Before Product-Market Fit: Which Search Tasks Can Wait?

Before product-market fit, a startup should do a small amount of SEO: get the technical basics right once, publish a few pages for people who already search for the problem you solve, and have a founder write up what sales calls teach. Put off anything that assumes your positioning is settled, such as big keyword campaigns, programmatic pages and link building. For some startups, search isn't the right channel yet at all, and it's better to know that before spending on it.

Why the timing matters

SEO is slow and it compounds. A page you publish in March might start bringing visitors in the fall and keep doing it for years. That's the appeal, and before product-market fit it's also the risk. Say a Seattle team selling invoicing software to contractors pivots to job costing four months in. Most of the pages, keywords and links built around "invoicing" stop being useful the day the homepage changes.

So the rule at seed stage is simple: do the parts that stay useful whatever you end up selling, and keep the parts that depend on your positioning small.

Do now: the technical basics, in one afternoon

These carry over to any version of the product.

1. Pick one domain and stick to it. Choose www or the bare domain, redirect the other, and serve everything over HTTPS. Moving domains later costs far more than deciding now.
2. Keep preview and staging builds out of Google. They can get indexed if anything links to them, so put them behind a password or send a noindex header.
3. Separate the app from the marketing site. Login screens, signup steps and app routes should be noindex. Marketing pages should be indexable and should arrive as HTML with the text in it, not as an empty shell that fills in after scripts run.
4. Set up Google Search Console and submit a sitemap. Run the URL Inspection tool on your homepage to confirm Google sees the text you think it sees.
5. Write page titles that say what the product does. "Acme: The future of work" tells a searcher nothing. "Job costing software for small contractors | Acme" does.
6. Check mobile speed with PageSpeed Insights. Marketing sites built on heavy animation can score badly, and it's easy to fix while the site has five pages instead of fifty.
7. Track signups and demo requests as conversions, by landing page. Later you'll want to know which pages bring buyers rather than readers.

Google's own Search Essentials covers the baseline rules in a few pages and is worth one careful read by whoever owns the site.

Do now: three to six high-intent pages

Not a blog yet. A handful of pages for people who have the problem today and are looking for something to fix it.

Find the topics in your sales calls, not in a keyword tool. Write down the exact words prospects use before they've heard your pitch. Say a Bellevue startup sells scheduling software for dog groomers. Its prospects might say "grooming appointment app", "how do I cut no-shows" and "something better than a paper book". Check those phrases in a keyword tool, then turn the ones with any search demand into pages:

• A homepage that names the category in the buyer's words, not your pitch deck's.
• One page for each customer type you're actually selling to now. If you serve mobile groomers and salons differently, that's two pages.
• A pricing page, even if pricing is early. Explain how it works and who each option is for.
• One honest comparison against what prospects use today, which is often a spreadsheet, a paper book or a general-purpose tool.

A question can deserve an answer even when only a few prospects ask it. At this stage, fit matters: does the page explain a problem the product solves today, and can someone act on that answer? Keep a note of the conversations behind each topic so you know what to revisit if the product changes.

Do now: founder-written content

Founders know things nobody else at the company knows yet: why the existing tools fail, what customers tried first, what they say when they cancel. Write one post a month from that material, or record a conversation and have it edited. The topics come straight from calls: a teardown of the usual workaround, the questions every prospect asks in the first five minutes, what changed your mind during your first twenty customer interviews.

The test for pre-PMF content is whether it pays off without search. A good founder post is one you'd send to a prospect after a call, share in a community you belong to, or hand to a new hire on day one. If it ranks later, that's a bonus. If you can't picture sending it to anyone, skip it.

Postpone until the product settles

• Keyword research across hundreds of terms, and a content calendar built on it.
• Programmatic pages, where one template is filled out for every city, industry or integration.
• Link building, including buying a link package from us or anyone else. Links strengthen pages that already convert. They can't fix positioning that's still moving.
• Broad educational content: glossaries, "what is" posts, beginner guides to the whole category.
• Pages for other countries or languages.

Wait to expand until sales and support can describe a repeatable buyer problem. For each proposed page, write down the customer question, the product feature that answers it and the next action a reader can take. If those three pieces keep changing, maintain the existing page and record what you learn rather than commissioning a large batch. Stable product details give a writer something useful to explain and keep a small team from rewriting its entire site after each release.

The guide SEO vs PPC for Seattle Businesses: Where the Budget Goes Further can help you think through acquisition spending while your product is changing.

When SEO is the wrong channel

Be honest with yourself about these, because they come up a lot at seed stage.

• You're creating a category. If buyers don't know your kind of product exists, they don't search for it. Founder-led sales, communities and events create the demand first, and search catches it later.
• Your market is a short list. If you sell to a few hundred named companies, a founder emailing the right people will do more than any ranking.
• You need to learn fast. Search feedback takes months. Customer calls and small paid tests answer messaging questions in weeks. (We don't run ads for clients, so that isn't a pitch.)
• Your runway is short. If the money runs out before a new page could realistically rank, spend on what closes deals this quarter.

SEO does make sense before product-market fit when prospects already search for the problem in words you can see in your call notes. In that case the handful of pages above are cheap to build and keep working through the next pivot, as long as the problem stays the same.

The guide Which Product Pages Should a Software Team Build First? can help you prioritize pages that explain a working product.

A plan for the next two quarters

• Month one: technical basics, Search Console, and three to six high-intent pages.
• Every month after: one founder post, a look at the Search Console queries, and an update to the pages using phrases from new calls.
• Every quarter: ask whether the positioning moved. If it held and the pages bring signups, that's the signal to scale search.

When that signal shows up, our SaaS SEO service covers the scaling part: keyword research across the category, content on a schedule, and links to the pages that convert. Until then, keep it small and keep your URLs stable.