**Honest answer first: the published data on GTM engineer pay is thin, US-centric, and measures what companies advertise rather than what they actually pay. Two analyses are worth knowing. Bloomberry, analyzing 1,000 GTM engineering job postings, reported a median of $127,500 per year, calculated only from postings that published a salary range (Bloomberry). Steven Moody, who hand-checked 491 US full-time GTM engineer roles of which 235 included salary data, found a median advertised base of around $150,000** with data current through May 2026 (Steven Moody).
Anyone quoting you a single confident number for this role is either summarizing one of these two datasets or making it up.
In short
- Two credible analyses, two different medians: $127,500 and ~$150,000 base. Both are US job-posting data.
- Both measure advertised ranges, not offers, not total compensation.
- The spread matters more than the middle: Moody found an $84k gap inside the same job title.
- Seat beats skill list. Roles closer to software engineering advertised $218,750 median base; outbound and growth roles, $135,000.
- AI demand carried no pay premium in the data (r = +0.07).
- Outside the US, there is essentially no published benchmark. Negotiate on scope and proof instead.
The two datasets, side by side
| Bloomberry | Steven Moody | |
|---|---|---|
| What was analyzed | 1,000 GTM engineering job postings | 491 hand-checked US full-time roles; 235 with salary data; 433 postings for skills analysis |
| Period | Postings over the past year; growth measured Jan–Sep 2024 vs Jan–Sep 2025 | Current through May 2026 |
| Headline pay figure | Median $127,500 | Median base ~$150,000 |
| Top figures cited | Vercel $252,000; OpenAI $250,000; LILT AI $221,500 | Pure software-engineering seats: $218,750 median base |
| Method notes | LLM-assisted extraction from postings; 100 LinkedIn profiles reviewed separately | Agencies, staffing firms, non-US and duplicate roles removed; LLM-assisted classification, verified programmatically; 28 hiring-team interviews |
| Stated limits | Median based only on postings that published a range | Midpoints of advertised ranges, not offers or total comp; ~28% of salary data from pay-transparency states; pay not predictable from skills or archetype (cross-validated R² ≈ 0) |
Sources: Bloomberry, Steven Moody.
Why the two numbers disagree
They are not contradicting each other; they are measuring slightly different populations.
- Different time windows. One looks at 2024–2025 postings; the other is current through May 2026.
- Different filters. Moody removed non-US roles, agencies and staffing firms, and manually deduplicated. Broader samples pull in lower-paying and less specialized listings.
- Base vs. total. Moody reports advertised base; posting medians elsewhere may mix base and on-target earnings.
- Disclosure bias. Only some postings publish ranges at all, and roughly 28% of Moody's salary data comes from mandatory-disclosure states, over-representing those employers.
Practical consequence: use the numbers as a range of plausible US advertised bases, not as an expected value for you specifically.
The finding that should change your plan
The most useful number in either dataset is not a median. It is the spread.
Moody's analysis classifies each posting into a "seat": pure software engineering, embedded software engineering, data, RevOps, or outbound/growth. The pay difference between seats is large: roles closest to software engineering advertised a median base of $218,750, while outbound and growth-focused roles advertised $135,000. As the analysis puts it, an $84k gap sits inside one job title (Steven Moody).
Two more findings deserve attention because they contradict popular career advice:
- Roles mentioning Clay advertised about $17k less ($148K vs $165K). That is a correlation about which kinds of roles mention Clay, not a reason to avoid the tool; Clay remains one of the most requested platforms in the field. Read it as evidence that tool fluency alone is not what is being paid for.
- AI demand was unpriced. Pay was flat across every level of AI requirement in the postings (r = +0.07), even as Anthropic mentions rose to 37.8% of postings in Q2 2026 from about 3% before Claude Code.
Meanwhile the market is clearly short of people: Moody reports 30% of roles still open after three months, and Clay says about 100 GTM engineering listings go live every month (Clay). Demand is real; the pricing of specific skills inside that demand is noisy.
What this data cannot tell you
Be precise about the gaps, especially if you are about to make a career decision on them:
- Offers, not ads. Posted ranges are opening positions. Negotiated outcomes are not in either dataset.
- Total compensation. Bonus, commission, equity and benefits are largely invisible here.
- Non-US pay. Both are US-centric. If you are hiring or being hired in Latin America, Europe or Asia, these figures describe a different market.
- Contract and agency work. Explicitly removed from Moody's set, and priced very differently in practice.
- What "GTM engineer" means at that company. The same title covers a Salesforce administrator, a Python-writing data engineer and an outbound operator.
- Individual predictability. Moody's own cross-validation found pay essentially unpredictable from skills or archetype (R² ≈ 0). Averages describe the market; they do not describe your offer.
If you see a confident salary "calculator" for this role, ask for its sample size and collection method before trusting it. For a role this young, most of them are estimates layered on estimates.
If you are outside the US
There is no reliable published benchmark for GTM engineers in most countries, and a US median is not a number you can reasonably demand. What works instead:
- Anchor on scope, not geography. What systems will you own, what metric are you accountable for, how many people depend on your work.
- Ask for the band. "What is the range for this role, and what separates the bottom from the top of it?" is a normal question and a fast way to find out if a company even has bands.
- Price the alternative. What would the company pay an agency, or a US hire, to get the same outcome? That is your real comparison, and it is often the honest basis for a raise conversation.
- Understand how you will be paid. Contractor invoicing, currency and fees change your take-home more than a few thousand dollars of headline salary. See getting paid as an international contractor and remote AI jobs for LatAm.
The MitHub way to raise your own number
MitHub's capability ladder is a better predictor of what you can charge than any market median, because it describes what a company is actually buying:
| Rung | What you sell | How pay conversations usually go |
|---|---|---|
| Doer | Hours of manual execution | Compared against the cheapest person who can do the task |
| Director | Output produced by directing AI and systems | Compared against tool subscriptions |
| Designer | Systems other people rely on | Compared against the cost of the problem you removed |
| Owner | A business result you are accountable for | Compared against the revenue you influence |
The jump that changes pay is Designer to Owner: moving from "I built the workflow" to "I own the number the workflow moves." That is also the jump that requires evidence, which is why we insist on documented case studies. A candidate who can say "here is the leak I found, the system I built, and what happened to the metric over eight weeks" is negotiating on something no benchmark can price.
A practical twelve-month plan:
- Pick the seat you want. If you want the top of the band, the data says that seat is closer to data and engineering. Read do GTM engineers need to code? before deciding.
- Close the specific gaps for that seat, in order, using the GTM engineer skill stack.
- Own one metric per quarter, end to end, and record the baseline before you start.
- Publish two case studies with situation, path, result and evidence.
- Re-negotiate on scope, not on market articles. Bring the number you moved.
Learning the foundations at MitHub is free, and talent never pays a fee to get a job. Start with the Faculty of Revenue Reverse Engineering, and if you are still mapping the role itself, read what is GTM engineering first.
One last caution on numbers: Moody's data also shows this job churns hard, with about 25.9% of GTM engineers leaving within 90 days, roughly eight times the marketing rate. A high salary in a role with no mandate, no data access and no owner is not a win. Ask what you will own before you ask what it pays.