You built the product quickly — maybe with Cursor, Lovable, Bolt, or Claude Code — and took it from idea to working product in a few weekends. You launched. A handful of people visited. A few signed up. Product Hunt produced a temporary spike. You posted about it on X. Then everything went quiet.
This is the point where a lot of builders start looking for an SEO agency. It feels logical: you need traffic, SEO creates traffic, an agency does SEO. But there's a missing question — do you have a marketing system that's ready to be scaled, or are you still trying to discover what works? Those are completely different problems, and an agency can be excellent at the first one while still being the wrong answer to the second.
The uncomfortable truth about early-stage marketing
Most vibe-coded products don't initially have an execution problem. They have a diagnosis problem. The founder doesn't yet know which customers feel the problem most acutely, which language makes them pay attention, which searches signal genuine buying intent, which pages convert, which channels fit the product, or whether the real bottleneck is traffic, positioning, or conversion. Hiring people to execute before answering those questions doesn't remove the uncertainty — it makes the uncertainty more expensive.
This is why the agency decision so often gets made too early. The founder thinks they're purchasing growth; what they're really purchasing is activity around an unvalidated strategy. More articles, more reports, more keyword lists, more meetings — potentially very good work, aimed at a target that may still be moving.
The make-or-buy test
Before hiring an agency, identify your actual constraint.
If the constraint is time — you already know which pages bring qualified visitors, which messages convert, and you have evidence that organic discovery contributes to signups or revenue, but you simply can't execute enough of the proven work yourself — that's a strong case for an agency. You're buying capacity to scale a system you already understand.
If the constraint is specialist expertise — a regulated industry, complex international requirements, heavy local-search dependence, or a serious technical SEO problem — a specialist agency may solve something a general-purpose tool or solo founder can't handle reliably. That's also a sensible purchase.
If the constraint is uncertainty — you don't know which channel works, which searches matter, or why visitors aren't converting — this isn't a capacity problem yet. It's a learning problem, and it needs small, inexpensive experiments that shorten the distance between an idea and evidence: diagnosing the site, fixing the technical foundation, and testing a handful of specific pages before committing to a large monthly engagement.
Agencies are not the enemy
I want to be clear about this, because the internet loves false choices. SEO agencies aren't obsolete. AI hasn't removed the value of human judgement, specialist expertise, or industry relationships — the best agencies have seen patterns across hundreds of companies and can spot problems a first-time founder would miss. They're also genuinely useful for work that remains stubbornly human: relationship-driven link building, digital PR, editorial quality control, regulated content, large-scale production, and cross-functional campaigns. The problem isn't agencies. The problem is asking an agency to scale something before you know whether it deserves to be scaled.
What I would do before signing a retainer
If I had just launched a vibe-coded product, I'd spend the next 90 days reducing uncertainty.
First, fix the technical floor. Before producing more content, make sure search engines can properly discover and understand the product — sitemap submission, indexing and crawlability, page titles and descriptions, canonical tags, internal links, mobile performance, structured data that actually describes the page, and Search Console wired up. Not glamorous, but it stops every future page starting with a handicap.
Second, choose three narrow questions. Not a broad keyword like "project management software" — a new product won't outrank established companies for a category that broad. Look for three specific questions asked by the exact person the product was built for: "how do freelance developers track billable hours without spreadsheets," "what's the simplest CRM for a two-person consultancy." A specific question isn't just lower competition — it reveals a specific problem, which makes it easier to write a useful answer and connect it naturally to the product.
Third, publish one genuinely useful answer per question. Not ten generic AI-generated posts — three pages that could reasonably be the best answer available for those exact questions, each with a direct answer near the top, first-hand knowledge or original examples, honest limitations, and a relevant next step. AI can help research, structure, and edit. The founder still has to supply the perspective that makes it worth reading — that distinction matters more as the internet fills with competent but interchangeable content.
Fourth, measure behaviour, not production. Publishing a page isn't the outcome. Track whether it gets indexed, earns impressions, attracts relevant visitors, and moves any of them toward the product. You don't need perfect revenue attribution at this stage — you need enough evidence to distinguish useful work from activity.
When I would hire the agency
Once I could say something like: these three pages attract the right people, this topic produces qualified signups, we understand the message, now we need help expanding the system — that's a far stronger brief than "we launched, traffic is low, we need SEO." The first gives an agency a working signal to amplify. The second asks it to discover your market, positioning, channel, and execution plan all at once. Some agencies can do that, but it's a different engagement, a different level of strategic involvement, and usually a larger budget than straightforward SEO execution.
Where AI marketing tools fit
The interesting opportunity isn't "AI versus agencies." It's what happens between doing everything manually and being ready to hire a full team. Early-stage founders need a system that can inspect the product and its current visibility, diagnose the most important constraint, recommend the next piece of work, help execute it, measure what happened, and use that result to choose the next action. That's the gap we're building Chad to fill — not another dashboard of hundreds of recommendations, and not sixteen disconnected agents to manage. Just an answer to: what's preventing this product from being discovered, and what should we do about it first?
The decision in one sentence
Hire an SEO agency when you have a working organic playbook and need help scaling it. Before that, use the cheapest and fastest path available to discover what the playbook should be. Fix the technical foundation. Test specific queries. Publish useful answers. Measure what attracts the right people. Then scale the evidence — not the hope.
Also published on Medium.