SEO for B2B SaaS: what to do at each stage

Stage determines the work, not company size

The usual advice is stage-blind. It describes what to do without asking where the company is, which is why so much of it produces nothing. Organic compounds, and compounding is only an asset when the thing being compounded stays true. Before positioning settles, content is a liability that will need rewriting.

Stage	Stable enough to build on	Do this	Not yet
Pre PMF	Almost nothing	Crawlable site, brand terms, notes from sales calls	Any content programme
Early Series A	Competitors, integrations	Comparison, alternatives, pricing, integration pages	A blog
Late Series A	Problem space	Content types, link rules, first complete cluster	Scaled publishing
Series B+	Category position	Update near-ranking pages, consolidate, build tools	–

The useful question is not “what should we do for SEO” but “what is stable enough here to be worth compounding”.

This assumes the general principles are already familiar. If not, the operator’s guide to B2B SEO covers them first.

Pre product-market fit: almost nothing

The correct amount of SEO work here is close to zero, and this is genuinely hard advice to accept when organic is cheap and the runway is long.

The reason is that your positioning will change, probably several times. Content written against positioning that changes has to be rewritten or deleted, and deleted content that ranked is worse than content never written, because you also lose the links.

What is worth doing: get the site technically crawlable, claim your brand terms so people who hear about you can find you, and write down what you learn from sales conversations. That last one is not SEO, it is the raw material for SEO later, and it is far more valuable than any post you could publish now.

Early Series A: commercial intent only

Positioning has firmed up. Now the highest-return work is the pages closest to a purchase decision, because they convert immediately and rank faster than educational content.

• Comparison pages against the two or three competitors that come up most in sales calls.
• An alternatives page for the incumbent your buyers are leaving.
• A pricing page that actually explains the model rather than saying “contact us”, since pricing is among the most searched terms for any B2B product and refusing to answer sends the visitor to a competitor who does.
• Integration pages for whatever your buyers already run.

Deliberately not yet: a blog. Skip it entirely at this stage. It consumes the budget that should be building the pages above, and it will not rank for anything valuable for months.

Late Series A into Series B: architecture and the first cluster

This is the point at which building a system pays off, because there is now enough volume of work to need one.

Define the content types as a closed set. Write the link rules into the brief template. Build one complete cluster around the problem space you want to own, pillar plus five to eight spokes, linked as a hub.

Hire or contract an internal owner before you contract production capacity. Content produced without someone internal who can answer product questions is generic by necessity, and generic content in B2B does not convert regardless of how well it ranks.

What that architecture contains, quarter by quarter, is covered in the B2B SEO strategy piece.

Series B and beyond: depth, tools, and consolidation

Now scale is worth it, and the constraint shifts from whether to publish to whether the existing pages are working hard enough.

Three activities carry most of the return. Update the pages that nearly rank, since a page at position eleven is far cheaper to move than a new page is to create. Consolidate overlapping pages, because two pages at position fifteen usually make one at position six. And build the free tools, which by now have the domain authority behind them to actually rank and earn links.

This is also the stage where AI-search visibility starts to matter commercially rather than theoretically. Assistants cite passages, so the work is making sections self-contained, stating claims before elaboration, and naming entities specifically rather than hedging them out.

The underlying economics that make this worth funding at all are set out in the B2B SaaS channel maths.

What each stage costs, and who does it

Stage does not just determine the work. It determines who should do it, and getting that wrong is more expensive than getting the tactics wrong.

Stage	Right owner	Common mistake
Pre PMF	A founder, an hour a week	Hiring an agency to produce content against positioning that will change
Early Series A	One in-house marketer, briefed narrowly	A generalist asked to do brand, demand and SEO at once
Late A into B	In-house owner plus contracted production	Contracting the strategy as well as the writing
Series B+	A small team, with the system documented	A large team with no written architecture, producing volume

The middle two rows are where most programmes are staffed wrongly

The pattern across the rows: the thinking stays inside as early as you can afford it, and the production moves outside as soon as the thinking is stable enough to brief against. Reversing that, which is the common shortcut, produces content that reads fluently and answers no product question specifically.

The cost question has a similar shape. Pre product-market fit the right spend is close to zero, and that is genuinely hard advice to accept when a competitor is publishing weekly. At Series B the right spend is whatever keeps a documented system running, which is usually less than the same output would cost from an agency and considerably more than nothing.

The two mistakes that recur at every stage

Stage-specific advice covers most of it, but two errors appear at every stage in different clothing, and they cost more than any tactical choice.

Doing the next stage’s work early

At pre PMF it is publishing a blog. At Series A it is building an architecture for content that does not exist. At Series B it is starting a second cluster before the first is finished. The pattern is the same: work that would be correct later, done now, producing nothing and consuming the budget for work that would have paid.

Measuring against the wrong stage’s expectations

Judging a pre-PMF site on traffic. Judging a Series A programme on pipeline in month four. Judging a Series B programme on publishing volume. Each of these ends a programme that was working, because the number being watched was never the number that stage produces.

Both come from the same root: treating the stages as a description of company size rather than as a description of what is stable enough to build on. A well-funded Series B company that has just changed its positioning is, for this purpose, back at the start.

The practical test before starting any piece of work: name the thing it depends on being stable, and ask whether that thing has actually stopped moving. If it has not, the work is early regardless of what the funding round says.

Reading your own stage honestly

The hardest part of stage-matched work is not knowing what each stage requires. It is admitting which stage you are actually in, because the answer is frequently earlier than the funding round suggests.

Four questions settle it, and none of them are about revenue.

• Has the positioning changed in the last six months? If yes, content written against it will need rewriting, and you are effectively pre product-market fit for this purpose regardless of your stage.
• Can someone internal answer a product question in writing, this week? If not, any content programme will produce generic material, and the fix is hiring rather than publishing.
• Do you know which two competitors come up most in sales calls? If not, the comparison pages that carry early-stage pipeline cannot be written yet.
• Is there a page a rep would send today if it existed? If yes, that is the next page, whatever the keyword volume says.

The last question is the most useful and the least used. A page that a salesperson would forward on a live deal has a value that no keyword tool will show, and it is almost always closer to the buying decision than whatever the content calendar had scheduled.

The pattern across all four: they test whether the thing you would build on has stopped moving. That is what stage means here, and it is why a well-funded company that just repositioned is, for content purposes, back at the beginning.

What to stop doing at each stage

Stage-matched advice usually lists what to start. The more useful half is what to stop, because most teams are already doing work that suited an earlier stage and nobody has said so.

On reaching	Stop	Because
Product-market fit	Writing about the problem in general terms	You now know who buys and why. Write for them, not for the category
Early Series A	Publishing on a schedule for its own sake	Cadence without a cluster produces volume nobody finds
Late Series A	Treating every page as equally important	Three to five pages will carry the commercial traffic. Find them
Series B	Adding pages as the default response to any gap	Updating a page at position eleven beats creating a new one

The last row is the hardest change to make and the highest return

The Series B row deserves emphasis because it inverts a habit that worked earlier. Early on, more pages genuinely is the answer: you have too few, and each new one adds coverage. Past a certain size the calculation flips, and a page already at position eleven is a far cheaper win than a page that does not exist.

The signal that you have crossed that line: you have pages ranking on page two that nobody has looked at in a year. If that is true, the next quarter of work is in the existing library rather than the content calendar.

Moving between stages without restarting

The transitions are where work gets thrown away, and they are more avoidable than they look. Each move to the next stage should keep most of what came before, and the parts that do not survive are predictable.

What carries forward at every transition: the architecture, the internal owner’s product knowledge, the technical foundation, and any page tied to how the product works rather than to how it is positioned.

What does not: anything written against a positioning that has since changed, comparison pages naming competitors you no longer meet, and pricing explanations that describe a superseded model.

~70%

of pages survive unchanged

~20%

need the positioning updated

~10%

should be removed rather than fixed

Higher survival than most teams expect, which is an argument for building earlier rather than waiting

That first number is the useful one, because the fear of writing content too early usually assumes it will all be wasted. Most of it is not. The waste concentrates entirely in positioning-dependent pages, which is a reason to sequence those late rather than a reason to publish nothing.

The practical version at any transition: list the pages that name a competitor, state a price, or describe who the product is for. Those are the ones to review. Everything else almost certainly still stands.

The pattern across every stage

One thread runs through all of it: build the thing closest to the buying decision that is stable enough to still be true in a year.

Pre product-market fit, almost nothing is stable enough. At Series A, your competitors and your integrations are. At Series B, your category position is. The work follows the stability, not the other way around, and programmes fail most often by doing Series B work on a pre-product-market-fit company.

Matching the work to the stage is most of what I do in growth consulting engagements.