A hiring manager sends over a GTM engineer req with a $220K OTE on it — $110K base, $110K variable, quota-carried, same plan as the account executives sitting two desks over. Every candidate worth interviewing will read that as a $110K job with a lottery ticket attached, and most will never reply. The posted market for this role is priced like engineering, not like sales. If you budget it like sales, you are quoting a number the market does not recognise and you will lose six weeks finding that out the slow way.
Here is how base/variable actually splits for a technical revenue role in 2026, what the posted offers say, and how GTM engineer packages compare to RevOps and AE packages at the same level.
The number on the req is base — not OTE
Start with what is actually on job posts. Posted US GTM engineer roles at venture-backed companies advertise base-only ranges centred on $151,200–$220,000 — that range is base salary, with no variable component quoted at all. The practical consequence: a candidate reading your post is anchoring on the guaranteed number. If your $220K is an OTE, you are competing against companies offering $220K before anyone hits a target.
Where variable pay does show up for GTM engineers, it sits on top of base rather than carving base in half. One benchmark set shows a base range of $96,000 to $172,000 with additional pay of $48,000 to $90,000 layered over it. That is a bonus, structurally — not a commission plan. And at AI companies paying the top of the market, the variable slice is 25%–30% of an OTE averaging $230,000–$310,000. Roughly 75/25. Never 50/50.
If the guaranteed half of your offer is smaller than someone else’s whole offer, the pay mix is the problem — not the pitch.
Three pay structures, one job family
GTM engineering sits between three established comp models, and the data makes the gap obvious. Here is the same seniority tier priced three different ways:
| Role | Structure | Base | Variable / total |
|---|---|---|---|
| GTM engineer (venture-backed, posted) | Flat base, no variable quoted | $151,200–$220,000 | Not posted |
| GTM engineer (AI co., top of market) | ~75/25 base-heavy | — | 25%–30% variable on $230K–$310K OTE |
| RevOps | Salary plus discretionary bonus | — | 76.92% receive a yearly bonus |
| SaaS AE (median) | ~50/50 quota-carried | — | 53:47 base:variable on $190K OTE |
| Enterprise AE (self-reported) | 50/50 quota-carried | $130,000 | $260,000 OTE |
| Sales engineer (BLS, all industries) | Base-weighted | $124,900 median | — |
| Sales engineer (software publishers) | Base-weighted | $153,410 median | — |
Read the AE rows against the GTM engineer rows. An enterprise AE at $260,000 OTE has exactly half of it guaranteed — $130,000 base. A GTM engineer at the posted median has $151,200–$220,000 guaranteed and no quota. Those are not variations on a theme. They are different jobs with different risk profiles, and the market has already decided which one GTM engineering is.
The RevOps number is the one most hiring managers should pay attention to, because it is where GTM engineer plans are drifting: salaried, with a bonus that 76.92% of people get. Not everyone — but most. That is a comp plan designed to reward contribution to a number without pretending the individual controls it. Which is precisely the honest description of what a GTM engineer does.
Why 90/10 and 80/20 beat 50/50 for this role
Variable pay works when the person being paid can move the number on their own. An AE can. They own the account, run the cycle, and close or don’t. Half their package at risk is defensible because the lever is in their hand.
A GTM engineer builds the lever. They wire the enrichment pipeline, the routing logic, the agent that drafts the first-touch sequence, the data model the forecast runs on. Pipeline moves because of that work — but it moves two quarters later, across a team of fifteen reps, through a funnel the engineer doesn’t sit in. Attributing 50% of an individual’s pay to that chain of causation produces one of two outcomes: the plan pays out regardless, in which case it was base with extra paperwork, or it doesn’t, in which case your best technical hire leaves in month nine.
That is why the plans that work cluster at:
- Flat base, no variable. The default on posted venture-backed reqs. Simplest to approve, easiest to compare, zero ambiguity for the candidate.
- 90/10 with a company-performance kicker. The bonus tracks company or GTM-wide attainment, not an individual quota. Cheap to administer, reads as upside rather than risk.
- 80/20 or 75/25 with MBOs. Appropriate at the senior end and at AI companies paying the top of the band, where the 25%–30% variable is tied to delivery milestones and system outcomes rather than closed-won.
Anything past 70/30 and you are not paying a GTM engineer — you are asking a builder to accept an AE’s risk without an AE’s control, and the top of the candidate pool simply declines.
What actually happens when you post an AE-style OTE
Three failure modes, all of them expensive.
You filter for the wrong candidate. People who are comfortable with half their pay at risk are, mostly, salespeople. People who build systems have spent their careers in base-heavy comp and read a 50/50 split as instability. You will get volume — just not from the pool you wanted. That is the spray-and-pray version of a search, dressed up as a comp plan.
Your offer loses a side-by-side comparison it should win. A candidate holding a competing offer at the posted base median of $151,200–$220,000 will compare your guaranteed half against their whole. A $220K OTE at 50/50 is a $110K offer in the only column that matters, and $110K is below even the all-industry sales engineer median of $124,900 — let alone the $153,410 median inside software publishers, which is the sector your req is actually competing in.
You rebuild the plan in month four. When the first bonus cycle lands and nobody can agree what the engineer’s quota attainment was, finance and the hiring manager renegotiate — usually by converting most of the variable to base. You have now paid the cost of a comp redesign and a trust deficit to arrive at the structure the market would have given you on day one.
How to budget the req instead
Build base first, then decide whether a bonus adds anything. A workable 2026 frame:
| Level | Base | Variable treatment | All-in target |
|---|---|---|---|
| Mid — first systems hire | $150K–$175K | Flat base or 10% company kicker | ~$150K–$190K |
| Senior — owns the GTM stack | $175K–$220K | 10–20% MBO on delivery milestones | ~$190K–$265K |
| Staff / lead — AI-native, top of market | $200K–$230K+ | 25%–30% variable | $230K–$310K OTE |
The senior row is where most reqs land, and it maps cleanly onto the posted base band of $151,200–$220,000 with an additional-pay layer of $48,000–$90,000 on top. Note what that layer is not: it is not half the package, and it is not quota. Price the base to win on its own, then treat variable as the thing that keeps a good hire for a third year.
If you are setting bands across the wider GTM org at the same time, the salary benchmarks and the GTM engineer hiring guide cover level definitions and scope boundaries alongside the numbers.
Writing MBOs a technical hire will accept
If you are running 80/20 or 75/25, the MBO has to be something the engineer controls. Four that hold up in practice:
- Shipped systems. Lead routing live and under a defined SLA by end of Q2. Binary, dated, verifiable.
- Data quality. Account match rate or enrichment coverage above a stated threshold, measured from the warehouse — which is why these reqs often overlap with data warehouse and RevOps scope.
- Rep-facing adoption. Percentage of the sales team using the workflow weekly. Pays for usefulness, not output volume.
- Company attainment kicker. Same multiplier the rest of GTM gets. Aligns incentives without pretending the engineer owns a territory.
What to avoid: individual pipeline-sourced quota, closed-won attribution, and anything requiring a three-way argument about multi-touch models at bonus time. If the plan needs an attribution debate to pay out, it will not pay out — and the hire will read that correctly.
Three checks before the req goes live
- Is the posted number base or OTE? Say so explicitly in the post. Ambiguity costs you replies from exactly the candidates who scrutinise offers hardest.
- Does your guaranteed figure clear the technical floor? If base alone doesn’t beat the $153,410 software-publisher sales engineer median, you are below market for the technical skill before you have priced the GTM layer at all.
- Have you compared against the right neighbour? Benchmark your plan against the RevOps model — salaried with a bonus 76.92% of people receive — not against AE comp at 53:47 on a $190K OTE. Same revenue org, entirely different risk contract.
Get the mix right and the search gets faster, because the offer stops needing a defence. Approved base, a bonus framed as upside, and a number candidates can compare in one glance — that is how these roles close in days, not weeks, and how you keep top-tier technical talent past the first bonus cycle.
