Draft Workforce-Planning Scenario Memos

Workflow area: HR & Recruiting · Tier: Advanced · Hands-on time: ~45 minutes

Scenario memos go wrong in two ways: the numbers live only in someone's head, or the numbers live in a model nobody can explain. The discipline here is strict separation — Excel owns the arithmetic, the AI owns the structure and prose, and you own the assumptions, stated in writing where they can be challenged.

When to use — and when not to

• Use it for the recurring planning cyclesannual budget, mid-year reforecast, or a leadership request to compare defined options.
• Use it when the assumptions are decided or decidablegrowth targets, attrition estimates, salary inflation. The memo's job is to show consequences of assumptions, not to hide them.
• Skip the AI for the raw modellingheadcount math (FTEs × loaded cost × months) belongs in Excel where formulas are auditable. AI that "calculates" your scenario totals will be confidently wrong; it drafts around numbers you computed.
• Skip it for restructure executiondeciding who is affected by a restructure is a human, legal and ethical process. This playbook compares aggregate scenarios; it never touches named individuals.

The recipe

1. 1.

   Build the base model in Excel. One sheet, aggregate only: rows by department × level band, columns for current FTE, average loaded cost (salary + benefits + overheads), planned backfills, and expected attrition. No names. Add a small assumptions block at the top — attrition rate, salary inflation, hiring lead time, cost per hire — each in its own labelled cell so scenarios can flex them.

2. 2.

   Define the three scenarios in numbers, not adjectives. Copy the base sheet three times:

   • •
     Growth — e.g. +15% FTE in product and sales, backfill all attrition, 8-week hiring lead time.
   • •
     Freeze — no new requisitions, backfill only critical roles (define "critical" now, in writing), natural attrition runs.
   • •
     Restructure — e.g. −10% cost via span-of-control consolidation, with one-time restructuring cost estimated per affected FTE.

   Let Excel compute the 12-month cost line, end-of-year FTE, and cost delta versus base for each. Sanity-check totals by hand — one wrong cell reference here invalidates the memo.

3. 3.

   Stress-test the scenario design before writing. Paste only the assumptions block and the three scenario summary tables into Claude or ChatGPT:

   Copy

   Here are my workforce scenario assumptions and computed outcomes
   (12-month view). Act as a skeptical CFO and CHRO reviewing them.
   
   - Which assumptions are doing the most work in each scenario, and
     what happens directionally if they are wrong by 25%?
   - What second-order effects are missing? (e.g. freeze → attrition
     rises above the assumed rate; restructure → backfill of lost
     critical skills; growth → manager span and onboarding load)
   - Which risks are asymmetric — cheap to insure against, expensive
     to ignore?
   - What would each scenario do to delivery capacity, not just cost?
   
   Do not recalculate my numbers. Critique the model design.
   

   Fold the good challenges back into the Excel model — often this adds a sensitivity row (attrition ±2 points) that becomes the most-discussed number in the meeting.

4. 4.

   Draft the memo from the final tables.

   Copy

   Draft a 1.5-page workforce scenario memo for the executive team.
   Use ONLY the numbers in these tables — do not compute, round, or
   extrapolate anything yourself.
   
   [PASTE ASSUMPTIONS BLOCK + THREE SCENARIO SUMMARY TABLES]
   
   Structure:
   - Context: why we are choosing between these three, one paragraph.
   - Scenario comparison table: 12-month cost, cost delta vs base,
     end FTE, capacity impact, top risk — one row per scenario.
   - One paragraph per scenario: what it buys, what it risks, what
     it forecloses.
   - Sensitivities: the two assumptions that most change the answer.
   - Recommendation placeholder: "[LEADERSHIP DECISION]" — do not
     write a recommendation.
   - Decision needed and by when.
   
   Tone: neutral between scenarios. No euphemisms — "restructure"
   stays "restructure", not "rightsizing journey".
   

5. 5.

   Verify and decide the recommendation yourself. Check every number in the memo against the spreadsheet — models transpose table cells more often than you would believe. Then write the recommendation paragraph in your own words, or leave it genuinely open if the memo's job is to frame the choice. The AI structured the comparison; the recommendation is a leadership act.

Make it yours

• Capacity view for delivery orgs. Add a "deliverable roadmap items" line per scenario — executives often care more about what ships than what it costs, and the freeze scenario's real price is usually there.
• Quarterly phasing. Break the 12-month view into quarters; a freeze looks cheap in Q1 and expensive in Q4, and phased tables surface that honestly.
• Board-pack version. Feed the finished memo through pb-19's doc-to-deck workflow for the three-slide version: comparison table, sensitivity chart, decision slide.

Quality check

• All arithmetic lives in Excelevery memo number traces to a cell, none were generated by the model.
• Assumptions are printed in the memoanyone can challenge the attrition rate because they can see it.
• Scenarios are neutral until the recommendationthe drafting didn't smuggle in a preference.
• The sensitivity section existsreaders see which assumption swings the answer.
• No individual-level detail anywhereaggregate bands only, in the model and the memo.