for buyers, programmers and press.

Lesson from working with filmmakers: treat AI as a creative engine, not a finish line. Use clear briefs, bake in constraints (aspect ratio, resolution, typography), and set measurable goals (audience preference, print quality).

1. What you’ll need
   • Reference images or a moodboard (3–6 frames showing tone, lighting, color).
   • An image-generation tool (Midjourney, Stable Diffusion web UI, or an AI art service) and a simple editor (Photoshop/Canva).
   • Film title, tagline, and preferred aspect ratio: 27×40 in / 2:3 or 1200×1800 px for web.
2. Step-by-step
   1. Define the core: one emotion + one visual hook (example: “lonely lighthouse at dusk; silhouette; cinematic drizzle”).
   2. Run 8–12 prompt variants using different focal points, lighting, and color palettes.
   3. Pick the top 3 outputs, upscale to print resolution (300 DPI) and composite in an editor to add title, credits, and legal text.
   4. Test on two surfaces: social-size (1200×1800) and print mockup (3000–4500 px on long edge).

Copy‑paste prompt — primary (drop film details into brackets):

“Poster for a short film titled ‘[FILM TITLE]’. Genre: [DRAMA / THRILLER / SCI-FI]. Central image: [brief hook, e.g. ‘silhouette of a woman on cliff with lighthouse behind’]. Mood: cinematic, melancholic, high-contrast, warm highlights and teal shadows. Composition: single dominant focal point, rule-of-thirds, dramatic backlighting, soft film grain. Color palette: deep teal, warm amber accents. Camera: 35mm lens, shallow depth of field, rim lighting. Typography space at bottom center for title and tagline. Style references: vintage movie poster, analog film texture, subtle vignetting. Output: high-resolution, 2:3 aspect ratio, print-ready (300 DPI).”

Two prompt variants

• Minimal variant: “Minimal cinematic poster: single silhouette, monochrome teal-orange palette, bold negative space, title area clear, 2:3, 300 DPI.”
• Illustrative variant: “Painterly cinematic poster, dramatic sky, ethereal light rays, textured brush strokes, warm highlights, 2:3, high detail.”

Metrics to track

• Time-to-final: hours from first prompt to approved poster.
• Iterations: number of prompts until top 3 identified (target: ≤12).
• Audience preference: % positive feedback from a small panel (target: ≥70%).
• Technical: final file at 300 DPI, long edge ≥3000 px.

Common mistakes & fixes

• Overloaded prompts → simplify to one mood + one visual hook.
• Low resolution → always upscale or generate at high-res default.
• Unreadable title area → reserve clear negative space in prompt and composite text manually.
• Uncanny faces → avoid close-up faces in prompts or use reference photos for compositing.

1-week action plan

1. Day 1: Create moodboard and define one-sentence creative brief.
2. Day 2: Run 8–12 prompt variants and collect results.
3. Day 3: Select top 3, upscale, and composite title treatments.
4. Day 4: Small-panel review (5 people) and collect feedback.
5. Day 5: Final refinements: color grade, type, legal text; prepare print files.
6. Day 6: Produce social-size assets and one physical print test.
7. Day 7: Deploy to festival submissions, social, and press kit.

Keep iterations focused and always move to manual compositing for final type and legal blocks — that’s where perception and professionalism land.

Your move.

Quick win (under 5 minutes): Paste this prompt at the end — get a two-sentence hook for your About page you can publish immediately.

Nice title — clear, outcome-focused. That’s exactly the use case: short, repeatable prompts that produce measurable copy.

The problem: founders either produce bland bios or over-share irrelevant detail. The result: low engagement, weak trust signals, few leads.

Why it matters: your founder story is often the first emotional connection. Done right it increases time-on-page, conversions and demo/email signups. Done wrong it kills credibility.

Hard lesson I’ve used with clients: a concise, benefit-led narrative plus one human detail outperforms long chronological histories every time. Readers want relevance first, story second.

Step-by-step: What you’ll need

1. One clear outcome (e.g., more email signups, demo requests, trust for partnerships).
2. 3 facts: role/title, problem solved, one human detail (hobby, origin, motivation).
3. An AI tool or editor to refine text.

How to do it (copy, paste, publish)

1. Use this primary prompt (copy-paste into your AI):
   “Write a 150–200 word founder story for an About page. Start with a 1-line hook that states the specific problem the founder solves. Include three short paragraphs: (1) what they do and why it matters to customers, (2) one personal detail that builds trust, (3) a one-sentence call-to-action inviting readers to sign up or request a demo. Tone: confident, warm, business-focused. Keep plain language for a non-technical audience.”
2. Ask for a 2-sentence version for the page header: “Create a 2-sentence headline and subhead that summarize the founder’s mission and benefit.”
3. Swap in your specifics (role, problem, human detail). Review for clarity, remove jargon, publish.

What to expect: a publish-ready About paragraph plus a short header. Time: 5–20 minutes. Outcome: faster edits, consistent messaging, clearer CTAs.

Metrics to track

• Time on page (aim +20% over baseline)
• Click-throughs on About CTAs (email/signup/demo) — track weekly
• Bounce rate from About page (aim for -10% within 2 weeks)
• Leads attributed to About page (qualitative: mentions in sales calls)

Common mistakes & fixes

• Too much history — Fix: compress chronology into one sentence focused on benefit.
• Vague outcomes — Fix: replace “help companies” with “reduce IT costs by X%” or “cut onboarding time by Y days.”
• No CTA — Fix: add one clear next step (email, demo, signup) in one sentence.

1-week action plan

1. Day 1: Run the primary prompt, pick one version to publish (5–20 minutes).
2. Day 2–3: A/B test header vs previous header (simple headline swap).
3. Day 4–5: Monitor metrics, collect qualitative feedback from 3 colleagues/customers.
4. Day 6–7: Iterate copy based on metrics and feedback, re-publish.

AI prompt you can copy-paste right now

“Write a 150–200 word founder story for an About page. Start with a 1-line hook that states the specific problem the founder solves. Include three short paragraphs: (1) what they do and why it matters to customers, (2) one personal detail that builds trust, (3) a one-sentence call-to-action inviting readers to sign up or request a demo. Tone: confident, warm, business-focused. Keep plain language for a non-technical audience.”

Your move.

Smart question: keeping your stack to AI tools that already plug into Zapier prevents dead ends and makes results measurable fast.

What to connect (and when)

• General LLMs (write, summarize, classify): OpenAI (ChatGPT/GPT-4 family), Anthropic Claude, Google Gemini, Microsoft Azure OpenAI, Cohere. Pick one primary; keep a cheaper “fast” model as backup.
• Transcription/meeting notes: Fireflies.ai, Fathom, Avoma, AssemblyAI, Otter.ai. Use when voice/video is involved.
• Document/receipt extraction (OCR + AI): Nanonets, Rossum, Veryfi, Mindee, Docsumo. Use when you need structured fields from PDFs, invoices, IDs.
• Knowledge bases (context for good answers): Notion, Confluence, Google Drive/Docs. Use Zapier search actions to pull relevant notes and feed to the LLM.
• Core admin apps to automate around: Gmail/Outlook, Google Calendar/Outlook Calendar, Slack/Teams, Google Sheets/Airtable, HubSpot/Salesforce, Asana/Trello.

Why this matters

• Consolidate 80% of admin: email triage, meeting prep, note summaries, CRM updates, document filing.
• Trackable savings: 5–10 hours/week within 30 days, with error rates you can measure.

Do / Do not

• Do start with one general LLM and one specialist (transcription or document extraction) to avoid bloat.
• Do enforce structured outputs (JSON or bullet fields) so Zapier can route cleanly.
• Do store your house style and rules in Zapier Storage/Tables and inject them into prompts.
• Do draft emails/events first; require manual approval before sending live for the first 2 weeks.
• Do trim inputs with Formatter (e.g., top/bottom 150 words) to cut costs and noise.
• Don’t send sensitive data to third-party AIs without redaction. Mask names, amounts, IDs.
• Don’t rely on one-shot prompts. Chain: classify → summarize → act.
• Don’t skip logging. Write every AI action to a Sheet for audit and learning.

Worked example: Inbox → Draft reply → CRM update

1. Trigger: New email in Gmail with label “Leads.”
2. Search context: Find related contact in HubSpot (or Salesforce). Pull last activity notes from Notion.
3. Classify + summarize (LLM): Use OpenAI/Claude/Gemini to produce: intent, urgency, contact role, 3-bullet summary, and next action.
4. Path: If intent = “book meeting,” create a calendar invite draft; if “pricing,” attach your pricing one-pager link; if “support,” create a ticket.
5. Draft reply: LLM writes a 120–180 word email in your tone with 3 short options for subject lines.
6. Approve: Send draft to Slack for one-click approve/edit, then Gmail sends.
7. Log: Update CRM, add summary to contact, append line to Google Sheet with outcome, time saved (estimate), and model cost.

Copy-paste prompt (use in your LLM step)

Paste into an OpenAI/Claude/Gemini action. Replace bracketed parts with your details.

“You are my executive admin. Follow these rules: 1) Output , , , , , and . 2) Keep the draft 120–180 words, warm-professional, no jargon, use British English, and offer 3 subject lines. 3) If missing info, ask 1 clarifying question at the end of the draft. 4) Never invent facts; only use provided context. Input starts now. COMPANY STYLE: [paste your style/tone bullets]. CONTEXT: [recent CRM notes or Notion page text]. EMAIL: [paste the incoming email body].”

What you’ll need

• Zapier account (multi-step Zaps enabled).
• Accounts for your chosen LLM (OpenAI/Claude/Gemini) and any specialist tools (e.g., Fireflies.ai, Nanonets).
• Access to Gmail/Outlook, Calendar, CRM, and a Sheet for logging.

How to set it up (10 steps)

1. Create labels/folders to filter target emails (e.g., “Leads,” “Vendors,” “Internal”).
2. Build a Zap: Trigger = New Email in Label.
3. Add Formatter steps to trim signatures/threads (keep top and most recent bottom 150 words).
4. Search CRM for contact; fetch last note. If none, create contact.
5. Pull company style/tone from Zapier Storage/Tables (editable without touching the Zap).
6. LLM step with the prompt above; request structured fields.
7. Paths: route by to Calendar/Docs/Helpdesk actions.
8. Draft email in Gmail (don’t auto-send yet). Push preview to Slack for approval.
9. On approval, send email and update CRM with the summary and next action.
10. Log to Google Sheet: timestamp, intent, time saved (minutes), model used, token/cost estimate, manual edits (yes/no).

Metrics to track

• Time saved per item (baseline vs. automated).
• Email reply time (median) and response rate.
• Error rate: % of drafts needing major edits.
• Model cost per email, per meeting, per document.
• Meetings booked and no-show rate (post-automation).

Common mistakes & fast fixes

• Messy outputs. Fix: demand JSON-like fields and validate with Formatter before routing.
• Runaway token costs. Fix: summarize context to 300–500 words before the main prompt; prefer “mini/flash” models for classification.
• Hallucinated facts. Fix: include “never invent; if missing, ask 1 question” in prompt; compare against CRM fields.
• Too many tools. Fix: cap at 1 LLM + 1 specialist until KPIs improve for 2 consecutive weeks.

1-week action plan

• Day 1: Pick your primary LLM and one specialist (transcription or document extraction). Connect accounts in Zapier.
• Day 2: Implement the Inbox → Draft → CRM Zap. Keep manual approval on.
• Day 3: Add Calendar path for “book meeting.”
• Day 4: Add a Sheet log and a daily summary to Slack.
• Day 5: Roll out a second Zap: receipts to Sheet using Nanonets/Rossum (extract date, vendor, amount, category).
• Day 6: Connect Fireflies.ai (or similar) to auto-post meeting summaries to CRM and Slack.
• Day 7: Review metrics; switch one classification step to a cheaper model if quality holds.

Insider tip: Keep your tone guide, product elevator, and pricing blurb in Zapier Storage. Refresh once; all Zaps inherit it—no rebuilds, consistent voice.

Your move.

You’re aiming to use AI to find higher‑paying gigs faster — good focus on speed and rate. Let’s make it practical.

Quick win (under 5 minutes): Paste the prompt below into your AI, get a list of high‑pay keywords and a ready‑to‑use search string. Add it to your job board saved searches with a minimum budget filter. Expect fewer, better leads today.

Copy‑paste prompt:

“I’m a [your service: e.g., brand designer / copywriter / developer]. Produce: 1) 20 buyer‑intent keywords that correlate with higher budgets (e.g., audit, migration, compliance, conversion, strategy, retainer, M&A, enterprise, replatform, turnaround). 2) 10 decision‑maker titles (Founder, CEO, VP, Director, Head of…). 3) A search string I can paste into job boards that includes the keywords and titles, and excludes low‑budget signals (student, free, cheap, $5, test project). Output as comma‑separated lists I can copy.”

The problem: Most freelancers spend hours sifting low‑budget posts and send generic proposals. That’s wasted pipeline time and weak positioning.

Why it matters: Filtering for intent and budget at the top of the funnel increases average contract value and cuts time‑to‑call. You close fewer, better deals.

Lesson from the field: The fastest lift I see is a three‑part system — premium positioning, strict lead filters, and outcome‑based proposals. AI does the heavy lifting if you give it the right prompts.

What you’ll need:

• A chat AI (any reputable option)
• Access to 1–2 job boards or marketplaces
• A simple spreadsheet for tracking
• Calendar reminders for follow‑ups

Do this next — step‑by‑step

1. Define a premium positioning line (10 minutes)Use AI to turn your service into an outcome. Prompt: “Write 3 positioning lines for a [service] that promise a measurable business outcome in 12 words or less. Include the audience and the result. Example format: ‘I help [who] achieve [business result] with [service].’” Choose one and use it in every proposal.
2. Build your high‑pay keyword map (10 minutes)Ask AI for: high‑budget keywords, decision‑maker titles, and exclusions. Add them to saved searches. Pro tip: Keywords like audit, conversion, migration, replatform, compliance, strategy, retainer, scale, enterprise, due diligence tend to attract bigger budgets than generic “logo/blog/app.”
3. Set minimums and alerts (5 minutes)On each platform: set a minimum budget (e.g., $3,000+) or hourly floor. Turn on instant notifications. Respond within 30–60 minutes of posting to lift reply rates.
4. AI triage: score leads in seconds (15 minutes)When a post lands, paste it into your AI with: “Score this lead 0–100 for fit based on budget, decision‑maker seniority, urgency, business impact, and clarity. Explain the score in bullets. If under 70, tell me why I should skip.” Only pursue 70+.
5. Proposal factory (15 minutes per proposal)Prompt your AI: “Create a 150‑word proposal using this positioning line: [paste]. Include: 1) 2–3 outcome bullets with numbers; 2) a 3‑step plan; 3) one smart scoping question; 4) social proof in one line; 5) invite to a 15‑minute call.” Keep it tight and business‑focused.
6. Three‑tier pricing with anchors (10 minutes)Prompt: “Propose three packages (Good/Better/Best) for [project], each with clear deliverables, timeline, and value drivers. Anchor to outcomes, not hours. Include a Minimum Project Fee of $[your floor]. Suggest one fast‑start bonus if they book a call this week.”
7. Follow‑up sequence (5 minutes)Prompt: “Write three concise follow‑ups for a proposal: Day 2 value add (share a quick idea), Day 5 objection handle (budget/timing), Day 9 close (clear next step). 60–90 words each.” Schedule them.
8. Warm lane: curated outreach (30 minutes weekly)Have AI surface 10 ideal companies per week and draft one‑sentence insights. Prompt: “Based on my niche [describe], list 10 ideal client profiles and a one‑sentence insight I could send each that references a likely missed opportunity. Keep it respectful and specific.”

Insider signals that usually mean bigger budgets:

• Language: audit, overhaul, replatform, conversion, due diligence, compliance
• Titles: Founder, CEO, COO, VP, Director, Head of
• Context: acquisition, expansion, launch, migration, performance plateaus
• Formats: retainers, multi‑phase, discovery sprints, roadmaps

Metrics to track weekly:

• Qualified leads per hour (target: 3–5)
• Response rate to proposals (target: 30%+)
• First‑call rate per 10 proposals (target: 3–5)
• Average contract value (target: rising week over week)
• Time‑to‑first‑call (target: under 72 hours)
• Effective hourly rate (revenue ÷ actual delivery hours)

Common mistakes and fast fixes

• Applying to everything. Fix: enforce a minimum budget and 70+ triage score.
• Selling activities, not outcomes. Fix: lead with business results (revenue, risk reduction, time saved).
• Underpricing. Fix: set a minimum project fee and use three‑tier anchors.
• Slow follow‑up. Fix: pre‑write and schedule a 3‑touch cadence.
• Long proposals. Fix: 150 words, one call to action, proof in numbers.

One‑week action plan

1. Day 1: Lock your positioning line. Set your minimum project fee.
2. Day 2: Generate keyword map and exclusion list. Create saved searches with budget filters and alerts.
3. Day 3: Build your AI triage prompt. Score 20 posts. Pursue only 5–7 at 70+.
4. Day 4: Send 5 outcome‑based proposals using the factory prompt.
5. Day 5: Package your three‑tier pricing. Add a fast‑start incentive.
6. Day 6: Warm lane: send 5 curated outreach notes with one‑sentence insights.
7. Day 7: Review KPIs. Trim weak keywords. Refine prompts. Repeat.

What to expect: Cleaner pipeline, faster responses, and a higher average contract value within 7–14 days as your filters and proposals tighten.

Your move.

Quick win: Use AI to produce a clean, single-page onboarding doc in 10–20 minutes that reduces back-and-forth, speeds payments, and sets clear expectations.

The problem: onboarding documents are inconsistent, take too long to create, and leave clients confused about next steps.

Why this matters: a consistent onboarding sheet reduces time-to-first-deliverable, improves client satisfaction, and lowers admin hours — directly impacting revenue and capacity.

What I’ve learned: start simple. A one-page, templated onboarding that’s tailored per client wins every time. AI handles the copy and structure; you handle the specifics and approvals.

1. What you’ll need
   • Service summary (one sentence per service).
   • Standard deliverables, timeline, pricing terms, and client responsibilities.
   • Brand voice (formal/friendly) and logo/file placeholders.
   • Access to an AI writer (ChatGPT or similar) and a document tool (Google Docs/Word).
2. How to do it — step-by-step
   1. Collect the assets above into a single folder.
   2. Use this AI prompt (copy‑paste) to generate a draft:

AI prompt (paste into your AI tool):

“You are a professional client onboarding specialist. Create a one-page onboarding document for [SERVICE NAME] that includes: a short welcome (20–30 words), scope and deliverables (bulleted), timeline with 3 milestones and durations, client responsibilities (bulleted), payment terms, communication preferences (who, how, response times), and next steps with the first action item. Tone: friendly but professional. Keep it under 300 words. Use placeholders like [CLIENT NAME], [START DATE], [PRICE].”

1. Refine the AI output: replace placeholders, check dates/pricing, ensure compliance, and add your logo.
2. Create a template from the final doc so you can reuse and swap in specifics per client.
3. Automate delivery: attach the filled template to your welcome email or client portal and include a simple checklist and signature/request to confirm.
4. Pilot with one client, collect feedback, and adjust.
5. Scale by making the template a reusable file and storing it in your CRM or project tool.

Metrics to track

• Time to produce onboarding doc (target: <20 minutes).
• Client confirmation rate within 48 hours (target: >80%).
• Time-to-first-payment (target: reduce by 25%).
• Admin hours saved per month (target: track before/after).

Common mistakes & quick fixes

• Too generic: fix by adding 2–3 client-specific bullets (objectives, constraints).
• Overlong doc: trim to one page and highlight next action clearly.
• Not updating template: schedule monthly 10-minute reviews to keep details current.
• Poor delivery timing: send onboarding immediately after contract signature—automate it.

7-day action plan

1. Day 1: Gather service summary, pricing, and client-responsibility list.
2. Day 2: Run the AI prompt and generate 2 variations per service.
3. Day 3: Review and pick the best; swap placeholders with real examples.
4. Day 4: Create a reusable template and save to your docs/CRM.
5. Day 5: Automate sending via email template or portal.
6. Day 6: Pilot with one new client and collect feedback.
7. Day 7: Measure metrics and iterate on copy or process.

Your move.

—Aaron

Hook: You want notes converted into actionable tasks in Todoist or Notion — fast, reliable, and measurable. Good focus.

The problem: Notes are messy, inconsistent, and sit in a backlog. That friction means ideas never become done.

Why it matters: Turning notes into tasks systematically increases execution. You’ll reduce missed deadlines, speed decisions, and prove time saved with simple KPIs.

My core approach (what I’ve learned): Don’t automate everything; automate the extraction and routing. Use AI to parse notes into structured task items, then push to Todoist or Notion via an automation tool (Zapier/Make/Shortcuts) or native integration.

Quick checklist — do / do not

• Do: Standardize a note format (title, context, action verb, due/horizon, priority).
• Do: Use AI to extract intent, not to decide priority automatically.
• Do not: Over-automate due dates without human review.
• Do not: Push every extracted line as a task — filter for “actionable” only.

Step-by-step (what you’ll need, how to do it, what to expect)

1. Gather tools: a notes app (Apple Notes/Obsidian), OpenAI/GPT access (or built-in AI), Todoist and/or Notion accounts, and Zapier/Make for automation.
2. Create a short note template: Title / Context / Action / Due (optional) / Tags. Use this consistently for new notes.
3. Build an AI extraction step: send the note text to GPT to return JSON with fields: title, task, due, project, priority, notes.
4. Map the JSON to Todoist fields (content, due date, project, labels) or Notion database properties and create the item via Zapier/Make.
5. Test with 10 real notes, review results, refine the prompt and mapping, then enable auto-run for new notes.

Copy-paste AI prompt (use with GPT/OpenAI):

“You are an assistant that converts raw meeting notes into a JSON list of actionable tasks. For each actionable item, return: title, task (one-sentence action starting with a verb), due (YYYY-MM-DD or null), project (short name), priority (low/medium/high), tags (comma-separated), and notes (context). Ignore non-actionable statements. Output only valid JSON.”

Worked example

Note: “Discuss Q3 budget with finance, decide on hiring by May 20, follow up on vendor quote.”

• Todoist task: “Discuss Q3 budget with finance” — due 2025-05-20 — project: Finance — priority: high
• Notion entry: Task row with Title, Due: 2025-05-20, Status: To Do, Tags: finance, hiring, vendor

Metrics to track

• Tasks created per week from notes.
• Task completion rate within due date.
• Average time from note creation to task creation.
• False-positive rate (non-actionable items created as tasks).

Common mistakes & fixes

• Too many tasks: add an AI filter that flags actionability (yes/no).
• Wrong dates: require human confirmation for auto-set due dates.
• Duplicates: check recent tasks in project before creating new ones.

1-week action plan

1. Day 1: Standardize note template and pick automation tool.
2. Day 2: Create AI prompt and test extraction on 10 notes.
3. Day 3: Map fields to Todoist/Notion in Zapier/Make; run dry tests.
4. Day 4: Review results, tweak prompt and mapping.
5. Day 5: Enable automation for selective notebook or tag.
6. Day 6: Measure KPIs (tasks/week, completion rate).
7. Day 7: Tweak rules (dates, priorities) and scale to more notes.

Your move.