Where the GTM engineer role came from, and why it exists now
Clay says it coined the GTM engineer title in 2023 and counts about 100 GTM engineer job listings going live every month (Clay, checked 2026-11-10). Bloomberry's analysis of GTM engineering and RevOps postings found a 205% year-over-year increase from January to September 2025 (Bloomberry).
The honest version: it overlaps heavily with RevOps
Bloomberry compared the responsibilities in GTM engineer and RevOps engineer postings and found 9 of the top 10 in common. RevOps postings lead with CRM ownership and forecast accuracy; GTM engineer postings lead with building automated workflows and integrating the stack, with outbound and prospecting emphasized. It also found almost no overlap with sales engineering, which is customer-facing. So a GTM engineer is best understood as the builder half of RevOps, pointed at pipeline.
What a GTM engineer actually builds
The six systems a GTM engineer is usually hired to ship, and how each one quietly breaks. If you are hiring, ask candidates how they found out one of their builds broke.
| Build | Inputs | Output | How it breaks |
|---|---|---|---|
| TAM list and ICP scoring | Firmographics, technographics, closed-won and closed-lost accounts | Every account in the market, scored for fit with a reason | ICP drifts from what actually closes; nobody re-scores after a pricing or segment change |
| Enrichment waterfall | A record missing email, phone or title; several data providers in priority order | The record filled by the first provider that returns a verified value, at the lowest cost | Provider order never revisited; credits burn on records that should have been filtered first |
| Signal plays | Hiring, funding, job changes, tech-stack changes, site visits, product usage | A trigger that starts a specific play for a specific account, with the reason attached | Too many signals fire, reps learn to ignore the alerts |
| Lead and account routing | Territory rules, owner of record, account tier, rep capacity | Every new lead or signal lands with exactly one owner, fast | Edge cases (existing customer, open opportunity, partner-sourced) route to the wrong person |
| CRM data model | Objects, custom fields, stages, required properties, dedupe keys | A CRM that agents and reports can trust: one account per company, current stages, clean fields | Fields multiply, duplicates creep in, stage definitions mean different things to different teams |
| Agent workflows | Prompts, tools, data access, approval rules | An AI agent doing research, drafting, scoring or CRM updates on a schedule or a trigger | No evaluation loop: output quality drifts and nobody samples it; agents write to fields they should not |
A sample week in the role
An illustrative week (not a survey result) for a GTM engineer at a 50 to 200 person B2B company with an outbound team. Notice how little of it is building something new.
| Day | Main block | What ships or gets decided |
|---|---|---|
| Monday | Pipeline and signal review with sales leadership; triage the request queue | Top three requests turned into builds, the rest declined or deferred with a reason |
| Tuesday | Build: a new signal play (for example, accounts that just hired a RevOps leader) | Trigger, enrichment, routing and a draft message template, tested on 50 accounts |
| Wednesday | Maintenance: enrichment waterfall costs, failed syncs, duplicate accounts | Provider order adjusted; a dedupe rule tightened; a broken webhook fixed |
| Thursday | Agent quality review: sample 25 agent emails and 25 CRM updates | Prompt or permission changes for anything wrong, off-brand or written to the wrong field |
| Friday | Adoption and impact report; office hours for reps | What ran, who used it, meetings and pipeline attributed; next week's build chosen |
GTM engineer skills and tools, by layer
Think of the stack in four layers. Tool shares come from Bloomberry's analysis of GTM engineer job postings.
| Layer | What you need to be able to do | Tools named in postings |
|---|---|---|
| Data | Source accounts and contacts, build enrichment waterfalls, verify emails and phones, dedupe against the CRM | Clay (the most-named tool), Apollo (29% of postings), ZoomInfo, intent and visitor-identification tools |
| Orchestration | Move data between systems, trigger plays on events, handle failures and retries | Zapier (39%), n8n (28%), webhooks, Python scripts |
| Systems of record | Design objects, fields, stages and routing; write back safely; respect validation rules | HubSpot (52%), Salesforce (45%), SQL against a warehouse |
| Execution and AI | Sequences, call workflows, prompts and agents, evaluation of AI output | Outreach (49%), Gong (23%), LLM APIs and AI coding tools |
- Code: SQL and Python each appear in 38% of GTM engineer postings (Bloomberry). You do not need to be a software engineer, but you need to fix your own broken script.
- Commercial judgment: knowing which play is worth building. Most practitioners came from SDR or RevOps roles (Bloomberry).
GTM engineer vs RevOps vs sales ops vs GTM agents
"GTM engineer vs RevOps" is the comparison buyers and candidates search for, but in 2026 there is a fourth option on the org chart: AI agents that do some of the same work. Here is how the four compare.
| GTM engineer | RevOps | Sales ops | GTM agents | |
|---|---|---|---|---|
| Primary output | Working systems: lists, waterfalls, plays, routing, agent workflows | Process, governance, forecasting, reporting | Territories, quotas, comp plans, CRM admin for the sales team | Executed work: researched accounts, sent outreach, booked meetings, updated records |
| Measured on | Adoption, meetings and pipeline from what they built, hours saved | Forecast accuracy, pipeline visibility, process adoption | Rep productivity, quota attainment, data accuracy | Qualified meetings and pipeline per dollar |
| Builds or runs? | Builds, then hands over | Designs and governs | Administers and supports | Runs, continuously |
| Best when | Your GTM motion needs custom data and plays as a lasting edge | You have several teams and need one source of truth | You have a sales team large enough to need planning | You need the work done now, with human review, before a team is hired |
The roles are not either-or: in teams that run agents, the GTM engineer or RevOps lead usually sets the agents' data access, approval rules and quality checks. See GTM 2.0 and the best GTM agents compared.
GTM engineer salary in 2026: what the data says
GTM engineer salary figures online disagree widely because they measure different things: posted ranges, self-reported survey pay, modeled estimates, base or total. Here are four sources we read on 2026-11-10, with what each one counts.
| Source | Figure | What it counts | Date |
|---|---|---|---|
| Bloomberry job-posting analysis | $127,500 median | Salary ranges posted in GTM engineering job ads that published pay | Updated 2026-01-25 |
| State of GTME survey (OneGTM) | $135,000 median; middle half $105,000 to $165,000 | Self-reported base from 61 US in-house GTM engineers; survey of 228 people, Nov 2025 to Feb 2026 | Updated 2026-09-30 |
| Supered posted-range pull | $151,200 to $220,000 (median low and high ends) | 13 posted base ranges at 10 venture-backed companies hiring systems builders; a sample, not a survey | 2026-10-02 |
| US Bureau of Labor Statistics | $124,900 median | Sales engineers, the nearest established federal occupation (no GTM engineer category exists) | May 2025 data |
What moves the number
- Location: in the State of GTME survey the US median base was $135K against $75K for respondents outside the US.
- Coding depth: respondents who rated their coding 7 or higher out of 10 reported a $135K median, against $90K for those at 3 or below.
- The job behind the title: Supered found three clusters in postings: campaign and list roles near $100K, systems builders in the $137,000 to $250,000 range, and software engineers on GTM teams at AI labs posting up to $405,000.
How to benchmark an offer (or a budget)
- Match the job, not the titleCount how many responsibilities are requests from one team versus systems the whole team runs. Price against the matching cluster above.
- Separate base, bonus and equitySurvey figures here are base. About half of State of GTME respondents reported no bonus, and two thirds held under 0.10% equity.
- Use posted ranges firstA posted range is the employer's own number. New York, for example, requires employers with four or more employees to post one; ask for it everywhere else.
GTM engineer job description template (copy and edit)
Written around outcomes rather than a list of tools, which attracts builders. Replace the bracketed parts, and benchmark the pay range against the salary section first.
About the role
As our GTM engineer you will design, build and maintain the systems that create and work our pipeline: our target account list, enrichment, buying-signal plays, lead routing, CRM data model and AI agent workflows. You report to [Head of RevOps / VP Sales] and work daily with [SDR, AE and marketing] teams. Success is measured by what reps actually use and the pipeline it produces, not by tickets closed.
What you will do in the first 90 days
- Audit our CRM data model, dedupe rules and enrichment spend, and fix the top three issues.
- Rebuild our TAM and ICP scoring from closed-won and closed-lost data, with a plain-English reason for every score.
- Ship two signal-triggered plays (for example [new funding] and [new leader in role]) end to end, from trigger to routed task or sequence.
- Put a weekly quality review on every AI agent or automated write-back to the CRM.
What you bring
- [2 to 5] years in RevOps, sales ops, growth or as an SDR who automated their own work.
- Hands-on with [Clay or similar], [HubSpot or Salesforce] administration, and an automation tool such as [Zapier or n8n].
- Working SQL; Python or JavaScript for scripts and API calls.
- Experience building or evaluating LLM prompts or agents, including how you checked their output.
- A portfolio: two systems you built, who used them, and what changed.
How to become a GTM engineer
Bloomberry's review of 100 GTM engineer LinkedIn profiles found the most common path starts as an SDR or BDR who automates their own prospecting, then RevOps and sales ops. Nearly all were self-taught.
- Automate your own job firstIf you are an SDR or ops analyst, build the list, enrichment or follow-up workflow you wish you had. A real problem with a real user beats any course.
- Learn one tool per layerA data and enrichment tool, one CRM in depth, one automation tool, and enough SQL to query a warehouse. Add Python when a no-code step fails you.
- Ship something other people runTurn your workflow into something three teammates use without you. That is the line between an operator and a GTM engineer.
- Measure itTrack usage, meetings and pipeline from what you built. That record is what the higher-paid postings ask for.
When to hire a GTM engineer vs buy agents with a Forward Deployed Engineer
A GTM engineer builds the machine; an agent platform with an embedded engineer rents you a machine that is already running. Several agent vendors now offer engineers this way: Attio lists onboarding with forward-deployed engineers on its Ultra plan, Outreach mentions FDE resources to help deploy agents, and 11x offers managed deployment (vendor sites, checked 2026-11-10).
| Your situation | Lean toward | Why |
|---|---|---|
| Custom data and plays are your competitive edge, and you have 12 months of runway for the role | Hire a GTM engineer | The systems and the knowledge stay in-house and compound |
| You need pipeline or a clean CRM this quarter, and nobody in-house can build it | Buy agents with an engineer | You get the outcome while you decide whether the role is worth a full-time hire |
| You already have a GTM engineer who is buried in maintenance | Both | Agents take the repetitive execution; your engineer sets their rules and reviews quality |
| You sell mostly by phone | Check the calling layer specifically | Most agent platforms stop at email and hand calling to a rep-operated dialer |
Frequently asked questions
What does a GTM engineer do day to day?
Builds and maintains the systems behind pipeline: target account lists and scoring, enrichment waterfalls, signal-triggered plays, lead routing, the CRM data model and AI agent workflows. A typical week mixes one new build with maintenance, quality review of automated output and a report on what reps actually used.
What is the difference between a GTM engineer and RevOps?
RevOps governs the revenue process: CRM ownership, forecasting, reporting and alignment between teams. A GTM engineer builds working automations and data systems, usually focused on outbound and pipeline. Bloomberry found 9 of the top 10 responsibilities overlap, so many companies treat GTM engineering as the builder function inside RevOps.
How much does a GTM engineer make?
It depends on what the source counts. Bloomberry's job-posting analysis found a $127,500 median; the State of GTME survey reports a $135,000 US median base; Supered's sample of posted ranges at venture-backed companies centers on $151,200 to $220,000. The BLS median for sales engineers, the nearest federal occupation, was $124,900 in May 2025.
What should a GTM engineer job description include?
The systems to own, a 90-day plan, how success is measured (adoption and pipeline) and a posted pay range. Use the template above.
- 1.Clay, GTM Engineering: What It Is, How It Works, and How to Hire (checked 2026-11-10) (2026)
- 2.Bloomberry, I analyzed 1000 GTM Engineering jobs, here is what I learned (updated 2026-01-25; checked 2026-11-10) (2026)
- 3.OneGTM / State of GTME, GTM engineer salary data from the 2026 State of GTME survey, 228 respondents (updated 2026-09-30; checked 2026-11-10) (2026)
- 4.State of GTME, The State of GTM Engineering report and methodology (checked 2026-11-10) (2026)
- 5.Supered, GTM Engineer Salary 2026: What the Postings Pay (posted ranges pulled 2026-10-02; checked 2026-11-10) (2026)
- 6.US Bureau of Labor Statistics, Occupational Outlook Handbook: Sales Engineers, median pay May 2025 (checked 2026-11-10) (2026)
- 7.Attio, Pricing (Ultra: onboarding with forward-deployed engineers), checked 2026-11-10 (2026)
- 8.Outreach, Pricing (FDE resources to help deploy agents), checked 2026-11-10 (2026)
- 9.11x, Platform (managed deployment options), checked 2026-11-10 (2026)
Spend your reps' day on conversations, not ringing phones.
TwinsAI parallel-dials up to 10 lines at once, validates every number with the carrier before it rings, and bridges a live prospect to your rep in under 400ms. It works inside HubSpot, Salesforce, Close, Attio, Outreach, Zoho, HighLevel, Pipedrive, Apollo and SmartMoving, with no seat minimum.