Product
Salary Counter Composer
You have a written offer and want more. Paste the base, your target, the role, and 2-3 leverage points. Get back a four-part counter email (with a softer alternate), a 3-opener follow-up call script for the recruiter, and a prioritized lever ladder tuned to the employer type. Uses only the market comparables YOU supply. Never invents numbers, never fabricates market data, always ends on interest not ultimatum.
About this tool
What does salary counter composer do?
You have a written offer and want more. Paste the base, your target, the role, and 2-3 leverage points. Get back a four-part counter email (with a softer alternate), a 3-opener follow-up call script for the recruiter, and a prioritized lever ladder tuned to the employer type. Uses only the market comparables YOU supply. Never invents numbers, never fabricates market data, always ends on interest not ultimatum.
How does Salary Counter Composer work?
Salary Counter Composer is an AI counter offer email tool. You paste the base offer, your target number, the role, the employer type (startup, big_tech, non_tech, non_profit), and 2 to 3 supporting facts (a competing offer, unique skills, deferred comp). One structured pass returns four artifacts you can send today: a four-part counter email (opener, gratitude, ask, close), a softer alternate version of the same email, a follow-up call script with 3 recruiter openers, and a prioritized lever ladder (sign-on, RSU refresh, PTO, earlier review) tuned to the employer type you picked. Every number in the output traces back to a number you typed.
How is this different from asking ChatGPT to write my counter offer?
A freeform ChatGPT prompt does three things that hurt you in a negotiation. First, it confidently cites made-up salary bands ('the market for this role is 180 to 220') without knowing your city, level, or company class. Second, it does not check whether your ask is defensible; it happily writes a 40% counter that the recruiter will silently discount. Third, it defaults straight into the exact recruiter cliches they clock in the first sentence ('commensurate with my experience', 'I am thrilled and grateful for the opportunity', 'as we discussed'). Salary Counter Composer refuses to invent market data, computes a calibration verdict against the Kalzumeus 10 to 20% band, and strips the classic recruiter cliches from every draft it returns.
Does it invent salary market data or comparables I did not give you?
No. Every number in the output is traceable to a field you filled in: base_offer_amount, target_amount, and (if you supplied one) competing_offer_amount. The market frame sentence in the email is built strictly from the supporting facts you pasted, not from a training-data guess. There are no made-up percentiles, no fabricated 'the market pays X for this role at your level', and no invented 'similar candidates get Y'. If you did not give it a comparable, it does not pretend to have one, and the email argues from your supplied facts instead.
What is the ask calibration verdict?
The tool computes your ask as a literal delta_percent from the base offer to your target, then classifies it against two anchors from real negotiation data: the Kalzumeus 10 to 20% band (patio11's widely cited defensible counter range) and the Metaintro study's 12.45% average successful counter. Verdicts come back as conservative, reasonable, aggressive, or out_of_band. If your ask lands out_of_band (say, a 35% jump above the base), the tool suggests a defensible fallback number inside the reasonable band, plus one line on why that number is more likely to land without inviting a rescinded offer.
What is in the follow-up call script?
Three tone-calibrated openers for the exact pushbacks a recruiter will actually use on the call. The 'outside our range' opener responds without capitulating and pivots to a non-base ask. The 'budget is fixed' opener acknowledges the constraint and asks what IS open (sign-on, RSU refresh, earlier review). The 'revisit at first review' opener accepts the frame in principle but pins down a specific date, a specific number, and a written commitment. Each opener is 2 to 3 sentences in speaking cadence, ready to read straight off a script during the actual call.
What is the lever ladder?
A prioritized list of 5 to 7 non-base asks you can push on when the base salary is capped: sign-on bonus, RSU refresh, earlier performance review, additional PTO, delayed start date, remote or hybrid flex, relocation assistance. Each item comes with one line on why it works (naming WHO can approve it without a fresh HR sign-off) and a verbatim 1-liner sample_ask you can paste. The employer_type field picks the top opener automatically: startup leads with equity, big_tech leads with RSU refresh, non_tech leads with sign-on plus PTO, non_profit leads with PTO plus start date plus remote flex.
Which tier should I pick, Quick or Deep?
Pick Quick (15 produde tokens) when you want a defensible starting draft you will edit yourself before sending, or when the ask is small and the relationship is easy. Pick Deep (40 produde tokens) when the number actually matters: Deep runs a two-pass polish that internally rewrites for cadence, cross-checks your target against the competing_offer_amount you supplied, and calls out every cliche in the forbidden list by name so the email reads like you wrote it. The Metaintro study puts the average successful counter around 12.45%, a meaningful full-year raise on any real offer, so the higher tier pays for itself on a single accepted counter.
Is it free, and do I need a subscription?
No subscription. Each run costs produde tokens: 15 for Quick, 40 for Deep. Tokens are a prepaid platform credit you buy once and spend across any produde tool, so you pay per counter draft instead of a monthly negotiation-coach retainer. If a run fails, the full token cost is refunded to your wallet automatically. Most people write one or two counter emails per job search, so a small top-up covers the whole search plus the next few years of raise conversations.
What does it cost?
From 15 tokens. Pick a tier on the form above. Produde tokens are a prepaid platform credit; buy them once and spend on any tool.
Do I pay if a run doesn't complete?
No. You only pay for results. If a run can't finish, its full token cost stays in your wallet automatically.
Is my data stored?
Your input and the run output are saved to your account so you can browse history. They are not sold, shared with advertisers, or used to train shared models. See privacy for details.