Hire your first marketer when founder-led work has already produced one channel that reliably generates qualified conversations, you can explain why it works, and the thing stopping you scaling it is hours rather than uncertainty. Before that point a marketer has to discover your market for you, which is not delegation. It is abdication.
Founders usually arrive at this decision asking who: generalist, product marketer, growth lead, agency. That question matters, but only second. The harder one is whether the business yet has something a marketer could amplify.
A marketer cannot manufacture product-market fit. They can systematize trust, sharpen the story, and turn a repeatable signal into a machine. Until there is a signal, the function belongs in the founder's hands, supported by AI tooling and fractional help where judgment is thin.
The question underneath the question
Hiring too early is expensive. So is hiring too late. The cost is rarely just salary. It is diluted positioning, a role that drifts, and a founder who concludes marketing does not work when what actually happened is that the first attempt was under-resourced and undefined.
Joel Wright's talent debt framework describes the mechanism well. A function should move up a rung only when it becomes the constraint on the next milestone. Until then the company runs on founder effort and the debt accrues quietly. Marketing is one of the last functions founders hand over, because it sits so close to positioning, trust and category learning. AI lets you fake the output for longer than before. It does not clear the debt.
What that debt looks like in practice: the founder is still writing the posts, running the calls, testing the messaging. Performance is uneven but learning is happening. Then, under pressure, the founder hires, hoping the new person will do the learning and the execution at once. That costs twice. You lose the founder's direct read on the market, and you pay someone to rediscover what you should already know.
Hire for leverage, not relief. If you are hiring mainly to escape the work, the business is probably not ready.
The readiness test
Five checks. Clear four and you are ready. Clear fewer and a full-time generalist will spend their first two quarters doing your discovery.
One, founder-led sales has closed real deals and you can describe the buyer without hedging. Not "mid-market maybe." You want five to ten customer conversations repeating the same pain, paying customers rather than pilots, and an explanation of why buyers moved now, not just why they liked the demo.
Two, one channel already repeats. Founder content, outbound that converts after a warm touch, partner referrals, events, search, community. Scale is not the test. Repetition is. If one channel works even inconsistently, a good marketer can systematize it. If none does, you need more founder discovery, not headcount.
Three, you can explain why it works well enough to repeat it deliberately. The message, the audience, the format. If the channel works and nobody knows why, it is luck with a dashboard.
Four, marketing is eating founder hours that belong elsewhere. Roughly eight to twelve hours a week on work a strong marketer could own: content, customer proof, positioning drafts, launch coordination, channel tests. Below that, keep it founder-led. Also measure energy, not only time. If marketing is fragmenting product focus and sales follow-up, it is already a drag.
Five, you have nine to twelve months of runway for the experiment. A first marketing hire is a learning bet, not a campaign purchase. If you cannot afford the time to find out what works, you cannot afford the hire.
There is a sixth test worth applying to yourself rather than the company: can you write the scorecard? A first hire whose mandate is "own marketing" is a role designed to fail. If you cannot name three concrete deliverables for the first 90 days, the problem is your clarity, not the candidate pool.
If you fail the test
Stay founder-led. That does not mean doing everything yourself forever. It means judgment stays with the founder while execution gets borrowed selectively.
Fractional help works for strategy, positioning and short-term execution, at the cost of depth and availability. An agency works for specific deliverables, campaign production and overflow, at the cost of context and accountability. For most early software companies, agencies are useful for throughput and not for judgment; fractional leaders are useful for judgment and not for daily ownership.
AI has genuinely changed this math. Drafting first-pass copy, summarizing research, repurposing content, all cheap now. That makes a small team with good judgment more viable than it was three years ago. What has not become cheaper is strategy. Use AI to compress research cycles, generate message variants, analyze call transcripts and map how buyers actually ask their questions. Do not use it to decide what earns trust.
The trap is that AI extends the window in which a founder appears to be managing marketing. The function looks covered, the debt keeps growing, and the company still has not learned its market.
Senior, junior, or neither
For most early-stage software companies the safest default is a senior individual contributor who still ships. Not a head of marketing with no hands-on bandwidth. Not a junior coordinator who needs more direction than you can give.
The obvious counter-argument is cost, and it usually backfires. Junior hires need process, feedback and worked examples. Early-stage companies have very little of any of those, which is exactly why the same hire that would thrive at a 200-person company flounders at a 12-person one.
The operator debate here is real and worth reading honestly rather than flattening. SignalFire argues startups get this wrong by over-indexing on growth before message and market fit, and favors an experienced product marketer where category education matters. Alex Kracov argues close to the opposite for many startups: a junior generalist, hired with safeguards like a take-home plan and a contract-to-hire path, on the grounds that a resume tells you very little and the ability to run many things in parallel tells you a lot. Ada Chen Rekhi emphasizes hiring for potential rather than a perfectly matched resume.
Those are not contradictory positions. They describe different stages of company maturity and different levels of market clarity. Read them as a map rather than a verdict.
The practical version: do not hire for status. Hire for compression. The first marketer should shorten cycles, not add meetings.
Match the profile to the motion
Stage 2 Capital ties the profile to the go-to-market motion rather than a generic org chart, which is the right frame. Three common shapes:
Sales-led B2B
The first hire needs product marketing instincts even if the title says generalist. They shape messaging, enable sales, and turn customer proof into narrative that a rep can actually use. This person sits between founder, sales and product.
Product-led growth
Lifecycle and content skill, plus enough analytical judgment to work across onboarding, activation and retention. In PLG, marketing is behavior design, not distribution. The marketer has to understand the product funnel, not just the top of it.
Technical categories
Security, infrastructure, data, developer tools. Here the hire earns trust with clarity rather than polish, translating technical substance into buyer language without flattening the product. This is the version of the job we see most often, and it is mostly about taking the founder's early trust-building and making it repeatable.
If you want a decision rule: weak message, start with product marketing. Clear message and a visible funnel, bias toward growth. Need both and can only afford one, take the senior generalist. Still working out the market, stay founder-led.
What to pay
Public salary data for "first marketing hire" is thin, so the honest method is triangulation across live job postings, compensation platforms like Pave and Carta, and benchmarks from startup hiring firms. The ranges below are working figures for US-based early-stage software companies in 2026, not universal bands.
Early generalist, seed stage. $110k to $150k base, 0.15% to 0.50% equity.
Senior individual contributor, Series A. $140k to $185k base, 0.10% to 0.35% equity.
Experienced product marketing or growth lead. $170k to $220k base, 0.05% to 0.20% equity.
These assume a real operator rather than a junior coordinator. If a candidate comes in above range, the question to ask is what work they remove from you and what systems exist in 90 days that do not exist now.
Equity should reflect stage and scope. Someone taking on undefined work early usually deserves meaningful upside, not just salary. And a cheaper hire at the wrong stage is a false economy: narrow execution against an unclear strategy produces activity you then have to pay someone else to undo.
How to evaluate
Most founders interview marketers as though hiring either a content writer or a VP. The first hire is neither. Test for judgment, speed, structure and proximity to revenue.
Give a thin brief and ask for a 30-day plan. Look at their assumptions, their sequencing, and what they choose to validate first.
Hand them a broken homepage, a vague value proposition, or a weak sales deck. Ask what they would change and why.
Walk through a past launch. Do they talk in tasks, or in hypotheses, outcomes and what they learned?
Run an AI workflow test. Ask how they would use AI for research, drafting, segmentation, competitive analysis and answer-engine visibility while keeping judgment their own. A strong candidate will not say AI does everything. They will tell you where it helps, where it does not, and what lightweight system they would build around it.
Keep any take-home small. One page, one ICP, one outcome. Ask for message, channel and measurement. You are not buying polished copy. You are finding out whether they think in systems.
Kracov's "button-clicker test" is the useful frame: the difference between someone who can execute a sequence and someone who can shape one. For a first hire you need both, and the second is the rarer half.
The first 90 days
Do not start the role with "own marketing." That sentence is how the function fails. Use a contract instead.
Days 1 to 30 are for clarifying the ICP and the narrative. What comes out: a one-page ICP summary, a positioning memo, and a message map built from customer interviews and your own notes.
Days 31 to 60 are one channel test. What comes out: a defined experiment with a baseline, a hypothesis, a documented process and weekly reporting.
Days 61 to 90 are for operationalizing trust. What comes out: a publishing cadence, a customer proof library, and a workflow you no longer have to hold in your head.
By the end you should have a clear positioning narrative, a tighter website and sales deck, one or two channel experiments with documented results, a simple reporting rhythm, and reusable messaging for sales and founder outreach.
If the hire is right, you should feel less cognitive load, not more. Less time explaining the product, more time making decisions. The failure signal is motion without leverage: more content, more meetings, more opinions, no clearer message, no repeatable learning.
What never transfers
The founder does not leave marketing after the hire. Some work stays founder-owned permanently, particularly in technical B2B: your point of view on the category, direct customer conversations, high-trust public writing, the narrative decisions, and the relationships that carry credibility.
The marketer does not replace that. They make it durable. Stop founder-led trust entirely and the company loses the thing that made it credible. Hand nothing over and you stay the bottleneck. The answer is a clean division of labor, not a handover.
That is the part we tend to get hired for: work that stays in the founder's voice, built so it keeps running without the founder holding every piece of it.