aaron

Strong point on using the after-tax effective rate as your yardstick. Let’s upgrade that into a simple, AI-powered decision rule you can run in 15 minutes and reuse every quarter.

The real problem: You’re comparing a guaranteed return (debt payoff) to a probable return (investing) without a policy. That creates second-guessing and delays.

Why this matters: A clear rule increases net worth faster, cuts interest waste, and removes decision fatigue. The win is consistency, not heroics.

Lesson from the field: Don’t chase average returns. Set a hurdle rate, protect liquidity, and use triggers. Cash has option value; once you send it to a lender, it’s hard to get back.

Build your decision rule (copy this playbook):

1. Protect liquidity first: Top up emergency savings to 3–6 months of essential expenses. No exceptions.
2. Set your hurdle rate: Choose a conservative expected return (e.g., 5%). This is the bar investments must clear after tax. For each debt, use after-tax cost: rate × (1 − tax rate) if deductible; otherwise the full rate.
3. Segment your debts:
   • Toxic (non-deductible ≥ 10–12%): pay down aggressively before investing.
   • Middle (5–9% after-tax): choose based on risk comfort; see step 5.
   • Low (≤ 4% after-tax, often fixed mortgages): usually invest, keep minimums.
4. Add variable-rate logic: If a loan can reset higher, give it extra priority. Rising rates erase your investment gains quickly.
5. Decide the split: If middle-tier debt is close to your hurdle (within 1–2 percentage points), split extra cash 50/50 between paydown and investing for 12 months, then reassess.
6. Automation: Schedule two transfers on payday: one extra-principal payment (labeled “principal only”) and one investment contribution.

Insider tricks that move the needle:

• Statement-date timing (credit cards): Make the extra payment right after the statement closes to cut the average daily balance and interest immediately.
• Mortgage recast: After a large principal payment, ask your lender about a recast. It lowers your monthly payment without a refinance (small admin fee), boosting cash flow flexibility.
• Prepayment tags: Ensure extra payments are applied to principal, not future interest. Most lenders let you specify this online.

Use AI to run the decision — paste this:

“Act as my financial analyst. Here are my debts: [Debt name | balance | interest rate | fixed/variable | tax-deductible: yes/no | minimum payment | prepayment penalty: yes/no]. My marginal tax rate is [X%]. I have [Y] months of emergency savings. I can allocate $[Z]/month extra. Use a conservative expected investment return of [5%] and also test 3% and 7%.

Tasks:
1) Calculate after-tax cost for each debt and rank them.
2) Model four strategies over 10 years: (A) Avalanche (highest after-tax rate first), (B) 50/50 split between payoff and investing, (C) Liquidity-first until 6 months of expenses then Avalanche, (D) Invest-first if all debts are ≤ hurdle.
3) For each, show: debt-free date, total interest paid, projected investment value, net worth difference vs next best option, and the breakeven return that makes investing tie with payoff.
4) Flag any variable-rate risk and prepayment penalties.
5) Give a simple recommendation and a monthly automation plan I can implement tomorrow.
Keep the language plain and the output in a compact list.”

What you’ll need (10 minutes): balances, rates, deductible yes/no, variable vs fixed, minimums, prepayment penalties, your tax bracket, emergency-fund months, monthly extra cash.

What to expect: A clear winner in most cases. If results are close, the 50/50 split de-risks regret while compounding begins on both fronts.

Metrics to track monthly:

• Debt-free date: target month you become payment-free.
• Total interest saved: cumulative vs minimum-payments baseline.
• Net worth delta: investments plus cash minus debt vs last quarter.
• Cash buffer: months of expenses; never below 3.
• Breakeven return: the return needed to justify investing over paying a specific debt; lower is better for investing.

Common mistakes and quick fixes:

• Comparing pre-tax to after-tax — Fix: convert everything to after-tax before deciding.
• Underfunded emergency cash — Fix: pause extra debt/invest until 3–6 months is set.
• Ignoring variable-rate exposure — Fix: prioritize variable loans; run a +2% rate shock in AI.
• Overconfident return assumptions — Fix: test 3%, 5%, 7% and use the lowest to decide.
• Prepayment penalties — Fix: verify before sending lump sums; redirect to highest-cost debt if penalized.

One-week action plan:

1. Day 1: Gather your numbers and confirm deductible status, variable vs fixed, and penalties.
2. Day 2: Run the AI prompt above. Save the output and pick the top strategy.
3. Day 3: Set two automations for payday: extra-principal payment and investment contribution. Tag the debt payment as “principal only.”
4. Day 4: If you have a mortgage and plan a lump sum, call the lender about a recast option.
5. Day 5: Implement statement-date timing for any credit card balances.
6. Day 6: Create a 3-line dashboard: debt-free date, interest saved YTD, cash buffer months.
7. Day 7: Schedule a quarterly 30-minute review to rerun the AI with updated balances and rates.

Clarity beats complexity. Build the rule once, automate it, and let the math work while you sleep. Your move.

Agree with your latest: low-tech, repeatable, QA-driven works. Here’s how to make AI the policy enforcer so your UTMs are consistent every time — without turning this into a software project.

The risk: letting AI “invent” names leads to drift. The fix: feed AI a controlled vocabulary and a strict pattern, then have the sheet assemble and validate. AI suggests; your sheet decides.

Field lesson: teams that switch from freeform naming to a controlled list plus an AI validator cut duplicate campaign labels by 80–90% and cut UTM build time by 70%+. That’s the leverage you want.

What you’ll need

• One spreadsheet with two tabs: Build and Vocab
• Controlled values for source, medium, channel, audience, offer
• AI assistant for generation and QA (use the prompts below)

Implementation steps

1. Create your controlled vocabulary (Vocab tab)
   • Columns: Term | Allowed values | Short code (optional)
   • Examples: medium = email, cpc, social, referral. source = google, facebook, linkedin, newsletter.
   • Include a synonyms column to map stray entries (e.g., “Google Ads” → google, “FB” → facebook).
2. Lock inputs and auto-clean
   • In Build tab, create input columns: Base URL | Source | Medium | Audience | Offer | Month (YYYYMM) | Content | Term (optional)
   • Apply data validation on Source/Medium from Vocab lists to prevent off-pattern values.
   • Add helper cells to normalize any text pasted in: use lower case and hyphens only (e.g., =LOWER(SUBSTITUTE(TRIM(cell),” “,”-“))).
   • Add a VLOOKUP to map synonyms to approved values before finalizing.
3. Standardize the campaign name
   • Pattern recommendation: brand-offer-audience-channel-yyyymm (lowercase, hyphens).
   • Build it with a formula so no one types by hand, e.g., =TEXTJOIN(“-“,TRUE,brand,offer,audience,channel,month).
   • Put variants only in utm_content (creative, subjectline, size).
4. Generate the final URL safely (handles existing ?)
   • Google Sheets example: =A2 & IF(REGEXMATCH(A2,”\?”),”&”,”?”) & “utm_source=” & B2 & “&utm_medium=” & C2 & “&utm_campaign=” & D2 & IF(E2=””,””,”&utm_content=” & E2) & IF(F2=””,””,”&utm_term=” & F2)
   • Add URL encoding if you expect special characters: wrap values with ENCODEURL(value) where available.
5. Auto-dedupe on the fly
   • Use COUNTIF to append a short suffix when a duplicate campaign name appears: =D2 & IF(COUNTIF($D$2:D2,D2)>1,”-v” & COUNTIF($D$2:D2,D2),””).
   • Freeze the final name once a campaign goes live (protect the cell).
6. Use AI for batch naming (constrained)
   • Feed AI only approved values from your Vocab. Do not let it invent new mediums or sources.
7. Use AI for QA before launch
   • Have AI scan a list of proposed UTMs and flag any values not in your Vocab, wrong casing, spaces, missing fields, or bad separators.
8. Governance rule
   • When audience, offer, or month changes, create a new campaign name. Everything else goes to utm_content.

Copy-paste AI prompt — generator

Use only these allowed values. Medium: email, cpc, social, referral. Source: google, facebook, linkedin, newsletter. Pattern for utm_campaign: brand-offer-audience-channel-yyyymm, lowercase, hyphens only. Inputs: Brand: “EcoCo”. Offer: “spring-sale”. Audience: “subscribers”. Channel: “email”. Month: “202503”. Generate 10 campaign names that match the pattern exactly and provide a two-column table: campaign_name and utm_content suggestions (5-10 chars, lowercase, hyphenated). No other fields, no punctuation beyond hyphens.

Copy-paste AI prompt — auditor

I will paste a list of URLs with UTMs. Validate each against these rules: lowercase only, hyphens only, required fields present (utm_source, utm_medium, utm_campaign), utm_source and utm_medium must be in this allowed list: source=[google, facebook, linkedin, newsletter], medium=[email, cpc, social, referral]. Flag any value not matching. For each invalid URL, return: original_url, issue_description, corrected_url (apply lowercase, hyphenation, and allowed mappings where obvious). Do not invent new values.

KPIs to watch

• UTM hygiene score: 1 – (invalid_or_off-vocab UTMs / total UTMs). Target ≥ 98%.
• Distinct utm_campaign count vs planned campaigns. Target ≤ +5% drift.
• Time to produce 100 links. Baseline now, target < 20 minutes with the system.
• QA defect rate per 100 links. Target ≤ 2.
• Percent of encoded values (when special chars present). Target 100%.

Mistakes and fast fixes

• Existing query string ignored: your builder always adds “?” — fix with the separator check shown above.
• Uppercase and spaces creep in: run a normalizer column and reference that in formulas, not the raw input.
• Platform synonyms (fb/Facebook/FACEBOOK): map with a synonyms table and VLOOKUP before finalizing.
• Overusing utm_campaign for creative: move creative differences to utm_content; keep campaigns stable.
• Manual edits post-launch: protect final columns; changes create attribution gaps.

1-week rollout

1. Day 1: Build Vocab tab with allowed lists and synonyms. Decide your campaign pattern.
2. Day 2: Set data validation and normalizer columns. Add the safe final URL formula.
3. Day 3: Use the generator prompt to create 20 options; paste into the sheet; finalize 5.
4. Day 4: Run the auditor prompt on the final URLs; fix issues; protect finalized cells.
5. Day 5: Launch 1–2 live campaigns using the sheet only. Time the build process.
6. Day 6: Export UTMs from analytics, compare distinct utm_campaign count to plan, log drift.
7. Day 7: Tighten vocab (add common synonyms), update rules, set monthly QA cadence.

Your move.

Quick win: spend 10 minutes now to turn vague promises into objective acceptance criteria — that single change will stop most scope disputes before they start.

The problem: verbal promises and fuzzy deliverables create scope creep, unpaid hours and eroded margins.

Why it matters: one unmanaged change can wipe out margin on a project and waste time on disputes. Clear, signed scope means predictable revenue, faster billing and fewer renegotiations.

My practical lesson: I’ve seen small businesses reduce scope disputes by >50% simply by adding a one-paragraph scope summary + 3 acceptance checklist items per deliverable. It’s low friction and high ROI.

What you’ll need

• A one-paragraph project bullet list (deliverables + exclusions).
• Milestone dates and payment terms (deposit, milestones, final).
• Your hourly rate and a contingency % for change-orders.
• Access to an AI text tool (copy/paste prompt below) and a lawyer for a quick review.

Step-by-step (do this now)

1. Write a 3–6 bullet project list: what you will and won’t do (5–10 min).
2. Run the AI prompt below to produce: a short agreement skeleton, explicit acceptance criteria under each deliverable, and a change-order clause (5–10 min).
3. Edit names, dates, prices; add your stop-work line: “no work on changes until signed approval & payment arrangement.” (10 min).
4. Attach the change-order form to every proposal and require client sign-off for out-of-scope work.
5. Send to your lawyer for a one-time review; save the finalized template for reuse.

Copy-paste AI prompt (use as-is)

“Draft a concise project agreement using these inputs: [paste project bullet list]. Include: deliverables with 2–4 acceptance criteria each, exclusions, milestones with dates, payment schedule (deposit/milestones/final), a change-order clause that explains how changes are requested, priced (hourly or fixed), and approved, estimated cost/time impact process, and a 3-step approval workflow. Keep language simple and client-friendly; highlight actions required by the client to avoid extra charges.”

What to expect

AI will deliver ~70–90% of the draft instantly. You’ll spend most time validating numbers, tightening acceptance criteria, and adding client details. After legal sign-off, rollout takes one client cycle.

Metrics to track

• % of proposals with signed scope before work starts.
• Number of scope-change requests per project.
• Time to approve a change-order (days).
• Revenue recovered from approved change-orders.

Common mistakes & fixes

• Vague deliverables — Fix: add 2–4 objective acceptance criteria per deliverable.
• Accepting verbal approvals — Fix: enforce signed or emailed approval plus payment terms before work begins.
• Relying on raw AI output — Fix: always edit numbers and run a legal review.

1-week action plan

1. Day 1: Draft three typical project briefs you sell most.
2. Day 2: Use the AI prompt to generate agreement skeletons for each brief.
3. Day 3: Create a one-page change-order template and stop-work clause.
4. Day 4: Legal review and finalize templates.
5. Day 5: Send next proposal with signed scope requirement.
6. Days 6–7: Measure baseline metrics and adjust language where approvals slip.

Your move.

Short answer: Yes — AI can produce onboarding sequences that move new buyers to value fast. Do one focused 3-email flow, measure activation, and iterate weekly.

The common problem: Buyers get polite, vague emails that don’t force a single clear action. Result: slow time-to-value and avoidable early churn.

Why this matters: Improving 7-day activation is the fastest, highest-leverage win for retention. If new buyers reach a meaningful outcome quickly they stick around and buy more.

Practical lesson: One action per email. Short copy. Clear timing. Track activation and fix the lowest-performing step first.

What you’ll need

• Buyer list with {{first_name}}, {{product_name}}, purchase date.
• Email tool with automation and conditional sends (segment by “completed action”).
• AI writing assistant (ChatGPT, Claude, etc.) to draft variants fast.

How to do it — step-by-step

1. Define the single 7-day activation (example: complete setup, use core feature once, or book a 15-min call).
2. Use the AI prompt below to generate a 3-email sequence: subject, preview, body (200–250 words), one CTA, timing, and a short success metric for each email.
3. Pick CTAs: Email 1 = low-friction setup task; Email 2 = lightweight feature use; Email 3 = human help or feedback capture.
4. Deploy: send Email 1 immediately; Email 2 at +48 hours to non-responders; Email 3 at +7 days to those still not activated.
5. Exclude buyers who complete the action from later emails with automation rules.
6. Review after 7 days (activation) and 30 days (cohort retention). Iterate one variable at a time (subject, CTA wording, or timing).

Copy-paste AI prompt (use as-is)

“You are an onboarding specialist. Create a 3-email onboarding sequence for new buyers of {{product_name}}. Objective: reduce time-to-first-value and increase 7-day activation. For each email provide: short subject and one alternative for A/B test, one-sentence preview text, body 150–220 words max, one clear CTA (single action), suggested timing (send immediately, +48 hours, +7 days), personalization tokens {{first_name}} and {{product_name}}, and a one-line success target for the email (e.g., 25% CTA click). Keep tone friendly, concise, and action-oriented.”

Metrics to track

• Open rate (target 40%+ for buyer lists)
• CTA click rate (target 15–30%)
• 7-day activation (primary KPI)
• 30-day retention / churn for cohort

Common mistakes & fixes

• Multiple CTAs — fix: one CTA per email.
• Generic language — fix: add product-specific examples and {{first_name}} personalization.
• Static timing — fix: pause the sequence for buyers who complete the action and accelerate for high-value buyers.

1-week action plan

1. Day 1: Run the prompt and pick CTAs.
2. Day 2: Edit copy, set up automation, include exclusion rules.
3. Day 3: Launch Email 1 to today’s buyers.
4. Day 5: Check opens and clicks; swap subject line if open <35% or CTA click <12%.
5. Day 7: Measure 7-day activation and adjust the weakest email (copy or timing).

Your move.

— Aaron

Quick win: Yes — use AI to create the skeleton while you focus on the analysis. That split speeds delivery and preserves control.

The problem: You spend hours structuring reports; the analysis is the value. AI can create consistent, appropriate outlines, but only if you give it the right inputs and guardrails.

Why this matters: Faster iteration, fewer structural revisions, clearer stakeholder reviews. You reduce time-to-decision and increase the chance your analysis drives action.

What I’ve learned: Use AI for format, headings, and recommended evidence placement. Keep human-in-the-loop for judgement, nuance, and interpretation. Combine a short brief + examples + constraints and you’ll get usable outlines first pass.

• Do: Give the AI the report purpose, audience, required length, and a few data points or file names.
• Do not: Assume the AI knows your audience priorities — tell it.

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

1. Prepare a 2–4 sentence brief: purpose, audience, key takeaway.
2. Collect core inputs: dataset names, KPIs, charts to reference, and one example report format you like.
3. Run the AI with a clear instruction to produce an outline with headings, suggested word counts per section, and where to place data citations.
4. Review and annotate the outline (5–15 minutes). Keep only structural edits — not phrasing.
5. Hand the annotated outline to your analyst team or use it to write your analysis; repeat if needed.

Copy-paste prompt to use (plain English):

“You are an executive report writer. Create a detailed outline for a 1,200–1,500 word report titled: [Report Title]. Audience: [Role, e.g., C-suite sales leaders]. Objective: [decision or insight required]. Include: suggested section headings, 1–2 sentence purpose for each section, recommended word count per section, placeholders for specific charts or tables (name them), and 3 recommended calls-to-action. Use a concise, business tone.”

Worked example (output you should expect):

• Executive Summary (150–200w): 2-sentence conclusion + 1 line of recommended action.
• Key Metrics Snapshot (150w): table of Revenue, GM%, Conversion, YoY change — reference Chart 1.
• Drivers Analysis (400w): channel performance, top 3 wins/losses with evidence bullets and links to Chart 2/Appendix A.
• Risks & Sensitivities (200w): 3 risks with impact and likelihood; recommended mitigations.
• Recommendations & Next Steps (200–300w): prioritized actions with owners and timelines.
• Appendix/Data Sources: list files and assumptions.

Metrics to track:

• Time to first outline (target <15 minutes)
• Outline acceptance rate (% accepted without structural edits)
• Revision rounds per report (target ≤2)
• Time from draft to final (days)

Common mistakes & fixes:

• Vague brief → AI produces vague outline. Fix: give audience and decision explicitly.
• Over-editing structure → creates churn. Fix: limit structural review to 15 minutes.
• Missing data placeholders → AI omits citations. Fix: include chart/table names in inputs.

1-week action plan:

1. Day 1: Draft 3 one-paragraph briefs for upcoming reports.
2. Day 2: Run AI to produce outlines; pick one and review.
3. Day 3: Use outline to write or collect analysis; annotate gaps.
4. Day 4: Finalize one report and measure time saved.
5. Day 5: Tweak prompt and checklist based on feedback; document template.

Next step: run the provided prompt with one brief and share the outline you get. I’ll show which edits reduce revisions and give you a template to reuse.

Your move.

Want feedback that students can act on — fast and at scale? Use AI to handle routine checks and deliver one clear revision step that actually moves writing forward.

The gap: Teachers spend hours on surface edits and long comment threads. That buries the coaching that changes thinking: argument clarity, evidence use, and revision strategy.

Why it matters: If you free 30–60 minutes per assignment by automating routine feedback, you can deliver targeted coaching to every student — improving draft quality, revision engagement, and final grades.

What I learned: Keep AI outputs small, specific and prescriptive. Two corrections + one 15-minute task beats a page of marginalia every time.

Do / Do-not checklist

• Do feed the AI a clear 3–5 point rubric and an example comment.
• Do limit feedback to 2 fixes + 1 focused revision task.
• Do anonymize when batch-processing.
• Do not send AI output verbatim without your light edit.
• Do not rely on AI for subjective judgments like voice or creativity.

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

1. Prepare: Create a 4-item rubric (thesis, evidence, organization, grammar). Keep labels explicit (e.g., Thesis: specific/partial/none).
2. Collect text: Have students paste one paragraph or a 300–500 word draft into your tool.
3. Run the prompt below for each paragraph/section. Expect a 40–80 word feedback block: 1 praise, 2 fixes, 1 15-minute task.
4. Review and tweak tone (30–90 seconds per student) then return feedback with a clear deadline for the 15-minute revision.
5. Require a short resubmit or reflection to track learning.

Copy-paste AI prompt (use as-is)

“You are an experienced writing tutor. Evaluate the paragraph below using this rubric: 1) Thesis (specific/partial/none), 2) Evidence (strong/weak/none), 3) Organization (clear/uneven/confusing), 4) Clarity & grammar (good/needs revision). Provide: one-sentence praise, two short corrective suggestions (each one sentence), and one concrete 15-minute revision task the student can complete now. Keep tone encouraging and concise. Here is the paragraph: [PASTE PARAGRAPH]”

Worked example

Student paragraph: “Climate change is bad because weather changes and crops fail. More people should care because it’s important for the future.”

Expected AI output: “Praise: You identify a clear concern about climate change. Fix 1: Turn ‘More people should care’ into a specific thesis — what action or understanding do you want? Fix 2: Add one concrete piece of evidence (statistic or example) to support the claim. 15-minute task: Rewrite the second sentence as a clear thesis and add one statistic or named example to support it.”

Metrics to track (KPIs)

• Average teacher time per submission (target: < 3 minutes).
• Revision uptake rate (students who complete 15-minute task).
• Average rubric score improvement between drafts.
• Student satisfaction with feedback (pulse survey).

Mistakes & fixes

• Mistake: Long, unfocused AI comments → Fix: Force 2 fixes + 1 task in prompt.
• Mistake: Using AI to grade voice → Fix: AI prepares notes; teacher assigns subjective grade.
• Mistake: Flooding with edits → Fix: Limit to 15-minute actionable task each round.

7-day rollout (practical)

1. Day 1: Create rubric and copy the prompt into your tool.
2. Day 2: Test on 3 paragraphs and tweak tone.
3. Day 3: Run for one assignment; review before sending.
4. Day 4: Require 15-minute revision from students.
5. Day 5–7: Track time saved and revision uptake; adjust prompts.

Start small: one paragraph per student, one rubric, one clear revision task. Measure time saved and draft improvement. Your move.

Hook: Good call on small, testable wins — start with one tiny routine and make it reliable before expanding.

Problem: You waste 10–30 minutes each morning collecting notes, emails and tasks. That friction kills momentum.

Why it matters: Reduce setup time, get a clear prioritized plan each morning, and free focused time for high-value work.

Quick lesson: I’ve seen non-technical users cut prep time by 70% with a single scheduled flow + a hotkey to open and paste into an AI app. The trick: predictable inputs, fixed timing, simple retries.

• Do start with one task (open AI page & paste text).
• Do test with dummy text; confirm each step manually first.
• Don’t send passwords or sensitive client info to any third-party AI until you control the flow.
• Don’t attempt complex API calls until the basic flow is stable.

What you’ll need

1. Windows PC, AutoHotkey installed; Microsoft account for Power Automate if you want scheduled collection.
2. An AI web app you use (test account or dummy data).

Step-by-step — AutoHotkey (local hotkey)

1. Pick a hotkey: e.g., Ctrl+Alt+A.
2. Manual test: select text, Ctrl+C, open AI site, click input, Ctrl+V, press Enter.
3. Create a .ahk file with these lines (paste as-is):

^!a::

Send, ^c

Sleep, 300

Run, chrome.exe –new-window “https://your-ai-site.example&#8221;

Sleep, 1500

Send, ^v

Sleep, 200

Send, {Enter}

Return

Step-by-step — Power Automate (scheduled)

1. Trigger: Recurrence (daily time you choose).
2. Action: Get files from OneDrive or read specific Outlook folder.
3. Action: Compose — concatenate needed text into one body.
4. Action: Send yourself an email with that body (or save .txt to OneDrive).

Worked example — Morning Summary

1. Power Automate runs at 7:55am, collects notes into one email.
2. At 8:00am you press Ctrl+Alt+A, AutoHotkey opens AI page and pastes the email body.
3. Ask the AI for a prioritized to-do list — you’re ready in 20 seconds.

Copy-paste AI prompt (use inside your AI app)

Summarize the following notes into a prioritized to-do list. For each item give: a one-line summary, priority (High/Med/Low), estimated time (Short: 0–15m, Medium: 15–60m, Long: 1–4h), and one sentence explaining why it matters. Then provide a suggested 3-step plan for the top two items: [PASTE NOTES HERE]

Metrics to track (KPIs)

• Time to prepare morning plan (before vs after).
• Flow success rate (%) — runs that finish without manual intervention.
• Manual edits per AI output (count per day).
• Tasks completed from AI list within the day.

Common mistakes & fixes

• Timing issues: increase Sleep values; use WinActivate before sending keys.
• Wrong text copied: add a brief clipboard-check step — paste into Notepad first for testing.
• Power Automate fails: check connector permissions and view run history to see where it stops.

1-week action plan

1. Day 1: Build AutoHotkey hotkey that opens the AI site and pastes dummy text. Verify reliability.
2. Day 2–3: Create Power Automate flow to collect notes and email them to you; confirm content quality.
3. Day 4: Connect steps — use hotkey to paste the emailed body. Track time saved each morning.
4. Day 5–7: Tweak pauses/filters, measure flow success rate and manual edits, iterate once per day.

Results you should expect: shave off 10–20 minutes daily on prep within a week; reduce morning decision friction; a repeatable flow you can expand.

Questions: are you using Outlook or another mail app for your notes? Tell me and I’ll give the exact Power Automate trigger and folder setup.

— Aaron

Your move.

Good call — your step-by-step mock interview approach is exactly the repeatable practice most PM candidates need. I’ll add a results-first version: make every session drive measurable improvement in how you frame problems, pick metrics and communicate trade-offs.

Why this matters

If you can shave 30–60 seconds off your framing, add one clear metric in every answer, and show one trade-off confidently, you’ll outperform most candidates. Those changes are small, measurable and interview-winning.

What you’ll need

• A chat AI (LLM) and a recorder or notes app.
• 5–10 prompt templates across domains (SaaS, consumer, mobile, marketplace).
• A compact rubric: Framing (30%), Metrics (25%), Trade-offs (20%), Solution clarity (15%), Communication (10%).

How to run a focused mock (step-by-step)

1. Set session goal: e.g., improve 7-day retention framing or extracting the right baseline metric.
2. Load an AI persona: “You are a senior PM interviewer at a Series B SaaS company.”
3. Start 20-minute interview: Candidate speaks aloud; AI asks 3–5 follow-ups. Record or transcribe.
4. Immediately score using the rubric (0–5 per area). Note one repeatable weakness.
5. Ask the AI to provide a 3-point improvement plan tied to that weakness; implement in the next mock.

Copy-paste AI prompt (use exactly)

“You are a senior product manager interviewing a candidate for a mid-stage SaaS PM role. Ask a single open-ended product-sense question. After the candidate answers, ask 3 probing follow-ups that focus on user segmentation, the single most important metric, and realistic constraints. Then give a 5-point, prioritized feedback summary: 1) top gap to fix, 2) exact phrase or sentence to use for framing, 3) one metric to add, 4) one trade-off to state, 5) a 60-second elevator answer example. Keep it practical and prescriptive.”

Metrics to track

• Average rubric score per session (target: +0.5 every 3 sessions).
• Time to first clear metric in answer (target: under 60s).
• Number of trade-offs explicitly stated (target: 1+ per answer).
• Interviewer or peer feedback consistency (agree on top 2 gaps).

Common mistakes & fixes

• Treating AI as final judge — Fix: compare AI feedback with one human replay each week.
• Over-preparing scripts — Fix: force a 60s cold answer before any notes.
• Vague metrics — Fix: require a baseline and target for every experiment proposed.

7-day action plan (practical)

1. Day 1: Build 5 prompts, set rubric, run 1 mock (record).
2. Days 2–4: Do 2 timed mocks/day. Score and correct one recurring gap each evening.
3. Day 5: Run a peer review — share 1 transcript, get human feedback.
4. Days 6–7: Focus drills on top 2 weaknesses (metrics framing, trade-offs) with 60s answers.

Expect small, measurable lifts quickly if you practice deliberately: improve framing, name a metric early, and state one trade-off each time.

—Aaron

Your move.