How to Counter a Job Offer (Without Losing It)

You got the offer. Your instinct is to accept before they change their minds.

That instinct is costing you money. Somewhere between $10,000 and $50,000, depending on how hard you flinch. Most engineers spend months grinding LeetCode to get to the offer, then fold the moment a recruiter says "we're very excited to extend this opportunity." Companies pre-approve a buffer before extending the initial offer. The first number is not a final number. It's an opening bid, and recruiters expect a counter. A 2024 Fidelity survey found that 85% of people who tried to negotiate got some improvement. Offer rescissions from polite, data-backed counters are vanishingly rare. They happen when candidates are rude, ask for something absurd, or come back after already accepting.

After a year of applying, grinding, and interviewing, the one thing you cannot afford to do is accept the first number.

Your Leverage Window Is Shorter Than You Think

The verbal call is not the moment to counter. When a recruiter says "we'd like to extend an offer," the right response is something like: "That's great news. When can I expect the written offer?" Then get off the phone.

The voice in your head saying "just be grateful" is not wisdom. It's anxiety pretending to be professionalism. Every hour after you receive the written offer that you spend not countering is leverage quietly evaporating.

Your negotiating power is maximum from the moment you receive the written offer to the moment you sign. Before the written offer, you haven't seen the full picture: equity grant, vesting schedule, sign-on, exact base, bonus structure. You cannot counter what you haven't read.

Once you have the written offer, reply: "Thank you. I'll review the details and get back to you by [specific date, 24 to 48 hours out]." This is standard. No company pulls an offer because a candidate asked for two days to review it.

Do the Research Before You Write a Word

You need two numbers before you write anything: a target and a floor.

Your target is the specific dollar figure you're aiming for. Your floor is where you walk. If you don't know your floor before you counter, you don't have a negotiation. You have a conversation that will end with whatever number they say next.

Levels.fyi is the most accurate source for total compensation in tech, broken down by company, level, and location. It's like Glassdoor but the numbers are real. Glassdoor covers broader industry ranges. The professional network Blind has recent offer data from people at the exact companies you care about.

Find the 50th percentile for your role, level, and location. That's your baseline. Your target should be the 75th percentile. Counter at or above your target, because you can come down but you cannot go up once you have named a number.

Concrete example: if Levels.fyi shows senior engineers at this company averaging $195K base in New York and the offer is $175K, counter at $205K. Not $185K. Anchoring low is how people leave money on the table by their own hand.

Email Beats the Phone (for You, Not Them)

Recruiters negotiate every week. You negotiate every few years. On a call, they are better at this than you are. They know how to create urgency, steer the conversation, and get verbal commitments before you're ready to give them. You're in a knife fight and they brought the knife. Email is how you bring a calculator instead.

Email gives you control over every word, removes real-time pressure, and creates a written record of everything agreed. If a recruiter pushes to discuss by phone, redirecting is fine: "I think more clearly on complex decisions over email. Would that work?" If they insist, agree, then confirm every detail in writing afterward.

What a Counter Email Actually Looks Like

Four parts: genuine enthusiasm, specific number, one real reason, open door.

No "hopefully." No apology. No "I was wondering if maybe." A specific dollar figure, not a range.

Hi [Name],

Thank you for the written offer. I'm genuinely excited about this role
and [specific thing: the problem space, the team, what you're building].

After reviewing the details and researching compensation for
[role/level] in [location], I'd like to ask for a base of $[X].
This reflects the 75th percentile for this level at comparable
companies, and I want the base to reflect the scope of the work.

Looking forward to finding something that works for both of us.

[Your name]

The reason doesn't have to be elaborate. It just has to be real. "Based on my research" works. "I'm weighing a competing offer in the same range" works. "I'm leaving unvested equity at my current company" works. A bare number feels like a demand. A reason makes it a conversation.

Ask in the Right Order

Not everything is equally movable, and the easiest win is usually not what candidates ask for first.

Sign-on bonuses often come from a separate budget entirely from base salary. This is non-obvious, and it changes how you should think about hitting a wall. Base salary increases require changing a pay band, which finance guards carefully. Sign-on bonuses often sit in a recruiting budget with a different approval chain. The same manager who cannot add $5K to base might have zero difficulty approving a $25K sign-on. Different drawer, same filing cabinet.

Patrick Star questioning whether companies charge extra normally or whether the 50% is real

The sign-on bonus really is from a different budget. It's actually extra. Worth asking for separately.

Pivoting to sign-on when base stalls is not a consolation prize. It's a different pool of money with a different gatekeeper.

Equity grants are the second lever. RSU increases are easier to get than base increases at public companies because the company's argument writes itself: if you believe in us, take more equity. "I'm happy to take less base if we can increase the initial grant" is a phrase that moves things.

Base salary is hardest. It sets a precedent, feeds into bonus calculations, and lives inside a band. Push on it first if that's where you have the most room. If it doesn't move after one counter, pivot to sign-on.

Other items that are nearly always negotiable: start date, early performance review (six months instead of twelve), remote flexibility, professional development budget. These don't move the total comp number dramatically, but they cost the company little and are almost never declined outright.

Use Competing Offers. Don't Invent Them.

A real competing offer is the most reliable source of pressure in a negotiation. More reliable than market data, more reliable than your interview performance.

Use it directly and honestly: "I have an offer from [Company] at [$X total comp]. I prefer this role for [genuine reason], and I want to see if we can get closer to that number before I decide."

Two rules. Only say it if it's true. And only use it if the competing offer is real ammunition: a comparable company at a comparable level. A Series A startup at half the total comp does not move a FAANG recruiter. A peer-tier offer at $220K moves a FAANG offer sitting at $195K.

Do not invent competing offers. Recruiters verify. They ask for redacted letters. Getting caught ends the negotiation and sometimes the relationship.

The best time to create this situation is before any single offer lands. Coordinate interview timelines so windows overlap. Once you have one written offer, email every other company immediately: "I've received a competing offer and my timeline has compressed. Can you accelerate?" Most will. For a breakdown of how specific companies structure their processes and timelines, the guides on Google's full interview loop and Amazon's process are worth reading before you schedule anything.

When They Push Back

The commitment trap. "I need you to commit before I can go to bat for you." This is asking you to surrender your leverage in exchange for a promise that more leverage is coming. It isn't. The answer: "Once you come back with the updated offer, I'll give you a quick decision. You'll hear from me fast."

The exploding offer. "This offer expires Friday." Push back directly: "That timeline doesn't work for me. I need until [five to seven days out]. Is that possible?" Most exploding offers extend when asked. Research from INSEAD found they reduce employer-candidate match quality by 8 to 13 percent, which is why many companies quietly extended deadlines all along.

"This is our best offer." Sometimes true. Test it with a sign-on pivot: "I understand base is at the band ceiling. Is there any flexibility on a sign-on to close the gap to my target total comp?" You find out quickly whether the wall is real or decorative.

The One Thing That Actually Costs You the Offer

Not countering with a higher number. Not asking for a sign-on. Not requesting time to decide.

What gets offers revoked is making it adversarial. Ultimatums, rude emails, or coming back to renegotiate after you've already said yes. If you've told a recruiter "I accept," the negotiation is over. Coming back after that isn't hard-nosed. It's breaking an agreement.

Counter once. If they improve the offer, you can counter once more if a genuine gap remains. After two rounds you've reached the end. Accept or walk. A third counter starts to feel like bad faith, and recruiters talk to each other.

You got to the offer by performing well through the technical rounds. SpaceComplexity is built for that part: voice-based mock interviews with rubric-graded feedback across problem-solving, communication, and code quality, so you arrive at the offer conversation from a position of genuine strength. The counter is the last mile. Don't leave it untouched.

The Short Version

Don't react to the verbal call. Wait for the written offer before you counter.
Research your target (75th percentile) and your floor before you write anything. Use Levels.fyi.
Counter by email. Recruiters are better negotiators on the phone than you are.
Counter format: enthusiasm, specific number, one real reason, open door.
Ask in order: sign-on bonus (easiest, separate budget), equity (moderate), base salary (hardest).
Competing offers are your best pressure point. Only use them if they're real.
Counter once, maybe twice. A third round looks like bad faith.
The one thing that actually costs you the offer: making it adversarial.