Salary Research

Trigger: /salary-research [company] [role] [level] [location]

Examples:

• /salary-research Google PM L5 Seattle
• /salary-research Anthropic Senior PM San Francisco
• /salary-research Stripe Growth PM L3 Remote
• /salary-research Meta Product Manager E5 New York

Role Type Detection

Before executing any step, read career-plan.md and detect the user's target function from their target role title:

• Product Management (PM): Product Manager, Group PM, Director of Product, VP Product, CPO
• Software Engineering (SWE): Software Engineer, Staff Engineer, Principal Engineer, Engineering Manager
• Design: Product Designer, UX Designer, Design Lead, Head of Design, VP Design
• Data Science / Analytics: Data Scientist, ML Engineer, Analytics Manager, Head of Data
• Marketing: Marketing Manager, Growth Marketing, Brand Manager, VP Marketing, CMO
• Customer Success / Sales: CSM, CS Manager, Director of CS, Account Executive, Sales Manager, CRO
• Other: Infer function from role description

All instructions below use PM as the default. When the detected function is NOT PM, substitute function-appropriate equivalents as marked with [FUNCTION-ADAPTIVE] notes throughout.

Inputs

Required:

• Company name
• Role title
• Level (company-specific level if known, e.g., L5, E5, IC4; or generic like "Senior")
• Location (city or "remote")

Optional:

• Specific comp components to focus on (e.g., "mainly interested in equity structure")
• Whether this is pre-IPO (changes equity analysis significantly)
• Your current comp (for comparison context)

Auto-read (do not ask the user for these -- read them directly):

• context-library/career-plan.md -- dream offer target, walkaway number, level preference
• context-library/experience-library.md -- to map user's experience to the appropriate level
• context-library/qa-master.md -- previously communicated comp expectations
• insider-data/company-intel/[company].md -- any comp data, band info, or negotiation intel (if exists)

Process

Step 1: Web Research - levels.fyi

Search for compensation data on levels.fyi:

• levels.fyi [company] [level] [role] total compensation
• levels.fyi [company] product manager [location]
• levels.fyi [company] [level] salary

Extract:

• Total compensation range (P25, P50, P75 if available)
• Base salary range
• Equity range (annual value)
• Bonus range
• Number of data points (more data points = higher confidence)
• Recency of data (data from 2+ years ago is less reliable)
• Location-specific adjustments if available

Step 2: Web Research - Glassdoor

Search for compensation data on Glassdoor:

• Glassdoor [company] [role] salary [location]
• Glassdoor [company] product manager compensation

Extract:

• Base salary range
• Additional cash compensation (bonus)
• Number of salary reports
• Note: Glassdoor often underreports equity, so treat as base salary reference primarily

[FUNCTION-ADAPTIVE] Use function-appropriate compensation data sources. SWE: levels.fyi (excellent SWE data), Glassdoor, Blind, teamblind.com. Design: levels.fyi, Glassdoor, AIGA salary survey, Coroflot salary guide. Data Science: levels.fyi, Glassdoor, Burtch Works DS salary study. Marketing: Glassdoor, Built In salary data, ANA salary survey. CS/Sales: Pavilion CS Compensation Survey, RepVue benchmarks, Bridge Group CS Comp Report, Glassdoor. Do NOT default to PM-specific search queries for non-PM roles.

NOTE: levels.fyi and Glassdoor have significantly less Design compensation data than SWE or PM. For Head of Design / Design Director targets, data sparsity is a real problem. Lower the confidence rating by one level (e.g., High → Medium) and supplement with: Dribbble salary survey, AIGA salary guide, and comparable-company estimation (anchor to the engineering leadership band at the same company, then apply the typical 10-15% Design discount). For CS roles, use Pavilion CS Compensation Survey and Bridge Group CS Comp Report as primary sources, as levels.fyi CS data is thin.

IMPORTANT: Marketing comp varies significantly by sub-function. Growth/performance marketing roles at PLG companies typically pay 10-20% above brand marketing roles at the same level and company. Product marketing comp is typically between growth and brand. When researching, segment queries by sub-function: search 'growth marketing director salary [company]' not just 'marketing director salary.' The ANA salary survey and Built In both allow sub-function filtering — use it. If the user is transitioning between marketing sub-functions (e.g., brand to growth), note the potential comp uplift as a positive factor.

Step 3: Web Research - Blind

Search for recent compensation data points on Blind:

• Blind [company] [level] offer [current year]
• Blind [company] product manager compensation [current year]
• teamblind.com [company] [level] TC

Extract:

• Recent individual data points (people sharing their offers/current comp)
• Negotiation outcomes (what people were able to negotiate up to)
• Any notes about band limits, refresher grants, or promotion comp changes
• Note: Blind data is self-reported and may skew high (selection bias)

Non-US Markets

If the target location is outside the US:

1. levels.fyi has limited international data -- use if available but note low confidence
2. Add Glassdoor [location] as primary source
3. Search for local compensation surveys: '[country] product manager salary survey [year]', regional salary platforms
4. Regional sources:
   • UK: Glassdoor UK, Reed.co.uk, Otta
   • Germany: Glassdoor DE, StepStone, Kununu
   • Singapore: Glassdoor SG, NodeFlair, MyCareersFuture
   • India: Glassdoor India, AmbitionBox, Naukri
   • Australia: Glassdoor AU, Seek, Hays Salary Guide
   • MENA: GulfTalent, Bayt.com, Hays Gulf Salary Guide, Robert Half Middle East Salary Guide.
5. Note comp structure differences: UK pension contributions (8-15%), European equity taxation, sign-on bonuses less common outside US
6. If company-intel comp data is US-only, flag: 'Comp ranges in the intel file are US-based. Apply location adjustment for [target location].'

Regional Comp Structures

When target market is non-US, show comp in the local framework:

UK: Annual gross salary + bonus + equity (if startup). Pension contribution (employer typically 3-8%). Private health insurance as benefit. Show in GBP. No 401k equivalent discussion.

Germany: Annual gross salary. Note: 13th-month salary common (effectively ~8% bonus). Weihnachtsgeld (Christmas bonus) at many companies. Company car value (5-10K EUR/year, common at senior levels). Betriebsrente (company pension). Public health insurance (employer pays ~7.3% of gross). Show effective total comp including all components.

India: CTC (Cost to Company) is the primary number. Break down: CTC = Basic + HRA + Special Allowance + PF (employer contribution) + Gratuity + Variable Pay. Show actual in-hand (take-home) number vs CTC. Variable pay is often 10-20% of CTC and may not be guaranteed. ESOP vesting schedules vary widely at Indian startups.

Middle East (UAE/Qatar): Tax-free income — gross = net. Include housing allowance (30-40% of base), education allowance, annual flight home allowance, end-of-service gratuity (UAE: 21 days/year for first 5 years, 30 days/year after). These benefits are substantial and standard.

Cross-Market Relocation (Non-US Current Comp to US Target)

If the user is currently compensated in a non-US market and targeting US roles (or any cross-market move where comp structures differ significantly):

1. Do NOT treat the current non-US comp as a pay-cut baseline. European, Japanese, and most non-US tech markets pay 40-60% less in total comp than US equivalents at the same level. A Senior PM earning EUR 130K in Munich is not "underpaid" -- they are at market for Munich. The comparison baseline must be the US market rate for the target role, not the user's current comp.
2. Add a "Market Transition" section to the output that shows:
   • Current comp in local currency and USD equivalent
   • US market rate for the equivalent role/level (from levels.fyi, Glassdoor)
   • The gap as both a dollar amount and a percentage
   • A framing note: "This gap reflects the US-vs-[market] comp differential, not a personal raise. Your target of $[X] is within the normal US range for this level."
3. Cost-of-living adjustment context: If the user is relocating from a lower-cost market to SF/NYC, note that the comp increase is partially offset by higher living costs. Provide a rough comparison: "SF cost of living is approximately [X]% higher than [current city]. Adjusted for living costs, $[US target] provides roughly [X]% more purchasing power than [current comp in local currency]."
4. Comp question coaching: Flag that the cross-market jump makes the comp question especially dangerous. Add: "CRITICAL: Never state your current non-US comp in interviews. The number will sound low to a US recruiter and anchor the negotiation against you. Always frame in terms of US market rate for the role: 'Based on levels.fyi, Senior PMs at [Company] earn $[X]-$[Y]. I'm targeting within that range.'" Cross-reference qa-master.md for any existing comp scripts.
5. Currency and tax notes: Note relevant differences in tax treatment, equity taxation, social contributions, and benefits that affect take-home pay comparisons. European comp often includes employer-paid social contributions (health, pension) that are separate from the stated salary, while US comp typically does not.

Step 4: Insider Data Check

Check insider-data/company-intel/[company].md for:

• Compensation bands or ranges mentioned in the profile
• Negotiation flexibility notes (is this company known for moving on comp?)
• Equity structure (RSU schedule, refresh policy, cliff details)
• Any insider tips about comp discussions

Insider Data vs User Research Cross-Check: After reading the company-intel comp ranges, compare them against any comp data the user has already documented in target-companies.md, qa-master.md, or career-plan.md. If the user's self-researched ranges (from levels.fyi, Glassdoor, Blind, or conversations with contacts) diverge significantly (20%+ gap) from the company-intel ranges, flag this explicitly: "The company-intel file lists [range] for this level, but your own research in [file] shows [range]. This divergence may be due to: (1) different level interpretations (company-intel may reference a higher level), (2) stale data in one source, (3) different comp components included (some ranges include equity at paper value, others at cash-equivalent). I recommend treating the LOWER range as more conservative and the HIGHER range as the ceiling in a strong-leverage scenario. Clarify which level the company-intel range refers to before anchoring your negotiation."

This cross-check prevents candidates from anchoring their expectations to inflated or deflated numbers from a single source. It is especially important when company-intel comp ranges include pre-IPO equity at paper value while the user's research reflects cash-equivalent estimates.

Step 5: Level Mapping

[FUNCTION-ADAPTIVE] Function-specific career ladder mapping. SWE: Junior -> Mid -> Senior -> Staff -> Principal -> Distinguished (L3-L8). Design: Junior -> Mid -> Senior -> Lead -> Principal -> VP. Data Science: Analyst -> Data Scientist -> Senior DS -> Staff DS -> Principal DS. CS: CSM -> Senior CSM -> CS Manager -> Director CS -> VP CS -> CCO. Marketing: Coordinator -> Manager -> Senior Manager -> Director -> VP -> CMO.

[FUNCTION-ADAPTIVE] CS/Sales roles often have OTE (On-Target Earnings) structures with variable comp tied to retention/revenue metrics. When detected: model total comp at 80%, 100%, and 120% attainment. Research accelerator curves above target. Factor in: ramp quota for new hires, measurement periods, and guaranteed variable in first quarter.

Using experience-library.md, map the user's experience to the appropriate level at this company:

• Years of PM experience
• Scope of work (IC vs team lead, 0-to-1 vs optimization, team size influenced)
• Impact level (metrics moved, revenue influenced, users affected)
• Compare against typical level expectations at this company (from levels.fyi career data or company-intel)

Career Changer Level Translation: If the user is coming from a NON-PM role (consulting, engineering, design, etc.), the level mapping requires special handling:

• Do NOT map consulting seniority directly to PM seniority. A McKinsey Engagement Manager (EM) manages teams and drives $200M+ engagements, which is Director-level scope in consulting terms. But transitioning to PM, the user will typically be hired at PM or Senior PM level (1-2 levels below their consulting seniority) because they lack PM-specific execution experience (sprint planning, A/B testing, shipping iteratively).
• Provide an explicit cross-function level translation: "[User's title] at [current company] = [equivalent PM level] for hiring purposes. Your consulting scope is [level], but PM roles will level you based on PM-specific experience, which is [0/limited/partial]. Expect to be hired at [level], not [higher level]."
• The comp implication is critical: if the user's current comp at their non-PM role exceeds the PM-level band at the target company, this is a GUARANTEED pay-cut scenario. Flag explicitly: "McKinsey EM comp ($245K) is at or above the P50 for PM at [company] ($XXK). You are trading seniority (and possibly comp) for a career change. Make sure the non-comp factors justify this."
• Research whether the target company has precedent for hiring career changers at higher levels. Some companies (especially those that value analytical rigor) may level a McKinsey EM at Senior PM rather than PM. Check recent PM hires from consulting backgrounds if this data is available.

Flag if there is a mismatch:

• "Your experience suggests [level], but you're applying for [different level]. This may affect comp range expectations."
• "At [company], [level] typically requires [X years / Y scope]. Your experience could support [level] or [level+1]."

Cross-Company Level Translation:

• When the user is coming from a company with a different leveling system, provide an explicit level translation table. E.g., Amazon L6 = Google L5 = Meta E5. This is critical because level titles mislead -- "Senior PM" means different things at different companies.
• If the user is at a level at their current company where comp is HIGHER than the equivalent level at the target company (common when moving between companies with different comp philosophies, e.g., Amazon L7 $350K-$550K vs a target where equivalent level pays $300K-$450K), flag this explicitly: "Your current level's comp at [current company] exceeds the typical range at [target company] for the equivalent role. You may be in a pay-cut scenario. See below."

Pay-Cut Detection:

• Compare the user's current total comp (from career-plan.md) against the target level's P50 at the target company.
• If current comp > target P50, flag: "Your current comp of $[X] is above the P50 ($[Y]) for this role. Expect a comp discussion where you are explaining why you are willing to move for less, not asking for more. See /negotiate pay-cut scenario handling."
• If current comp > target P75, flag more strongly: "Your current comp of $[X] exceeds even P75 ($[Y]) for this role. You are almost certainly taking a pay cut. Make sure the non-comp factors justify this move."
• Research whether the target company offers equity upside, refresher grants, or faster promotion velocity that could close the gap over 2-3 years.

Step 6: Synthesize and Position

Combine all data sources into a unified picture. For each data source:

• Note the confidence level (high/medium/low based on data point count and recency)
• Note any biases (Glassdoor skews low on equity, Blind skews high overall, levels.fyi is most balanced)
• Derive a recommended range that weighs the sources appropriately

Compare against the user's targets from career-plan.md:

• Dream offer target vs market reality
• Walkaway number vs market floor
• Is the user's target realistic for this company/level? If not, flag it with data.

Step 7: Generate Negotiation Context

Provide strategic context for the upcoming negotiation:

• Is this company known for negotiating? (from company-intel or Blind reports)
• What components are most flexible? (base is often banded; equity and sign-on often have more room)
• Recent hiring velocity (from target-companies or web search -- high velocity = more leverage for candidates)
• Any company-specific quirks (e.g., "Google's L5 band tops out at $X base" or "Anthropic includes equity refreshers annually")

Output

# Salary Research - [Company] [Role] [Level] [Location]
Generated: [date]
Data confidence: [high / medium / low] (based on data point count and recency)

## Market Data

### levels.fyi
- **Total comp range:** $[X] - $[Y] ([N] data points, last updated [date])
  - P25: $[X]
  - P50 (median): $[X]
  - P75: $[X]
- **Base range:** $[X] - $[Y]
- **Equity range:** $[X] - $[Y] per year
- **Bonus range:** $[X] - $[Y]
- **Confidence:** [high/medium/low] — [N] data points, [recency assessment]

### Glassdoor
- **Base salary range:** $[X] - $[Y] ([N] salary reports)
- **Additional cash comp:** $[X] - $[Y]
- **Confidence:** [high/medium/low] — [note: typically underreports equity]

### Blind
- **Recent data points:**
  - [Date]: [Level] offered $[X] TC ([base/equity/bonus breakdown if shared])
  - [Date]: [Level] negotiated from $[X] to $[Y]
  - [Date]: [Level] current comp $[X]
- **Confidence:** [high/medium/low] — [note: self-reported, may skew high]

### Company Intel (insider-data)
[If company-intel file exists:]
- Comp bands: [any data]
- Negotiation flexibility: [flexible / moderate / rigid]
- Equity structure: [RSU/ISO/NSO, vesting schedule, refresh policy]
- Notes: [any relevant tips]

[If no company-intel file:]
- No insider data available for [company]. Relying on public sources.

## Compensation Breakdown

| Component | Low (P25) | Mid (P50) | High (P75) | Notes |
|-----------|-----------|-----------|------------|-------|
| Base | $[X] | $[X] | $[X] | [banded at most companies] |
| Equity | $[X]/yr | $[X]/yr | $[X]/yr | [vesting schedule, type] |
| Bonus | $[X] | $[X] | $[X] | [target % if known] |
| Sign-on | $[X] | $[X] | $[X] | [typical for new hires] |
| **Total Y1** | **$[X]** | **$[X]** | **$[X]** | [includes sign-on] |
| **Total Annual** | **$[X]** | **$[X]** | **$[X]** | [steady state] |

[If pre-IPO company:]
**Equity risk note:** [Company] is pre-IPO. Equity values above are based on
last known valuation ($[X], [date]). Actual value depends on future outcomes.
Apply a [30-50%] discount when comparing to public company RSU offers.

**Illiquid equity analysis (for pre-IPO companies):**
- **Liquidity status:** [Can shares be sold on secondary markets? Is there a tender offer program?]
- **Exercise cost:** [For ISOs/NSOs: strike price * shares = $X out-of-pocket to exercise. This is real money at risk.]
- **Tax implications:** [ISOs: AMT risk on exercise. NSOs: taxed as income on exercise. RSUs: taxed on vesting but no exercise cost.]
- **409A valuation date:** [When was the last 409A? How stale is it?]
- **Dilution risk:** [What funding round? Earlier rounds = more future dilution. Note: a Series A grant will be diluted significantly by the time of IPO.]
- **Realistic value scenarios:**
  - Bull case (IPO at 2-3x current valuation): equity worth $[X]/yr
  - Base case (IPO at current valuation, minus dilution): equity worth $[X]/yr
  - Bear case (down round, acquisition at discount, or no liquidity event in 5+ years): equity worth $[X]/yr or $0
- **Cash-equivalent comparison:** "To compare this offer against a public company RSU offer, use the base-case equity value with a 30-50% liquidity discount. Your $[X] equity grant has a risk-adjusted annual value of ~$[Y], compared to $[Z] in immediately-liquid RSUs at a public company."
- **If NO valuation data is available (very early stage, pre-Series A, or no public funding data):** State: "Equity valuation data unavailable. Treat equity as lottery upside, not compensation. Evaluate this offer on cash comp alone (base + bonus + sign-on). If cash comp alone does not meet your walkaway number, this offer relies on equity that may never be liquid."

## Your Position

### Level Mapping
Based on your experience library:
- **Your experience:** [X years PM, scope summary, impact summary]
- **Maps to:** [Level] at [Company]
- **Confidence:** [high/medium/low]
- [If mismatch: "You're targeting [Level] but your experience suggests [Level]. Consider [action]."]

### Cross-Company Level Translation
| Your Current | Target Company Equivalent | Comp Implication |
|-------------|--------------------------|------------------|
| [Current company] [Level] | [Target company] [Level] | [Higher / Similar / Lower comp band] |

[If pay-cut detected:]
**Pay-Cut Alert:** Your current comp of $[X] at [current company] is above the [P50/P75] for [target level] at [target company]. This is a -$[delta] ([X]%) reduction. Ensure the non-comp factors (equity upside, career trajectory, mission, scope) justify this trade-off. See `/negotiate` for pay-cut negotiation strategy.

### Targets vs Market
| | Your Target | Market P50 | Delta |
|---|------------|------------|-------|
| Dream offer | $[X] | $[X] | [+/- $X] |
| Walkaway | $[X] | $[X] | [+/- $X] |

- **Dream offer assessment:** [Realistic / Stretch / Unrealistic] — [1 sentence with data]
- **Recommended ask:** $[X] total comp
  - Justification: [specific data source + your unique value that warrants this position in the range]
- **Walkaway number:** $[X] (from career-plan.md)
  - Assessment: [Above market floor / At market floor / Below market floor]

### Comp Expectations Alignment
[From qa-master.md:]
- You previously communicated: $[X]-$[Y] range
- Market data shows: $[X]-$[Y] range
- Alignment: [your stated range is within / above / below market]
- [If misaligned: "You stated $X-$Y but market data supports $A-$B. Adjust your stated range or be prepared to justify the gap."]

## Negotiation Context

### Company Negotiation Profile
- **Flexibility:** [flexible / moderate / rigid]
- **Most movable components:** [typically: sign-on > equity > base at this company, or whatever the data shows]
- **Band limits:** [any known hard caps, e.g., "Google L5 base caps at ~$210K"]
- **Equity specifics:** [vesting schedule, cliff, refresh policy if known]

### Hiring Velocity
- **Current hiring pace:** [high / moderate / low] (from target-companies or web search)
- **Leverage implication:** [high velocity = more leverage; low velocity = less urgency on their side]

### Recent Context
- [Any relevant recent events: layoffs, hiring freezes, expansion, IPO plans, funding rounds]
- [How this context affects negotiation: e.g., "Post-layoff hiring means they're being selective but roles that are open have real budget"]

## Next Steps
1. Use this data to anchor your expectations before any comp conversation
2. If you receive an offer, run `/negotiate [offer details]` for full counter-offer strategy
3. If you need to state expectations in a screen, use: $[recommended range] with source: "[data source]"
4. If your career-plan targets are misaligned with market data, consider updating career-plan.md

Example

Input: /salary-research Anthropic Senior PM San Francisco

Output would include:

• levels.fyi data: limited data points (Anthropic is newer/smaller), $280K-$400K total comp range from available reports
• Glassdoor data: $180K-$220K base range from 15 salary reports
• Blind data: 3 recent data points showing offers in $320K-$380K TC range for senior PMs
• Company-intel: Anthropic offers competitive equity with 4-year vesting, known to be moderately flexible on comp, strong emphasis on mission alignment
• Level mapping: user's 6 years PM experience + AI project work maps to Senior PM (IC3/IC4 equivalent)
• User's dream offer ($350K) is at ~P50 for this role -- realistic target
• Recommended ask: $370K total to anchor high, with $320K walkaway
• Negotiation context: Anthropic is hiring aggressively (high velocity = good leverage), equity is pre-IPO (apply 30% discount vs public company RSU comparisons)

Quality Checks

1. Multiple sources required. Never rely on a single data source. Always search levels.fyi, Glassdoor, and Blind at minimum. If data is sparse on one source, note it and weigh the others more heavily.
2. Data confidence ratings. Every data source must include a confidence assessment based on: number of data points, recency, and known biases. Do not present low-confidence data as if it were definitive.
3. Pre-IPO equity disclaimer. For any pre-IPO company, always include the equity risk note. Never present pre-IPO equity at face value without a discount discussion. This is a common candidate mistake and the skill must prevent it.
4. Level mapping honesty. If the user's experience does not clearly map to the target level, say so. Do not inflate the level mapping to make the user feel good. A honest "your experience maps to L4, not L5" prevents embarrassment in negotiation.
5. Career-plan alignment. Always compare market data against the user's targets from career-plan.md. If the user's dream offer is unrealistic for this company/level, flag it respectfully: "Your target of $X is above P90 for this role. It's achievable only with exceptional leverage (competing offers at $X+, rare expertise)."
6. Actionable ranges. The recommended ask must be a specific number with a specific justification, not a vague range. "Ask for $350K, justified by P60 positioning on levels.fyi and your 8 years of relevant experience" is actionable. "Ask for somewhere between $300K and $400K" is not.
7. Freshness. Compensation data ages quickly. Flag if the most recent data is more than 6 months old. Note any market shifts (layoffs, hiring booms) that may have moved the market since the data was collected.
8. No fabricated data. If web research yields no results for a company/level combination, say so. "No levels.fyi data available for [Company] [Level]. Relying on Glassdoor and comparable companies." Never invent data points to fill gaps.
9. Seamless handoff to /negotiate. The output should contain everything /negotiate needs. After running /salary-research, the user should be able to run /negotiate and the system uses the research automatically.
10. Pay-cut scenario detection. If the user's current comp exceeds the P50 for the target role, this MUST be flagged. Many experienced PMs moving between companies (especially from high-comp companies like Google/Meta to pre-IPO companies, or from inflated levels at one company to correctly-leveled roles at another) face pay cuts. The skill must surface this honestly so the user enters negotiations with eyes open, not surprised when the offer comes in "low."
11. Cross-company level translation. Always provide an explicit level mapping table when the user's current company uses a different leveling system than the target. "Amazon L6 = Google L5" is the kind of insight that prevents candidates from anchoring to the wrong band.
12. Cross-market relocation section required. If the user's current comp is in a non-US currency and the target role is in the US, the output MUST include the "Market Transition" section showing: current comp in local currency, USD equivalent, US market rate for the role, the gap as a percentage, and the framing note that this gap reflects a market differential not a personal raise. Without this section, the user is left without the context they need to frame comp discussions correctly.
13. Comp question danger flag for cross-market candidates. If the user's current comp is significantly below the US target range (e.g., EUR 130K vs $280-340K target), the output must include a bolded warning in the Negotiation Context section: "CRITICAL: Your current comp in [currency] will sound dramatically low to a US recruiter. NEVER state it. Always frame as: 'Based on levels.fyi, Senior PMs at [Company] earn $[X]-$[Y]. I'm targeting within that range.' Rehearse this redirect until it is automatic."

Career gap / returner comp adjustments: If career-plan.md shows a career gap (e.g., gap_years > 0, or explicit mention of career break/return), add these adjustments to the salary research output:

• Gap-adjusted benchmarks: Note that market rates may have shifted during the gap. Pull CURRENT rates from levels.fyi, Glassdoor, and Blind -- not rates from when the user was last employed. If the user's last comp data is from 3+ years ago, flag: "Your last active comp was in [year]. Market rates for [level] at [company type] have [increased/decreased/shifted] since then. Current data shows $[X]-$[Y]. Do not anchor to your pre-gap salary -- the market may have moved significantly."

Military Comp Structure Translation: If career-plan.md shows military background (active duty, veteran, reserves, or military titles like Captain, Major, Lieutenant, Colonel, etc.), add these adjustments to the salary research output:

• Military comp includes more than base pay. Military compensation includes: base pay + BAH (Basic Allowance for Housing) + BAS (Basic Allowance for Subsistence) + TRICARE (healthcare, equivalent to $500-$800/mo employer-paid premium) + TSP matching (retirement, equivalent to 5% employer 401k match). When converting to civilian total comp, calculate the TRUE total military comp including all benefits. An O-3 (Captain) with ~$65K base pay has a true total comp of approximately $90-110K+ when all allowances and benefits are included.
• Common mistake: military candidates anchor to their BASE pay and undersell themselves. A Captain earning $65K base pay who targets $90K civilian salary is leaving $30-50K+ on the table. Flag this: "Your military base pay of $[X] does NOT reflect your total compensation. Your TRUE total military comp including BAH, BAS, TRICARE, and TSP is approximately $[calculated total]. For civilian roles, target market rate based on your TOTAL years of leadership and analytical experience, not your military base pay."
• Defense tech clearance premium: Defense tech roles often pay a premium for clearance holders because the clearance itself has market value ($50K-$150K to sponsor, 6-12 month timeline). Research this differential when the target role is at a defense tech company or any company doing classified work. Note: "Your active [clearance level] clearance has significant market value. Defense tech companies factor this into their comp -- expect a 10-20% premium over non-cleared equivalent roles."
• Pre-gap salary anchoring risk: Flag if the user's comp expectations (from career-plan.md) are anchored to their pre-gap salary rather than current market rates. If the pre-gap salary is significantly below current market (market moved up during the gap), note: "Good news: market rates have increased since your last role. Your pre-gap comp of $[X] would map to approximately $[Y] in today's market. Target current market rates, not your historical number." If the pre-gap salary is above current market (rare, but possible in a down market), note: "Market rates for [level] have [compressed/shifted] since [year]. Current P50 is $[X] vs your pre-gap $[Y]. Adjust expectations to current market data."
• Return-to-work program comp data: If available, note that returnship programs (Path Forward, iRelaunch, company-specific) often pay 80-90% of market rate during the program period, with full-market conversion rates upon conversion. Include this context: "Returnship programs typically pay $[X]-$[Y] for [level] during the program (80-90% of market). Conversion offers are usually at full market rate. Negotiate the conversion comp at program entry, not at conversion time."

ZERO-EQUITY EMPLOYER DETECTION: If career-plan.md indicates the candidate's current or most recent employer has NO equity compensation (private company with no equity plan like Epic Systems, government, academia, non-profit, or bootstrapped startup with no equity grants), apply these adjustments:

• Flag in the output: "ZERO-EQUITY EMPLOYER: Current comp is 100% cash. Do NOT let this anchor the negotiation low -- total comp at equity-granting companies will be significantly higher."
• Generate a comp comparison table showing: Current cash-only comp (base + bonus) vs. Target total comp at equity-granting companies (base + equity + bonus). This makes the gap visible: e.g., "Current: $180K all-cash at Epic | Target: $280-340K total comp (base $190K + equity $70K/yr + bonus $30K) at comparable-level roles."
• Note: Candidates from zero-equity employers often have strong cash savings habits -- this can be used as a negotiation asset. Request MORE equity and LESS sign-on (there is no "forfeited equity" to buy out, so sign-on has less justification). Frame: "I have no unvested equity to replace, so I'd prefer to weight the package toward equity to align my incentives with the company's long-term success."
• Specific example: Epic Systems (private, no equity, Midwest-anchored comp) -- an Epic PM earning $170K all-cash is at market for Epic but well below P25 total comp for equivalent PM roles at equity-granting tech companies. The transition to a company with equity means a significant total comp increase even if base salary stays flat.
• Trigger: career-plan.md mentions employer with no equity, explicitly states no equity in comp, or lists a known zero-equity employer (Epic, government agencies, non-profits, academia).

HEALTH-TECH VS GENERAL-TECH COMP DIFFERENTIAL: Health-tech PM roles typically pay 10-20% below general tech PM roles at the same level due to smaller equity grants, narrower comp bands, and regulated-industry constraints. Apply these adjustments:

• Segment research by industry when the candidate has health-tech domain experience: "Health-tech target: $250-280K | General tech target: $280-320K at the same level."
• Flag when a health-tech domain expert is also applying to general tech: "Your health-tech comp range may anchor you below the general-tech market. Research both ranges separately. Do not let a health-tech offer set your expectations for general tech negotiations."
• This differential applies to companies like Epic, Cerner/Oracle Health, Veeva, athenahealth, and health-tech startups vs. general tech companies at equivalent levels.

Founder-to-employee comp reset (failed founder / shut-down company): If career-plan.md shows founder/CEO/co-founder experience at a company that shut down, failed, or was acqui-hired at low value, add a "Founder Comp Context" section to the salary research output:

• Do NOT anchor to the founder's previous salary. Founders at failed startups often paid themselves well below market (sometimes $0-$80K) or well above market (inflated CEO title with inflated comp at a funded startup). Neither number is relevant to the market rate for the target PM role. The anchor must be the ROLE's market rate from levels.fyi, Glassdoor, and Blind -- not the founder's historical comp.
• Flag the anchoring risk explicitly: "Your founder comp of $[X] at [startup] does not reflect the market rate for [target level] PM roles. Founders' self-set salaries are not comparable to market comp. Anchor your expectations to the role's market data: $[Y]-$[Z] based on [data sources]."
• If the founder's previous comp was BELOW market (common for bootstrapped/early-stage founders): Note: "Your previous comp was below the market rate for this role. This is an INCREASE scenario. Do not feel guilty about targeting market rate -- you were underpaying yourself as a founder. The market rate for [level] at [company] is $[X]-$[Y]. Target P50-P75."
• If the founder's previous comp was ABOVE market (common for well-funded startups with inflated titles): Note: "Your founder comp of $[X] included equity and/or a CEO-level salary at a funded startup. The market rate for [target level] PM is $[Y]-$[Z], which may be lower. This is a comp reset, not a pay cut -- you are moving from a founder role to an employee role at a different level. Adjust expectations to the employee market rate."
• Equity sensitivity: Founders are often equity-focused. Add context: "As a former founder, you may be tempted to over-weight equity in the offer evaluation. For PUBLIC companies, equity is comp -- value it fully. For PRE-IPO companies, apply a founder's skepticism: you know from experience that equity can go to zero. Apply a 40-60% discount to pre-IPO equity when comparing to your cash needs."