The math behind your job offer: why base salary is the worst number to compare

When you have two job offers in hand, the conversation always goes straight to base salary.

"Offer A is $180k. Offer B is $175k."

Offer B is actually worth $200k more over four years. You'd never know from the number everyone leads with.

The problem with comparing bases

Base salary is clean and comparable. Everything else is complicated.

Equity is the most obvious trap. 100k RSUs sounds like a lot until you understand:

The vesting schedule (4 years with a 1-year cliff means you walk away with $0 if you leave at month 11)
The strike price and current fair market value (for options)
Whether the company is public or private (and what "liquidity event" means in practice)
The valuation risk on private equity

Signing bonus is one-time, year 1 only. 401k match depends on vesting schedule and caps. Benefits (health, dental) are real money but hard to compare without actual quotes.

Nobody does this math before accepting. Everyone does it retrospectively, usually after it's too late.

What the new EverCV feature does

The new offer comparison tool (POST /api/offer-comparison) takes two offers and calculates 4-year total compensation in Python code — not an AI estimate. The math is:

4-year total = (base × 4)
             + equity_total (full 4-year vest)
             + signing (year 1 only)
             + (match_pct/100 × base, capped at match_cap, × 4)

That's it. Clean, transparent, inspectable. The test suite validates the math independently of the AI narrative. If Claude writes the wrong recommendation, the numbers are still correct.

Claude (Haiku) writes two things:

A plain-English recommendation: who wins, by how much, and whether the difference is meaningful
A list of intangibles to investigate — the things the math can't capture

What the intangibles look like

The AI prompt is specifically instructed to avoid repeating what's in the numbers and focus on what's hidden. The output looks like this:

"Does the equity in Offer B have a 1-year cliff? If you leave at month 11, you walk away with $0 from 25% of the 4-year package."

"Offer A mentions 'remote-friendly' — is that in the offer letter or a verbal promise from the hiring manager?"

"The 4% 401k match in Offer B assumes the company stays in business to pay it. At a pre-Series A startup, that's a real risk."

These are the questions that should be asked before signing, and most people don't ask them because nobody prompts them to.

The actual math on a real-ish comparison

Two common offers for a senior engineer right now:

| | Offer A (Stripe) | Offer B (Plaid) | |---|---|---| | Base | $200k | $185k | | Equity (4yr) | $400k RSUs | $500k RSUs | | Signing | $20k | $10k | | 401k match | 4% uncapped | 6% up to $10k |

Naive comparison: Offer A pays more ($200k vs. $185k base).

4-year total:

Offer A: $800k + $400k + $20k + $32k = $1.252M
Offer B: $740k + $500k + $10k + $40k = $1.290M

Offer B wins by $38k over four years despite the lower base salary. The equity delta and better 401k match swamp the $15k base difference.

Why this is in EverCV

The salary negotiation tool and the offer comparison tool address the two moments where engineers leave the most money on the table: when you have one offer (negotiate) and when you have two (compare correctly).

The math is free and public — you can do it in a spreadsheet. The tool makes it fast, catches the 401k match and signing-bonus math that people forget, and prompts you to investigate the intangibles that the numbers hide.

EverCV is building the infrastructure for the job search lifecycle — from application to offer acceptance. Try offer comparison at evercv.io.