A growth engineer ships production code that moves signup, activation, and retention metrics. A growth marketer ships campaigns, lifecycle flows, and attribution dashboards against the same metrics. A growth hacker is a 2010-era term for a scrappy generalist running channel-level experiments. The roles share a north star (revenue), but the artifacts ship to different stakeholders, the reporting lines diverge, and the 2026 salary bands differ by 30--50%. This article is the side-by-side matrix existing content refuses to give you.
What is the difference between a growth engineer, growth marketer, and growth hacker?
A growth engineer ships production code. A growth marketer ships campaigns and lifecycle flows. A growth hacker is a 2010-era label for a scrappy generalist running channel-level experiments before either of the modern roles existed.
The roles differ in three concrete ways:
- What they commit to. Engineers commit code to a Git repo. Marketers commit campaigns to a calendar. Hackers commit hacks to a Notion doc.
- Who reviews their work. Engineer output goes through a code review. Marketer output goes through a brand or legal review. Hacker output goes through a founder.
- Where they sit. Engineers sit in engineering or a growth pod. Marketers sit in marketing. Hackers historically sat next to the founder.
Sean Ellis coined 'growth hacker' in 2010, defining it as 'a person whose true north is growth.' Gergely Orosz's Pragmatic Engineer newsletter defines a growth engineer as a software engineer who 'thinks less about running experiments and more about building the infrastructure that enables those experiments.'
This is what growth engineering actually is: the engineering discipline applied to growth metrics, not a rebrand of marketing.
What does each role actually ship in a typical week?
The clearest way to tell these roles apart is to look at the artifacts they produce in a five-day sprint. The work product, not the title, is what matters.
Growth hacker (Monday--Friday):
- Spin up 5 channel tests across LinkedIn, Reddit, cold email
- Write 12 ad copy variants and load into Meta Ads
- Pull a GA4 report on yesterday's landing page
- Cobble a Zapier flow that pipes form fills into HubSpot
- Pitch the founder on a viral referral mechanic
Growth marketer (Monday--Friday):
- Brief the agency on Q3 paid creative
- Ship a new lifecycle journey in Iterable or Customer.io
- Review CAC and LTV by channel in a Mixpanel dashboard
- QA the latest landing page from the design team
- Run an attribution audit between GA4 and Salesforce
Growth engineer (Monday--Friday):
- Ship an A/B test on the signup flow in Next.js, gated behind a PostHog feature flag
- Write SQL in dbt to evaluate last week's onboarding experiment
- Build an internal tool that lets marketing self-serve campaign UTM generation
- Instrument events on a new product surface
- Pair with product engineering on the activation funnel
The overlap is the analytics layer. The divergence is whether you write the code that emits the events. The full growth engineer toolkit for 2026 is heavy on Cursor, dbt, PostHog, LaunchDarkly, and TypeScript.
Is growth engineering the same as growth hacking?
No. Growth engineering and growth hacking share a north star (move metrics, fast) but differ on durability, ownership, and craft.
Growth hacking is a tactic. Sean Ellis defined it in 2010 as scrappy, hypothesis-driven experimentation across the funnel. The classic artifacts are clever loops, channel arbitrage, and one-off hacks that scale until they stop working.
Growth engineering is a discipline. It treats growth like any other engineering problem: build infrastructure, write tests, ship to production, monitor in observability tools, refactor when the system gets complex. The artifacts survive after the person leaves.
Four concrete differences:
| Dimension | Growth Hacking | Growth Engineering |
|---|---|---|
| Time horizon | Days to weeks | Quarters to years |
| Output | One-off hacks | Reusable infrastructure |
| Failure mode | Hack stops working | Test framework outlives the team |
| Seniority of practitioner | Generalist | Senior software engineer |
In 2026, calling yourself a 'growth hacker' on LinkedIn signals seed-stage scrappiness. Calling yourself a 'growth engineer' signals you can ship code that finance will rely on next quarter.
How much does each role earn in 2026?
Compensation is the cleanest signal that these are different roles. The base salary spread between a senior growth engineer and a growth marketer is 30--50% in the US market as of April 2026.
Growth hacker / specialist: $80k--$130k base. The role barely shows up in modern levelling tables. Most people calling themselves growth hackers in 2026 are seed-stage founders or junior generalists.
Growth marketer: $75,615 (25th percentile) to $141,148 (75th percentile), per Glassdoor (April 2026) based on 166 reported salaries. Growth marketing managers earn $97,071--$175,645 per Glassdoor's manager band based on 875 reported salaries.
Growth engineer: Salary data is bimodal. Salary.com lists $84k--$135k for the title at the broad market. Senior growth engineers at Series B+ startups and FAANG-tier companies earn $150k--$260k+ base, with total comp pushing $400k+ once equity is included, consistent with senior software engineer bands on Levels.fyi.
The arbitrage opportunity in 2026: many companies still post 'growth engineer' job specs that require shipping React, SQL, Python, AND running paid campaigns, but pay at the marketing band. If you have the engineering chops, get hired into engineering.
Does a growth engineer report to engineering or marketing?
In 2026 the dominant pattern is solid line to engineering, dotted line to a Head of Growth or VP Marketing. But there is no industry consensus, and the answer is a leading indicator of how serious the company is about growth as a discipline.
Three real-world models:
- Embedded in marketing, dotted to engineering. Buffer's senior growth engineer reports to the Head of Engineering Operations but sits on the marketing team. Common at companies under 200 people.
- Independent growth org with its own engineers. Facebook's growth team is structured as a business unit under a GM, with backend, iOS, Android, and ML engineers reporting up through the growth VP to the CEO. Performance marketing sits inside growth.
- Bottoms-up hybrid. Pinterest treated growth as everyone's job, with a small core of designated growth engineers and marketers implementing initiatives proposed company-wide.
The smell test. If your growth engineer has no code review process, no on-call, and no engineering manager 1:1, they are a marketer in disguise and you are paying them too little.If they cannot get changes into production without a marketing approval, they are an engineer in disguise and you are paying them too much.
For a deeper breakdown of org charts at 50/100/250-person companies, see our growth team structure benchmarks.
Which role should a Series A startup hire first?
Hire a growth marketer first at Series A. Hire a growth engineer at Series B or after $5M ARR. This is the consensus from PostHog, Lenny Rachitsky, and most operator-investors writing about growth team composition in 2025--2026.
The reasoning is constraint-based. At Series A, your bottleneck is usually channel discovery and demand generation, not shipping speed. A growth marketer with paid, content, and lifecycle reps is the unlock. PostHog's guide to starting a growth team is explicit: 'until you've reached product-market fit, you don't need a growth engineer.'
The decision tree:
- Is your activation rate < 30%? Hire a growth engineer. Activation is a product engineering problem.
- Is your CAC stable but you cannot scale spend? Hire a growth marketer. You need new channels, not new code.
- Are you running fewer than 2 experiments per month because engineering is blocked? Hire a growth engineer. Engineering capacity is the constraint.
- Do you have less than $2M ARR and < 10k MAU? Hire a generalist (call them growth hacker, growth lead, or head of growth) until volume justifies specialization.
- Do you have product-market fit and $5M+ ARR? Hire both. Wire the engineer into the growth pod, the marketer into demand gen.
For a longer treatment, see who to hire first: growth engineer or growth marketer.
What kills each role in interviews?
Each role has a signature failure mode in interviews. Knowing the failure mode tells you what the job actually is.
Growth hacker fails when: they cannot produce an ICE-ranked backlog. The whole job is generating and prioritizing experiment ideas. If they cannot rank 5 ideas in real time, they have no system.
Growth marketer fails when: they cannot trace a number to its source. 'CAC dropped 20%' is not enough. The interviewer wants the GA4 segment, the attribution model, the SQL query, or the Iterable journey that produced the lift. Per Brian Balfour's essay on growth interview mistakes, candidates also fail by 'portraying themselves as order-takers' instead of strategic operators.
Growth engineer fails when: they cannot connect a pull request to a metric movement. The whole point of the role is closing the loop between code shipped and number moved. If a candidate walks through three projects and never names the metric, the experiment design, or the lift, they are a generic full-stack engineer applying for a growth role for the comp.
The other universal failure, per Balfour, is not articulating 'the why.' Why this project, why this company, why this metric. Junior candidates over-index on what they did. Senior candidates lead with why.
When should you call yourself a growth engineer vs a growth marketer?
If you ship code to production at least weekly and that code is the lever that moves a growth metric, you are a growth engineer. If your weekly artifact is a campaign, a brief, or a dashboard built on someone else's data pipeline, you are a growth marketer. If you are a founder doing both because you have to, you are a growth hacker, and that is fine until you raise a Series A.
The 2026 wrinkle is that AI tooling has compressed the boundary. Tools like Cursor, Lovable, and Replit let marketers ship landing pages, internal tools, and small product features through prompting. The floor on what marketers ship is rising. The ceiling on what engineers automate is also rising. Net effect: the role overlap zone is wider than it was in 2024, but the salary delta has not closed, because production code that survives a code review still requires the engineering rigor that prompting alone does not enforce.
| Dimension | Growth Hacker | Growth Marketer | Growth Engineer |
|---|---|---|---|
| Primary artifact shipped | Campaign hacks, copy tests, viral loops | Lifecycle flows, paid campaigns, attribution dashboards | Production code, feature flags, internal tools, experiment infra |
| Typical week | Spin up 5 channel tests, write copy, pull GA reports | Brief agency, ship Iterable journey, review CAC by channel | Ship A/B test in Next.js, write SQL for retention, build internal tool |
| Reports to | Founder or CMO | VP Marketing or Head of Growth | VP Engineering (dotted to Growth) or Head of Growth |
| Core stack | Zapier, Webflow, Meta Ads, ChatGPT | HubSpot, Iterable, Customer.io, GA4, Mixpanel | React/TS, SQL, Python, PostHog, LaunchDarkly, dbt |
| Avg US base (2026) | $80k--$130k | $97k--$175k (manager band) | $150k--$260k+ (senior) |
| Stakeholder shipped to | Founder, CMO | Demand gen, content, product marketing | Product, data, marketing |
| Origin | Sean Ellis, 2010 | Mid-2010s evolution of digital marketing ops | Facebook / Pinterest / Uber growth orgs, 2014+ |
| What kills them in interviews | No ICE-ranked backlog | Can't trace a number to a SQL query | Can't articulate a metric hypothesis |