Should an early-stage startup hire a growth marketer or a marketing agency?

Name the job before you pick a supplier

Hire a growth marketer or an agency only once the founders can name who buys, why they stay, and which piece of work is stuck. Neither contract will discover that.

Sam Altman, writing for Y Combinator, treats early use as learning, and growth before a product people love as a leaky bucket: people come in and do not stay. His signal is that users actively recommend the product to friends, and a net promoter score can measure that love. Enterprise and hard tech are included, because the sale gets easier once word has spread. Airbnb slogged for 1000 days before the product was loved, and word of mouth then carried growth. If the product is not loved, ads, marketing, and PR have to carry growth instead, and raising to grow a product people do not keep leaves the leak in place.

Arielle Jackson, in First Round Review, wants the actual problems, not a wish to grow or an investor’s nudge. One scoped project, a name or an online ad test, is a freelancer or an agency. An inside owner earns a seat when managing freelancers who are not talking to each other has become the founder’s job and the public work looks uncoordinated.

A growth marketer scales a motion you can already show

A growth marketer owns experiments on a funnel that already exists. Hire when a channel produces customers and tests are queueing, and skip the title while the site, the offer, or the signup still has to be invented.

Maya Spivak, in First Round Review, defines the seat as experiments on acquisition, engagement, and retention, with several routes ready or a freelancer, contractor, or agency to call. The bar is a decent website and an easy signup. They spread those materials, put paid distribution behind them, and test why people do not convert. Building the site and the signup is a different job.

Y Combinator’s growth office hours describe a different seat. Once retention is strong, often after about 15 to 20 people and at least two product managers, the next product manager may be a growth product manager working with engineers, not this marketer and not an agency.

Match the motion. Product-led work needs familiarity with ads, email, social, search, and paid search, not five expert crafts. Sales-led work needs case studies, credible copy, testimonials, and webinars. Brand work (reputation, voice, campaigns that ship) fits either motion. Founders who already design, and who already have a site and a brand, should hire the analytical gap. A hard-to-explain product needs a translator. See when to hire and which marketer fits.

The first person stays on the tools for at least a few quarters, shipping quick wins while you learn whom to hire next. Prefer a T-shaped marketer, because the motion, product line, or packaging can change within six months.

Arielle Jackson puts a full-time search at about three to six months, unless the person is a friend or former colleague with a strong working relationship. About ten people is her first-marketer bar only for a consumer brand, or a marketing-driven event such as organic, paid, PR, or content, not for a growth seat on a live channel. Enterprise SaaS adds marketing after a repeatable sales model, to sharpen the story and open channels, once the motion is solid and the founder starts looking for a sales leader. Consumer hardware that needs distribution, packaging, and pricing before ship needs a marketer earlier than a sales-driven enterprise company, for the launch, not to scale a live channel. Seek about five to eight years, still hands-on in one or two areas. Pass a leader who left the tactics and a junior who only wrote ad lines. Use Head of Marketing or Director. Vice president on day one limits later seniority, and the job is single player. Those headcount and title rules are a US venture norm from her 2019 interview, still on the page in its 2024 update. Do not open the search while the founders are still learning who stays, or if you cannot wait out several months. A past large team is a poor fit with no team. One small paid test is a weak case for the seat. Content or community can still justify a person before media spend is heavy. Press before a headline, a reporter, and a story is a positioning and product-marketing problem.

An agency is a scoped craft with a judge on your side

An agency, or a specialist freelancer, is the right buy for a project you can describe, date, and grade. It is a weak way to learn who the customer is.

Bowery Growth calls a traditional agency a weak way to validate a channel, because the model repeats one specialty across clients, and a shop that claims every channel at this stage is usually bluffing. Renting a team until a channel is repeatable fits only a bounded project the founder can grade. It fails when that team is asked to find the buyer. Ask for work at a company this early, still on the tools. A scale-up title is not that evidence.

Emily Kramer, in MKT1, keeps audience, story, positioning, and the lifecycle inside before any agency. Founders can run one or two contractors, not a stack. Content and search wait on a distribution path. Paid waits on a first set of ad lines and a landing page. Prefer a contractor or a small niche shop. Ask for the people, the same stage, the same sales motion, and a reference. Keep creative off the media buyer. Someone inside owns a program of site, content, paid, and search. One urgent artifact during the search can sit outside. That program cannot.

Say yes to one sentence: named person, artifact, date, and the signal that keeps or kills the work. Refuse a channel menu, a hidden login, an account only they can see, and any brief nobody on your team can approve. Compare a fractional operator and an agency and outsourcing versus an internal hire at seed.

Four contracts sit under the two titles

Hire neither while the founders are still learning who stays. Buy a scoped agency or freelancer for one artifact you can grade, a few days a week when one channel needs steady tests, and a full-time growth marketer only when that channel already produces customers.

Contract	Already true	You are buying	Leave it closed when
Founder keeps the work	You are still learning who stays	The next conversations and a small test	A live channel sits on the founder’s calendar
Scoped agency or freelancer	You can grade one artifact	A missing craft, with an end date	The gap is a weekly judgment
Part-time operator, a few days a week	One channel needs steady tests, short of a full seat	Hands in your accounts, easy to stop	The work is a single asset
Full-time growth marketer	A channel produces customers and tests are queueing	One metric, next to product	Site, offer, or buyer is still missing

How to hire a freelance growth marketer screens the days-per-week row. Marketing help without a full-time hire is when those days are the whole plan. Ignore marketplace ranges. Use the cell you can already check.

Run this sequence before you sign

Staff growth only after the product is recommended, the stuck work is named, and the contract matches a weekly rhythm or a single artifact. Stop at the first check you cannot answer.

1. If users do not actively recommend the product to other people, fix the product before you staff growth.
2. Hardware that needs distribution, packaging, and pricing before ship needs a marketer for the launch, sooner than a sales-driven enterprise company.
3. Stack-rank product, performance, creative, and communications, then lead with one and rent the rest.
4. No pass through your signup ends a growth interview.
5. Require a fixed scope, a maximum number of hours, and an end date.

What stays on the founder’s side of the table

Whoever signs, the company keeps the definition of a customer, the offer sentence, the accounts, and the stop decision.

Keep ad accounts, the site, the CRM, and analytics in the company’s name. A login only they hold means you kept an output and lost the learning. Each week, one metric, one bet, one stop condition, and a founder who approves anything a prospect will read.

Screen the person. Jackson wants an advisor or a friend in a last conversation before the offer, because founders misread marketing experience. If a line says they drove growth, ask which part they owned. Writing the ad lines is not owning the campaign. Ask when they last did the tactic. For a product marketer, Spivak wants work that is still live and the name of who decided. For a growth marketer, ask what they refuse. Decline a tactic whose point is to spam people who never asked, such as an address-book import.

When each option is the wrong buy

Skip a growth marketer when nothing is ready to distribute, when the gap is positioning or product marketing, or when you want a leader and have no team. Skip an agency asked to name the customer, asked for work you cannot grade, or asked to run the weekly call, and skip both while the founders can still run the next customer conversations and the next small test.

How I would take one side of this

I would take one named job with an end date, or a few days a week on one channel that already has a metric.

I am Piet Baudoin. I work alone, as Poldermarketing, remote, in Dutch and English. Hire me freelance for a bounded project or for set days each week. I am an AI-native growth marketer who builds and executes in the accounts. I am strong in AI, content, automation, and workflows. Google Ads and Meta Ads are newer for me: I set them up and review them. Scaling a large media program sits outside that.

When a freshly funded startup wants one person for the whole path, I take that as an all-in go-to-market engagement, fully remote, from the first message to the first customers. The free growth scan shows how the site reads to a buyer now. Positioning and messaging is the work when the words are the constraint. How I work is the shape of a project or a few days a week.