How do I handle a top rep who's threatening to leave?

Don't panic-match the offer. The threat itself is a diagnostic signal that something broke 60-90 days ago. Run a structured 48-hour stay conversation, isolate the real driver (autonomy, comp band, manager trust, or territory), make ONE concrete change you can deliver in writing, and be psychologically prepared to let them walk.

Counter-offers without root-cause repair create a 6-month flight risk and a precedent that quota-busters get raises by interviewing.

TL;DR Decision Frame

Why The Threat Itself Is Information

Top reps don't update their LinkedIn at 11pm on a Tuesday for fun. By the time you hear "I have an offer," they've usually been emotionally checked out for 8-12 weeks. According to Gartner's 2025 Sales Talent Survey, 74% of B2B sellers who receive a competing offer have already had two or more "trigger events" in the prior quarter — a missed promotion, a territory cut, a 1:1 cancelled three weeks running, or a comp plan rewrite they didn't see coming.

The offer is the symptom; the trigger is the disease.

This is why panic-matching fails: you're treating the symptom. The Bridge Group's 2025 SaaS AE Report (https://www.bridgegroupinc.com/blog/sales-development-report) tracks the post-counter-offer half-life at 11 months — meaning half of counter-offered reps are gone within a year anyway, but now you've also reset the comp anchor for everyone who hears about it.

And they will hear about it. Sales floors leak base salaries within 30 days; OTE within 60; retention bonuses within a quarter. Treating the offer as confidential is theater.

Plan as if every other rep on the team will know the package by month-end, because three of them already do.

The 48-Hour Stay Conversation (Do This Exactly)

Skip the "what would it take?" opener. That hands them the pen. Instead, run this script:

1. "What changed for you in the last 90 days?" — forces them off the offer and onto the trigger. Listen for org changes, missed promises, manager friction.
2. "If the offer didn't exist, would we be having this conversation?" — separates pull (the new role) from push (your problems). Push is fixable. Pull rarely is.
3. "Walk me through the new role — comp, territory, manager, ramp, equity vesting cliff, SPIFF program, peer pod size, who you'd report to, how their tech stack stacks up." — 60% of competing offers have hidden downsides (worse territory, longer ramp, equity that vests in 4 years instead of 1, no SPIFF program, smaller AE pod with no peer learning, manager you've never met, broken CRM hygiene, no SDR support).
4. "What would I need to put in writing in the next two weeks for you to commit through next fiscal?" — forces specificity and a time horizon you can plan against.

Take notes. Do not counter in the room. Tell them you'll come back in 48 hours with a written response.

Reps who refuse a 48-hour pause are already gone — let them go gracefully. Reps who use the 48 hours to escalate to "I need an answer by Friday" are negotiating, not deciding; that's also useful information. Reps who text you that night with "I've been thinking — what if we just talked about it again tomorrow?" are usually staying.

One operational note: invite their spouse/partner consideration explicitly. "Take 48 hours, talk to your partner, sleep on it." This isn't soft — it's structural. Career moves in dual-income households are rarely solo decisions, and pretending otherwise leads to last-minute reversals.

The Four Real Drivers (And What Actually Fixes Each)

Comp band mismatch. Pull Pavilion's 2026 Compensation Report (https://www.joinpavilion.com/compensation-report) and Bridge Group benchmarks. If your top rep's OTE is below the 65th percentile for their segment and tenure, you have a market problem — fix it for the whole tier, not just the threatener.

One-off raises leak within a quarter and trigger copycat threats. Specific check: are accelerators kicking in at 100% of quota or at 110%? A 10-point shift in accelerator threshold can mean $40K-$80K of variable for an enterprise AE at 130% attainment.

That's where most "underpaid" feelings actually live, not in base. Second check: how does the multiplier curve look between 130% and 200%? Many plans flatten there to control cost, which is exactly where your top rep operates and notices.

Third check: SPIFFs and contests — top reps don't just win them, they architect their pipeline around them. Cutting SPIFF budget without warning hits top reps harder than a flat base.

Manager trust collapse. No raise repairs this. Forrester's 2025 B2B Seller research found manager relationship is the #1 retention factor for top decile reps, ahead of comp. Options: peer-manager swap, skip-level sponsorship from a VP, or a 90-day "report to me" arrangement during a transition.

If you're the broken manager, own it directly: "I think I've been the wrong fit for you the last six months. Here's what I want to try." Reps almost never expect ownership from a manager — naming the rupture often resets the relationship faster than any structural change. Tactical move: stop running deal reviews as the next 1:1 and start running career architecture instead.

Ten of the next twelve 1:1s should not mention pipeline. Diagnostic question to yourself: when was the last time this rep volunteered information you didn't ask for? If the answer is "I can't remember," trust is already broken.

Role/scope ceiling. Promotion isn't always available, but scope is. "Starting Monday you own all $2M+ enterprise prospects in the East and you're the named lead on the Acme expansion" is concrete. "VP track in 18 months" is vapor — and reps know it.

Per LinkedIn's 2026 Workforce Report, 41% of voluntary AE departures cite "no path to bigger deals" as a top-3 reason. Concrete scope tools that actually move the needle: named-account ownership, deal-size floor increases, strategic-account committee seats, RFP lead, customer advisory board host, named partner-channel relationships, board-level reference calls, analyst briefings (Gartner/Forrester) on their accounts.

Pick the one your rep has been quietly resentful about not getting — usually visible in slack threads where they tag themselves into deals that aren't theirs.

Territory or comp plan damage. If you cut their territory or moved accounts in the last two quarters, just say so. "We took 14 logos out of your patch in Q2 and that was the wrong call. Here's what we're putting back." Reps respect named, repaired mistakes far more than spin.

Refuse the urge to defend the original decision; the rep already decided it was wrong, the only question is whether you can see it too. If the comp plan changed mid-year and they got the worse end of it, mid-year true-ups are not weakness — they are a signal you can actually reconstruct what you said and what you did.

The half-life of a comp plan rewrite without a true-up is one fiscal year before the best reps interview.

Counter-Offer Math (When Money IS The Issue)

If — and only if — diagnosis confirms a real market gap:

• Match to the 75th percentile of segment, not the offer. Matching the offer teaches reps that competitors set your comp.
• Tie ~30% of the increase to a 12-month retention bonus paid in two tranches (month 6, month 12). This is standard practice in late-stage SaaS per OpenView's 2025 SaaS Benchmarks.
• Document the new plan in writing within 5 business days. Verbal commitments evaporate at the next all-hands.
• Cap the team-wide cost: if matching this rep means you'd need to lift three peers within 60 days, just do all four now. Half-fixes cost more than full ones.
• Run the math on equity refresh, not just cash. A $40K refresh grant vesting over 3 years is often cheaper than a $25K base bump that compounds forever and resets benchmarks.
• Get a counter-letter, not a verbal commit. The rep should send YOU a written commitment ("I will remain at [company] through [date]") in exchange for the package. This isn't legally binding but the act of writing it changes the psychology.
• Loop in the CRO/CFO before you send anything. Surprising your finance team with a one-off comp adjustment is how you lose your own credibility — and the next save you need to make.

Bear Case: Why You Should Often Let Them Go

Here's the part most managers skip. Sometimes the right answer is "good luck, we'll miss you."

• The threat is the problem. A rep who weaponizes a competing offer once will do it every renewal cycle. You've trained them. Force Management's 2025 retention data shows reps who successfully leverage one counter-offer leverage a second within 22 months 58% of the time.
• You can't out-pay equity. If they're going to a Series B with meaningful options and you're public/late-stage, a $30K base bump doesn't compete with a 0.4% grant. Don't pretend otherwise. Run the napkin math with them: at a $400M outcome and 0.4% post-dilution, that's $1.6M pre-tax over 4 years. You can't match that with cash.
• Top reps mask team rot. Sometimes one hero rep is hiding the fact that the other seven are mediocre. Losing them forces the org reset you've been avoiding (see /knowledge/q175).
• Replacement cost is overstated. SaaStr's 2025 data pegs full AE replacement at 6-9 months of fully-loaded cost — real, but knowable. A bad retention deal is open-ended.
• Culture cost of caving. Within 30 days, two other reps will hear about the save. Within 60, one of them will test the same lever. You'll spend Q3 firefighting comp instead of selling.
• They may be right about you. The hardest case: maybe they should leave because your company isn't where their next move should happen. Top reps have better signal on market trajectory than most managers admit. If three of your last five top performers all left for AI-native vendors, the trigger is the company, not the rep.
• Boomerang optionality. Reps who leave well come back. Reps you panic-saved won't. Roughly 12% of senior AEs in late-stage SaaS are boomerangs per LinkedIn data; you only get that pool if you handle exits cleanly.
• Sunk-cost fallacy. "But we invested two years in their ramp" is exactly the wrong frame. The ramp is sunk. The only question is forward expected value of save vs. exit, and forward EV of a leverage-trained rep is structurally lower than a fresh hire who joins because they want to be here.

If you let them go: 60-day structured ramp-down, written knowledge transfer (top 20 accounts, open opps, champion contacts, mutual action plans, internal coach intros, partner relationships), backfill req opened the day they tell you, and a clean reference policy. Burning bridges with departed top reps is how you lose the next one — they talk.

The departing rep's LinkedIn post matters more for recruiting than any career page. Send them a handwritten thank-you note 30 days after they leave; no other vendor will, and you've just bought a future referral or boomerang.

The 30-Day Aftermath

Whether they stay or go, audit the trigger. Pull the last two quarters of 1:1 notes, comp statements, territory changes, and pipeline reviews. If you can't reconstruct what broke, your management cadence is the real problem — see /knowledge/q120 on accountability without micromanagement and /knowledge/q124 on coaching managers to coach.

If the trigger was a peer manager, see /knowledge/q125 for scale signals. If you're now down a rep mid-quarter, /knowledge/q180 covers the no-notice resignation playbook and /knowledge/q170 covers fast onboarding when you backfill. If the rep stays but you suspect they're coasting, /knowledge/q128 helps separate coaching from termination decisions, and /knowledge/q121 covers when to step into deals during the transition.

For the broader team-confidence reset after a high-profile near-departure, /knowledge/q175 is the playbook. If a PIP becomes the right call within 90 days of the save, /knowledge/q123 covers how to do it without burning the relationship.

Run a team pulse 30 days after the save or exit. Two questions, anonymous: "On a scale of 1-10, how fairly is comp set on this team?" and "On a scale of 1-10, how confident are you in your career trajectory here?" Track the deltas quarter over quarter. A successful save shows up as flat or rising scores 90 days later.

A panic-match shows up as falling scores within 60.

TAGS: retention, top-performer, compensation, team-stability, manager-trust, counter-offer, stay-conversation, sales-leadership, talent-strategy, equity-refresh, comp-design, sales-management