aaron

Hook: You don’t need more labels — you need a decision rule that turns your task list into calendar blocks, delegations, and measurable gains. Overlay the Eisenhower Matrix with value-per-hour scoring and you’ll stop firefighting and start compounding.

The problem: Urgency steals the mic. Without a numeric lens, everything feels critical, and Important/Not Urgent work never makes the calendar.

Why it matters: The win isn’t a prettier list; it’s throughput on the right work — revenue-driving projects shipped, needle-movers scheduled, noise delegated or dropped.

Lesson: Add one number: Value per Hour (VPH). Prioritize by VPH inside each quadrant. Then force next actions into your calendar or to a delegate with a micro-brief. That’s the conversion moment from plan to progress.

What you’ll need

• 6–12 tasks you might do in the next 14 days.
• A chat-style AI assistant.
• Your calendar and a place to assign delegations.

Rapid setup — 6 steps

1. List tasks with three fields per item: deadline (if any), expected business impact in dollars or a simple High/Med/Low, and rough effort hours.
2. Run the prompt below to classify by Eisenhower, compute VPH, and propose concrete next actions.
3. Accept the top 3 by VPH in Urgent & Important and block time this week.
4. Schedule one Important/Not Urgent item as a recurring deep-work block.
5. Delegate all Urgent/Not Important items using the AI-generated micro-briefs.
6. Drop or defer the rest explicitly — put a review date on deferrals.

Copy-paste prompt (robust)

Use this as-is. Expect a clean list you can execute today.

Classify the following tasks using the Eisenhower Matrix and compute a Value-per-Hour score. For each task, output: 1) Quadrant [UI, INU, UNI, N], 2) One-line reason (≤10 words), 3) Estimated impact (High/Med/Low or $), 4) Effort hours, 5) VPH = (Impact score per scale below) ÷ Effort, 6) One immediate action: Schedule (date/time), Block X hours (when), Delegate to [role] with 3-bullet brief, or Drop/Defer (date), 7) 90-day impact: High/Med/Low. Scales: High=$5k or 3 points, Med=$1k or 2 points, Low=$250 or 1 point. Then: a) Rank tasks by VPH within each quadrant, b) Propose calendar blocks for the top 3 overall this week, c) Generate a delegation micro-brief (3 bullets: desired outcome, key info, deadline) for all items in UNI. Tasks (include deadline if any, impact, effort hours): [paste tasks].

Insider trick: Ask the AI to highlight “fast wins” — VPH High with effort ≤60 minutes — and knock two out before lunch. Momentum is an asset.

Optional follow-up prompts

• Daily stand-up: “Given yesterday’s outcomes, re-evaluate my list, update VPH, and propose the single highest-leverage 90-minute block today. If my calendar is full, suggest the first reschedule to fit it.”
• Delegate-ready packets: “For each UNI task, produce a 4-line handoff: goal, definition of done, inputs/links I must provide, due date. Include a 1-sentence status check I can paste in chat/email.”

What to expect

• Clear top 3 actions with times you can commit to today.
• At least 1–2 items moved from Important/Not Urgent into the calendar.
• Delegation friction drops — the micro-briefs eliminate back-and-forth.

Metrics to track (weekly)

• Time allocation: ≥40% in Important/Not Urgent blocks.
• Delegation rate: ≥25% of tasks in Urgent/Not Important delegated within 24 hours.
• Start-lag on Urgent & Important: ≤24 hours from identification to first working block.
• Throughput: # of High-90-day-impact tasks completed.
• Calendar adherence: ≥80% of scheduled deep-work blocks honored.

Mistakes and quick fixes

• Everything looks urgent: Force impact-first ranking, then assign quadrants.
• Vague impact numbers: Use the points scale (3/2/1) — faster than dollars.
• Over-optimistic effort: Ask the AI for effort ranges and pick the 75th percentile.
• Delegations bounce back: Add “definition of done” and one approval checkpoint to every brief.
• Calendar overwhelms: Cap deep work to two 90-minute blocks per day; defer the rest.

1-week action plan

1. Day 1 (20 minutes): List tasks, run the robust prompt, calendar top 3, delegate UNI with micro-briefs.
2. Day 2–4 (10 minutes each morning): Re-run with updates, protect scheduled blocks, ship one Important/Not Urgent deliverable.
3. Day 5 (15 minutes): Review metrics above, adjust impact/effort estimates, drop or defer anything with Low impact and Low VPH.
4. Day 6: Add one recurring weekly deep-work block for your highest-ROI Not Urgent project.
5. Day 7: Summarize wins, note any misses, reset the list for next week using the same prompt.

Bottom line: Eisenhower tells you what bucket; VPH tells you what to do first. Convert that into calendar blocks and clean delegations, and the KPIs move.

Your move.

— Aaron

Quick win (under 5 minutes): Paste your best-performing headline into an AI prompt and generate 10 headline variants you can test immediately.

Problem: most side-giggers write one ad, pray it works, and wonder why conversions are low. You can’t rely on guesswork — converting ads are a numbers game: clear audience, tight offer, multiple creative variants, and rapid testing.

Why it matters: a 10–20% lift in conversion rate or a 10% drop in CPA turns a struggling side gig into profitable monthly revenue without more hours of work. This is where AI pays off: it multiplies creative options fast so you can find winners sooner.

My lesson: treat AI as a multiplier for hypothesis generation, not a magic copy button. Use it to create focused variants, test small, measure, then scale winners.

1. Define your goal & baseline — what is a conversion (sale, lead)? Pull last 30 days CPA, CTR, CVR. This is your benchmark.
2. Gather assets — best headline, 1–3 product images or a 15s video, 2 customer pain points, USP (one-liner), and target audience description.
3. Generate variants with AI — produce 5 headlines, 3 body copy variants (short, benefit-led, story), and 3 CTAs.
4. Create 6-8 ad combinations — mix headlines, copy, images; keep one variable per test where possible.
5. Launch low-budget A/B tests — allocate small daily spend per variant (enough to reach statistical signals: 200–500 clicks across tests if possible).
6. Measure and decide — run for enough traffic; pull CPA, CTR, CVR, cost per click (CPC), and return on ad spend (ROAS). Kill losers at 2x baseline CPA.
7. Scale winners — double spend on top performers and iterate creative every 7–10 days.

What you’ll need: ad account (Facebook/Google/Instagram), the best-performing headline, one image or short video, an AI chat tool, a spreadsheet to track results.

What to expect: in week one you’ll create 8–12 variants and get initial signals. Expect noise — look for consistent winners on CTR and CVR, not single-day spikes.

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

“You are an expert ad copywriter. Product: [brief product description]. Audience: [age, interest, pain point]. Offer: [discount, lead magnet, webinar, etc.]. Objective: [sale or lead]. Produce: 5 short headlines (max 30 chars), 5 medium headlines (30–60 chars), 3 body copy variants (short: 90 chars; standard: 125–150 chars; long: 200 chars) each with a distinct tone (direct, empathetic, humorous). Also provide 5 CTA options and a 2-line suggested description for the image. Include the primary benefit first and a clear next step.”

Metrics to track:

• CTR — which creative grabs attention
• CVR (conversion rate) — which creative completes the action
• CPA (cost per acquisition) — your primary profitability metric
• CPC and CPM — efficiency signals
• ROAS — revenue returned per dollar spent (if applicable)

Common mistakes & fixes:

• Testing too many variables at once — fix: one variable per test or structured combinatorial tests.
• Relying on impressions, not conversions — fix: measure CPA/CVR and tie to business goal.
• Letting an AI-generated idea run unedited — fix: human-edit for clarity and brand voice.
• Underfunding tests — fix: ensure each variant gets enough clicks to show a trend.
• Changing targeting mid-test — fix: lock targeting while comparing creatives.

1-week action plan:

1. Day 1: pull baseline metrics, collect assets, pick top headline.
2. Day 2: run the AI prompt above to generate headlines/copy.
3. Day 3: pair copy with images/video and build 6 ad variations in your ad manager.
4. Day 4: launch tests with small daily budgets per variant.
5. Day 5–6: monitor CTR/CPC; pause obvious losers.
6. Day 7: evaluate CPA/CVR; double budget on the top performer or iterate new variants.

Your move.

Short answer: You can build a reliable, low-tech AI workflow that writes personalized outreach, sends scheduled follow-ups, and stops when someone replies — without a developer. Focus on speed, repeatability, and one metric: reply rate.

The problemManual outreach is slow, inconsistent, and easy to let slip. That kills pipeline velocity and makes scaling impossible.

Why it mattersPredictable outreach turns activity into meetings. Even small improvements in reply rate (from 3% to 8%) materially increases qualified conversations and revenue opportunities.

Short lesson from the fieldKeep personalization tight (1–2 lines), automate status updates so follow-ups stop when someone replies, and run fast A/B tests on subject lines and first sentences.

What you’ll need

• Lead list (CSV, LinkedIn export, event list).
• Google Sheets (or simple CRM like HubSpot free tier).
• An automation tool (Zapier, Make) or your CRM’s sequences.
• Email account (Gmail/Outlook) or SMTP-enabled sender.
• AI assistant (ChatGPT or other LLM accessible via web or Zapier/OpenAI integration).

How to set it up — step-by-step

1. Prepare sheet: columns for Name, Company, Role, Email, KeyFact, Status, LastSent, Replies.
2. Create an automation: trigger = new row or status = ready. Action 1 = call AI to generate subject + 3 message variants (initial + 2 follow-ups). Action 2 = send initial email from your account. Schedule follow-ups at 3 and 7 days if Status stays “no reply.”
3. Reply detection: use your automation or Gmail filters to tag replies and update Status to “replied” (this cancels follow-ups).
4. Small batch testing: send 20–50 emails first. Review tone, opens, replies. Tweak copy via the AI prompt and redeploy.
5. Scale by batches of 50–100 after you hit a consistent reply rate you’re happy with.

Copy-paste AI prompt (use as-is)

You are a professional outreach writer. For this lead, generate: 1) a 6–8 word subject line; 2) an initial cold email (max 110 words) with a 1–2 sentence personalized opener referencing {KeyFact}, a one-sentence value statement showing outcome, and a single clear CTA asking for a 10–15 minute call; 3) two brief follow-ups (each 30–60 words) that reference the prior message, add one new micro-benefit or social proof, and offer the same CTA. Tone: warm, concise, non-salesy. Output as Subject:, Email 1:, Follow-up 1:, Follow-up 2:. Variables: {Name}, {Company}, {Role}, {KeyFact}, {Offer}.

What to expect (benchmarks)

• Open rate: 20–40%
• Reply rate: 4–12% (aim for 6%+ initially)
• Meeting set rate (from replies): 15–30%
• Unsubscribe rate: keep below 0.1%

Common mistakes & fixes

• Mistake: Pulling bad personalization. Fix: Only use verifiable facts and keep the personal line short.
• Mistake: No reply tracking. Fix: Automate reply detection and stop sequences immediately.
• Mistake: Sending too many touches. Fix: Limit to 2–3 touches and always add new value.

7-day action plan (exact)

1. Day 1: Import 50 leads into Google Sheets; add KeyFact for each (one-sentence).
2. Day 2: Build Zap: new row -> AI prompt -> send email; set follow-up delays (3, 7 days).
3. Day 3: Test send to 5 internal addresses; adjust tone and subject lines.
4. Day 4: Send first batch of 20 live prospects.
5. Day 5: Verify reply detection and stop logic; fix any failed sends.
6. Day 6: Review opens/replies; change subject line A/B if opens <20%.
7. Day 7: Tweak message copy with AI, then send next 30–50 based on learnings.

Your move.

Quick win (under 5 minutes): open your editor, create a 1080×1350 slide, add your logo top-left, a bold heading box centered, and a one-line footer with your accent color — save as “Cover Master.” Clone it next time you start a carousel.

The problem: inconsistent carousels cost time, confuse your audience, and kill repeatable growth. Teams redesign each post instead of templating once.

Why this matters: consistent templates reduce creation time, lift perceived professionalism, and make CTAs and headlines testable — which drives more saves, shares, profile visits and conversions.

My experience — short lesson: I had a client cut production from 3 hours to 25–35 minutes per carousel by locking four slide types and using AI for headlines and art direction. Saves rose 28% in three weeks because visuals and messaging became predictable and scannable.

What you’ll need

• Brand tokens: hex codes, two fonts (heading + body), logo file.
• Template editor: Canva, Google Slides, Figma or similar.
• 6 reference posts for mood/layout.
• AI assistant for copy + image art direction (Chat-style or image generator).
• Your phone for mobile checks.

Step-by-step (build & run)

1. Create a master file 1080×1350 with 20–30px safe margins; add logo, footer (handle + page #), and a small accent element.
2. Design and save 4 slide types: Cover, Text+Image, Quote, CTA. Keep one grid per type — no exceptions.
3. Set reusable components: heading frame (2–3 sizes), 3-line body box, image mask with a 20% overlay option, consistent icon spot.
4. Batch-populate 5 carousels: run the AI prompt below, paste text into templates, add images per art direction, export PNGs and test on your phone for legibility.
5. Schedule and measure. Iterate headline and CTA based on performance.

Copy-paste AI prompt (use this exactly)

“You are creating Instagram carousels for {brand_name}. Brand colors: {primary_hex}, {accent_hex}. Fonts: {heading_font}, {body_font}. Create 5 carousel variations, each 5 slides. For each carousel provide: Slide 1 headline (6–8 words) + 4 headline variations; Slide 2 one-sentence problem + one statistic; Slide 3 three one-line solutions; Slide 4 one short client/example line with a measurable result; Slide 5 single-line CTA with a clear benefit. Also provide a 2-line caption, 5 hashtags, and image art direction per slide (mood, color accents, subject, suggested crop). Tone: confident, helpful, simple. Keep each slide sentence under 12 words.”

Metrics to track

• Engagement rate (likes + comments) per carousel
• Saves and shares (primary quality signal)
• Profile visits and link clicks within 48–72 hours
• Time-to-create per carousel (target: 20–40 minutes)
• CTA conversion (download, signups, DMs)

Common mistakes & fixes

• Text too small: increase headings 10–15% and test on phone.
• Busy images: add a 20–30% dark or brand-color overlay behind text.
• Too many slide types: reduce to 4 core types for the first month.
• Weak CTA: one verb + one benefit (e.g., “Download checklist — save 3 hrs/week”).

7-day action plan (do-first)

1. Day 1: Build Cover Master + three other templates.
2. Day 2: Set brand tokens and collect 6 reference posts.
3. Day 3: Run the AI prompt for 5 carousels and review outputs.
4. Day 4: Populate templates, add images, export, test on phone.
5. Day 5: Schedule 2 carousels; monitor engagement for 48 hours.
6. Day 6: Iterate best headline + CTA; refine one template detail (type size/overlay).
7. Day 7: Batch 3 more carousels using the updated template.

Your move.

— Aaron

Quick win (under 5 minutes): open your editor, create a new 1080×1350 slide, place your logo top-left, add a bold heading box and a one-line footer with your accent color — save it as “Cover Master.” You now have a reusable cover you can clone instantly.

The problem: inconsistent carousels dilute brand recognition and cost hours to produce. Most teams redesign each post instead of templating once.

Why it matters: consistent templates reduce creation time, increase perceived professionalism, and make testing headlines and CTAs reliable. That drives higher saves, shares, and profile visits.

My experience — short lesson: set up 4 slide types and use AI to populate copy and art direction. That changed a client’s production time from 3 hours to 25–35 minutes per carousel and improved saves by 28% in three weeks.

1. What you’ll need
   • Brand tokens: hex codes, two fonts, logo file.
   • Design canvas: Canva, Google Slides, or any template-supporting editor.
   • Six example posts that match your desired mood.
   • AI assistant for copy + image art direction.
2. Step-by-step build (what to do)
   1. Create a master file with a 20–30px safe margin grid and fixed footer for page numbers/handle.
   2. Design and save 4 slide types: Cover, Body (text + image), Quote, CTA.
   3. Add reusable components: heading frame, 3-line body text box, image mask, accent element.
   4. Duplicate the templates and batch-populate content using the AI prompt below (5 carousels at a time).
   5. Export optimized PNGs sized for mobile, review on your phone, adjust contrast if needed.

Copy-paste AI prompt (use this to generate headline + slide bullets + art direction):

“Create a 5-slide Instagram carousel for a small business coach. Slide 1: short attention-grabbing headline (6–8 words). Slide 2: state the problem in one sentence + two quick stats. Slide 3: three concise solutions as bullets (one line each). Slide 4: a short client example (1–2 lines). Slide 5: clear CTA with an offer to download a checklist. Provide 4 headline variations and art direction notes for images (mood, colors, subjects, suggested crop). Tone: confident, helpful, simple.”

Metrics to track

• Engagement rate (likes+comments) per carousel
• Saves and shares (primary signal for quality)
• Profile visits and link clicks after posting
• Average time-to-create per carousel

Common mistakes & fixes

• Text too small: increase heading by 10–15% and test on phone.
• Too many slide types: cut to 4 core types for the first month.
• Weak CTAs: make it single action + one-line benefit.
• Images clash with text: add 20–30% darker overlay behind text.

1-week action plan

1. Day 1: Build master file and save 4 slide templates.
2. Day 2: Collect 6 style references and set brand tokens in your editor.
3. Day 3: Use the AI prompt to generate content for 5 carousels.
4. Day 4: Populate templates, export, test on phone, tweak spacing/contrast.
5. Day 5: Schedule posts and track metrics. Batch two more carousels if time permits.

Your move.

Hook: Build a weekly personalization engine that hits 35–55% opens, 3–8% replies, 1–3% positive replies—without tripping spam filters.

The real issue isn’t writing; it’s input control and guardrails. Most “personalization” fails because the trigger is vague, the claim is risky, and volume is scaled before the data is ready.

Why this matters: Deliverability is a compounding asset. Tight inputs + human checks + measured send ramps = durable reply rates and consistent meetings. That’s pipeline you can forecast.

Lesson from the field: One verifiable fact + one relevant benefit + one easy question outperforms long copy. The win comes from repeatable data hygiene and AI prompts that refuse to invent.

How to run it (end-to-end)

1. Define success (this week): meetings per 100 sends, positive reply rate, complaint rate, and bounce rate. Set thresholds: 1–3 meetings/100 sends, 1–3% positive replies, <0.2% complaints, <2% hard bounces.
2. Standardize your CSV: name, role, company, email, persona, trigger (≤18 words, source noted), benefit (role-specific, one sentence), proof (short, optional), timezone, reason (blank now).
3. Collect real triggers: job posts, product updates, quotes, funding, hiring, content topics. Enforce a 12–18 word limit. If you can’t verify it in 10 seconds, skip it.
4. Generate drafts in safe batches (50–100) using the prompt below. Output as CSV. Include a reason column with NEEDS_CHECK flags for anything uncertain.
5. QC before sending: skim 15–20 samples. Delete flattery, trim long lines, and normalize claims (avoid percentages unless you can prove them). If in doubt, use a neutral role insight.
6. Send for deliverability: plain text, 40–70 words, no links or images in the first touch. Stagger over 2–3 days. Keep daily volume consistent. Authenticate your domain and warm gradually.
7. Follow up twice: Day 3 nudge, Day 7 close-the-loop. Under 28 words each. No pressure language.
8. Classify replies fast: Use the classifier prompt to tag Positive / Neutral / Referral / OOO / Not a fit. Auto-draft a short human reply for Positive and Referral.
9. Scale only after stability: If two consecutive batches meet targets, increase volume 25–50%. If any metric slips, pause scale and fix inputs.

Copy-paste AI prompt (robust batch generator)

Prompt: “You create concise, human cold emails. Use only the facts provided. Do not invent or embellish. If any trigger or proof is vague, set reason to NEEDS_CHECK and write a neutral role insight instead. Input per contact: name, company, role, persona, trigger, benefit (one sentence), proof (short, optional). Output CSV columns: name, subject (6–8 words), email (max 70 words, plain text), reason. Email format: ‘Hi {name} — [one-sentence personal line referencing the trigger if safe; otherwise a neutral insight for the role]. [One-sentence value aligned to the benefit + optional proof]. [Soft one-question CTA].’ Tone: warm, direct, no flattery, no jargon, no links, no attachments. Examples of CTAs: ‘Worth exploring?’, ‘Open to a quick chat next week?’, ‘Should I send a 3-line summary?’”

Reply classifier + auto-drafter

Prompt: “Classify each reply as: POSITIVE (wants to talk), NEUTRAL (asks for info/time), REFERRAL (points to colleague), OOO (auto-responder), NOT A FIT, UNSUBSCRIBE, BOUNCE. Then draft a 2–3 sentence plain-text response for POSITIVE or REFERRAL only. Inputs: original email (benefit), recipient reply. Output fields: label, action_note, draft_response.”

Templates that convert

• First email (45–70 words): Hi {name} — {specific trigger}. We help {persona} {benefit}. {optional proof}. Open to a quick chat next week, or want a 3-line summary?
• Follow-up 1 (≤28 words): Re: {trigger}. If {benefit} is on your list, I can send a 3-line summary. Worth it?
• Follow-up 2 (≤28 words): Closing the loop — happy to park this. If {benefit} becomes a priority, want a one-pager later?

What to expect

• Batch 1–2: stabilize deliverability and message-market fit. Aim for 35–45% opens, 2–5% replies.
• Batch 3–4: lift with better triggers and subjects. Target 45–55% opens, 3–8% replies, 1–3% positive.
• Meetings trend: 1–3 per 100 sends once inputs are tight and volumes are steady.

Metrics that matter (track weekly)

• Open rate by subject and sender.
• Reply rate and positive reply rate.
• Meetings per 100 sends; time-to-first-reply (<48 hours target).
• Hard bounce (<2%), spam complaint (<0.2%), OOO rate (context for volume).
• Top 5 triggers by positive replies (double down next week).

Common mistakes and quick fixes

• Unverifiable triggers. Fix: mandate trigger source notes; if unsure, switch to neutral role insight.
• Over-claims. Fix: remove percentages unless backed by a public proof line.
• Links in first touch. Fix: offer a 3-line summary by reply; share links only after engagement.
• Scaling before stability. Fix: require two healthy batches before increasing volume.
• Ignoring OOO and referrals. Fix: use the classifier to queue smart resends and reach the referred contact.

One-week action plan (results in 7 days)

1. Day 1: Finalize CSV schema. Build 100 contacts with one tight trigger each. Write persona benefits and one proof line.
2. Day 2: Run the batch generator for all 100. Inspect the reason column; edit or remove NEEDS_CHECK.
3. Day 3: Send 50 emails (plain text). Start the reply classifier. Log metrics.
4. Day 4: Send remaining 50. Capture top subjects and triggers from replies.
5. Day 5: Send Follow-up 1 to non-responders. Book meetings from positives within 24 hours.
6. Day 6: Review metrics vs targets. Tighten benefits and proof. Prepare next 150 with improved triggers.
7. Day 7: Send Follow-up 2. If metrics meet thresholds, plan a 25–50% volume increase next week.

Insider edge: Add a “reason” column to every output. It’s the fastest way to spot weak triggers and keep the tone grounded. When in doubt, shorten. Short wins.

Your move.

Hook

Good question — focusing on both themes and sentiment is the right place to start. You can get actionable insights from open-ended responses without being a data scientist.

The problem

Open-ended answers are rich but messy: inconsistent language, varying lengths, and hidden themes make manual analysis slow and error-prone.

Why it matters

Extracting reliable themes + sentiment turns messy text into measurable KPIs you can act on (product fixes, messaging changes, support training). Fast, repeatable analysis scales decision-making.

What I’ve learned

Automate the heavy lifting with an LLM or topic model, but always validate with human review and a small labeled sample. That keeps precision high and false signals low.

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

1. Gather data: export responses to CSV (columns: id, question, response, metadata).
2. Sample & label: randomly label 200–500 responses for themes and sentiment to create a validation set.
3. Preprocess: trim whitespace, remove duplicates, keep original text; add length and metadata columns.
4. Run analysis: use an LLM or a topic modeling tool to extract themes and assign sentiment scores (see prompt below).
5. Cluster & summarize: group similar labels, count frequency, extract representative quotes per theme.
6. Human review: review top 10 themes and 200 edge cases, update rules or prompt, re-run if needed.
7. Deliver results: table of themes (name, count, %), avg sentiment per theme, top representative quotes.

Copy-paste AI prompt (use as-is with your LLM)

“You are a customer-insights analyst. For each survey response, do three things: 1) assign up to 2 concise theme labels (comma-separated), 2) give sentiment as Positive / Neutral / Negative and a confidence score 0-1, 3) return a single short representative quote (max 20 words). Output as tab-separated values: id [TAB] themes [TAB] sentiment [TAB] confidence [TAB] quote. Do not add extra commentary.”

Metrics to track

• Theme coverage (% of responses assigned a theme)
• Sentiment accuracy vs labeled sample (% agreement)
• Top-5 theme share (% of responses)
• Time per full analysis run (minutes)

Common mistakes & fixes

• Over-labeling: restrict to 1–2 themes per response.
• Ambiguous themes: merge similar labels into canonical names after clustering.
• Blind trust in scores: always validate with a labeled sample; recalibrate prompt if accuracy <85%.

1-week action plan

1. Day 1: Export data, create sample, set up CSV.
2. Day 2: Label 200–300 responses for validation.
3. Day 3: Run LLM batch with the prompt, get raw output.
4. Day 4: Cluster themes, compute counts, and extract quotes.
5. Day 5: Human review of top themes and 200 edge cases; adjust prompt/rules.
6. Day 6: Re-run and finalize theme list and sentiment metrics.
7. Day 7: Deliver summary dashboard and recommended next actions (top 3 fixes by impact & effort).

Your move. — Aaron

Your one-sentence brief is the right forcing function — it nails intent, scope, and CTA placement before the AI writes a word. Let’s turn that clarity into rankings and affiliate clicks.

5‑minute move (do this now): Add a short “At‑a‑glance Verdict” box to the top of your next post and tighten the title/meta for CTR. Paste this prompt:

AI prompt (copy‑paste): Create a 40–60 word “At‑a‑glance Verdict” for the keyword {PRIMARY_KEYWORD} with {INTENT} intent. Name the top pick, 2 key benefits, 1 drawback, and who it’s best for. Write a title under 60 characters and a 150‑character meta description that sets a clear expectation. Provide two CTA lines: “See price” and “Compare alternatives”. Keep it factual and non‑hypey.

The problem: Generic AI drafts mirror the SERP but miss the conversion path. That leads to low CTR, skim-and-bounce behavior, and weak affiliate clicks.

Why it matters: Affiliate revenue is a function of impressions → CTR → on‑page click‑through → conversion. Improving any one of those by a few points compounds fast.

Field lesson: Treat AI as your editor, then engineer two deltas the SERP lacks: an above‑the‑fold verdict box and unmistakable CTAs mapped to intent. Those two pieces lift CTR and outbound clicks without more traffic.

1. Intent triage (10 minutes) — Choose one buyer‑intent keyword. Write the brief sentence you outlined (intent, length, CTA spots). Add 3 objections your reader likely has. Expect: sharper intro and FAQs that pre‑empt bounces.
2. Capture SERP shape (10 minutes) — Open the top 3–5 results, copy their H2/H3s into your notes. Paste the headings into the prompt below to extract must‑have sections and gaps.
3. AI brief builder (copy‑paste) You are an SEO editor. Target keyword: {PRIMARY_KEYWORD}. Intent: {INTENT}. Here are H2/H3s from the top results: {PASTE_HEADINGS}. Produce: (1) 1‑sentence search intent; (2) required sections shared by top results; (3) content gaps we can own; (4) H2/H3 outline; (5) two CTA placements with copy; (6) title under 60 chars; (7) 150‑char meta; (8) 5 FAQs using related keywords; (9) fact‑check checklist. Keep it concise and practical.
4. Draft with conversion baked in (20 minutes) — Use the outline and run this prompt:

AI prompt (copy‑paste): Write a {WORD_COUNT} word SEO‑friendly affiliate post for {PRIMARY_KEYWORD} with {INTENT} intent. Start with a 3‑bullet “At‑a‑glance Verdict” box naming the top pick and who it’s for. Include H2/H3s from the outline, a comparison section, pros/cons, and a conclusion with a clear CTA placed after the top pick and again at the end. Add a 4‑question FAQ using related keywords. Suggest 3 internal link anchor texts. Do not invent specs; insert [Verify] where data is needed. Tone: helpful, authoritative, skimmable.

1. Edit for trust (30 minutes) — Compress the intro, add one unique proof (test note, screenshot, or quote), mark affiliate links as sponsored/nofollow, and keep CTAs visually consistent. Expect: higher on‑page click‑through.
2. Publish and instrument (10 minutes) — Set up an event for affiliate link clicks, add 2 internal links, and stamp “Updated: {month year}” near the top. Expect: clean attribution and freshness signals.
3. Optimize from data (15 minutes on Day 3 and Day 7) — If CTR is soft, test a sharper title/meta. If outbound CTR is low, move the first CTA directly under the top pick and add a one‑line “who it’s for.”

Metrics to watch (simple ladder)

• 48 hours: Indexed? Impressions started? If not, fetch/index and add one internal link from a crawled page.
• 7 days: SERP CTR for the primary query; aim to beat your site’s median. Outbound affiliate click rate per session; target steady improvement week over week.
• 14–28 days: Average position trend, organic sessions per post, affiliate clicks per 100 sessions, revenue per 1,000 sessions.

Common mistakes & fixes

• Thin, unverified specs — fix: add [Verify] tags in draft; confirm and replace before publishing.
• No above‑the‑fold answer — fix: use the verdict box so scanners get value in seconds.
• Vague CTAs — fix: action + outcome copy (“See price” / “Compare alternatives”) placed after the top pick.
• Mismatched intent — fix: mirror top‑ranking section order and only add gaps that deepen decision quality.

1‑week action plan

1. Day 1: Pick 3 buyer‑intent keywords. For each, write the one‑sentence brief and capture H2/H3s from top results.
2. Day 2: Run the brief‑builder prompt for all 3. Approve outlines and CTA placements.
3. Day 3: Generate drafts with the conversion‑first prompt. Add the verdict box to each.
4. Day 4: Edit for trust — insert proof, disclosures, internal links, and event tracking.
5. Day 5: Publish 1 post. Baseline metrics: impressions, CTR, affiliate clicks per session.
6. Day 6: Publish the second post. Add an internal link from an existing relevant page to each new post.
7. Day 7: Review CTR and outbound CTR. If below your median, test a new title/meta and move the first CTA under the top pick.

Short, controlled loops. Brief → draft → proof → publish → measure → tweak. Your move.