When should a startup hire its first growth marketer?

A growth hire needs a motion, not a round

Michael Seibel at Y Combinator says stay lean and keep burn low until you are swamped by happy, loyal, and ideally paying customers. Headcount and a funding round are the proofs he tells founders to ignore. Elena Verna and Andrea Wang, after interviews with a dozen growth leaders, hold full-time growth roles until you are expanding a fit you already have, with early proof the distribution method works. Casey Winters, in that piece, calls the job connecting more people to value that already exists.

When should a seed-stage startup hire its first full-time growth or marketing person?

A seed-stage startup should hire its first full-time growth or marketing person only when a segment already keeps the value and one path already produces demand you would run again. Adam Fishman, in March 2025, keeps a growth hire with the founders at pre-seed and seed. He allows one only as a maybe, for a curve that is already a vertical line; everyone else waits. Early users and a small revenue stream are not the signal. His 2022 note calls a great growth push the fastest way to kill a bad product. He staffed one paid channel before it had proved itself, and that channel then weakened.

For a marketing seat, Ken Babcock waits for a painkiller people adopt, a way to distribute it, and feedback that no longer blocks use. About 15 to 18 people and about 30,000 users came first at his company, and headcount was not the trigger. Unplanned growth from users was. Taru Wahlroos at Vendep sees the first B2B marketer most often from late seed into early Series A, then says the round matters less: paying customers, ideally outside the founders’ personal network; a sales motion you are about to repeat, with collateral and a channel test; and a founder whose week is already full of copy, the site, and outreach. Not pre-revenue, not a frequent pivot, not an unnamed buyer, rarely a VP, and not a hire whose only job is ads or a prettier brand. Maya Spivak sends a growth marketer to end users, and product marketing only after sales must convince a named buyer.

Arielle Jackson sets no weekly hour quota. A seed motion is ready to scale when you would run that path again on purpose.

Five checks, and a miss keeps the role closed

Open the seat only when a segment keeps the value, one path shows a signal you made on purpose, you can name the number, the site and the trail a stranger leaves are already in place, and you can carry the tests through a learning cycle while you stay close. Miss one, and the seat stays closed.

1. A segment keeps the value: they stay, pay again, book a second conversation, or refer without a campaign. Otherwise a bad tactic and a bad product look the same.
2. One path shows a signal you made on purpose: conversations from content, outbound at a steady rate, a paid test you would repeat, or usage that creates more usage.
3. You can name the number: signups, activated users, qualified conversations, or revenue from that path. Verna says the founding team writes the first model: one lever (acquire, keep, or get paid) by one motion (product, sales, or marketing).
4. Spivak will not hire a growth marketer to build the basics. You need a usable site and a signup a stranger can finish. Track a visit, a lead, an activation, a second action, and a source. Fishman says that without quantitative data, much of a growth toolkit does not work.
5. Jackson rejects a full-time hire for one short paid test. A better case opens three to five further channels. Content and community can come sooner: the limit is time, not media.

The way you sell moves the clock

How customers arrive moves the timing more than the round.

Jackson puts a consumer path, organic, paid, public relations, or content, near the first 10 people. That headcount is a US venture pattern; elsewhere, the five checks decide. Wahlroos separates B2B SaaS from consumer and ecommerce playbooks: a product-led motion toward smaller buyers and an enterprise motion with multi-month cycles and account-based marketing need different playbooks. Enterprise waits until the sale repeats, near a sales leader taking the founder’s seat. A hardware launch a month out is already late, because distribution and packaging are due. Spivak fills the gap the founders lack and wants range, because the motion may change within six months. If the buyer reads another language, the first scope is that language’s page, outbound, and proof. Otherwise you are hiring for a market you cannot yet sell to.

Where it sticks	First person	Not yet
Acquisition, and one path already converts	Growth marketer on that path	Offer, page, or signup still blocks
Activation or retention, and the loop is in the product	A product person who can ship the loop	Channel marketer who cannot ship it
Getting paid, and the sale already repeats	Story and leads, not a new brand team	Deals are still founder heroics
Consumer hardware	Distribution and packaging before launch	A head of growth, and no team
You cannot point to the stuck step	Founders. Do not hire to accelerate growth	Discovery

Verna and Wang say a new growth team starts on acquisition about nine times in ten, because that is where the biggest drop-off usually sits, and because getting paid waits until the team has earned trust. Match the first hire to the step that is stuck: acquisition, activation, engagement, or getting paid. Fishman also rejects the acquisition-only leader. His bias is a product-shaped hire, but the motion you found decides.

Use a smaller contract until the checks pass

Until the five checks pass, Jackson sends one discrete job, such as a name or a short paid test, to a freelancer or an agency. Weigh a freelance growth marketer against a full-time hire and outsourcing against hiring internally.

Form	Use it when	Stop it when
Fixed scope, end date	One page, sequence, or pilot	Scopes stack and the story has no owner
Set days each week	The metric is real, a full week is not	The work needs daily shipping
Full-time builder	All five checks pass	You wanted a manager, or the lever is in the product
Agency squad	A known channel, one bounded test	They only recommend their own channel

A fractional seat before a head of marketing owns one metric, not a department. To stop doing all the marketing, hand one converting motion to the early-stage marketer with depth on that gap.

Put a 90-day target on one page

If you cannot write a 90-day target on one page, do not open the search. Jackson puts that search at about three to six months unless the person is a friend or former colleague you have already worked with. Tristan Gillen expects no significant measurable output in month one, a channel signal in months three to six, and a verdict on the person around month six. Look inside first: Verna names analytics, product, marketing, engineering, or financial planning, and Fishman adds support, because an outsider copies the last company.

1. Who buys, why they buy, and which sentence has already produced a yes. If this is empty, the founders still own it.
2. One metric, one path, a kill rule, and the date you will read the result.
3. Access: analytics, the site, the CRM, any ad account, and three customers they may call in the first month.
4. What stays with you: the offer, who counts as qualified, and approval on anything a customer will see.
5. At day 90, review the test, not the person: keep it running only if the next test is obvious.

Hire a builder who still does the work

The first growth hire is a builder, not a head of growth asked to invent the model.

Verna starts with a builder (growth marketer, growth product manager, or lifecycle) on the path that already shows a signal. One-channel optimizers come only after that path holds; if the hypothesis failed, an innovator who invents loops comes next.

For a first marketing generalist, Jackson looks for about five to eight years, still tactical, titled Head of Marketing or Director, not an early VP. For this growth seat, Fishman will not open a Head of Growth search with nobody to lead. He hires a player-coach a notch below Director or Head of, who still ships. Gillen draws the other line: a growth lead finds and scales channels, while a head of marketing owns brand, communications, and the whole function. Spivak wants hands-on work for a few quarters, and a candidate who skipped your signup is out. Ask which experiment they ran, which tactic they refuse, and which part of the channel they owned. Ad text is not the channel. Skip Fishman’s celebrity leader: a big-company name who has likely not worked at this stage, and needs big budgets or big teams, advises your builder instead. Ask what they built from nothing, not what moved after they joined. Refuse a post that treats brand, performance, product, and communications as equals.

Leave the seat empty in these situations

Keep the seat closed when no segment stays, no path is worth running again, the buyer is unnamed, there is no quantitative trail, the loop is in the product and nobody can ship it, or the brief is fatigue, an investor request, or one bounded test.

You are late when a converting path has no owner, inbound sits unanswered, or the founder’s week is that motion. Gillen calls it late when the company is locked into channels that are hard to undo. Stop a habit channel that does not produce demand before the search, or the hire spends the cycle undoing it.

How I would time the first seat

For a freshly funded startup with a path, one freelance marketer can cover marketing, AI, and automation anywhere, remote, from the first message to the first customers, with no separate specialist for every part. I am Piet Baudoin, one person under the name Poldermarketing, AI-native, Dutch and English, strong in AI, content, automation, and workflows. Google Ads and Meta Ads are newer for me: I set them up and review them, and I do not scale a large paid program. I build and run the work, freelance or fractional, on a fixed scope with an end date or a few days a week. The free growth scan shows how the site reads to a buyer. How I work is the engagement. Positioning and messaging is where I start when the sentence is the constraint.