How to build a B2B content strategy

What a B2B content strategy actually is

A B2B content strategy is the plan that decides which topics you publish, for which buying stage, and to move which metric. It is not a calendar, and it is not a word count. It is the set of decisions that give every piece of content a job before anyone writes a sentence.

Most content programs skip this step and go straight to production. The result is volume without direction: fifty posts that each rank for nothing, convert no one, and compete with each other for the same thin keyword. A strategy exists to prevent that — to make sure the next piece fills a gap rather than adding to a pile.

That job assignment points somewhere counterintuitive. Most programs build from the top down — awareness first, because it is the easiest to write and draws the most traffic. The higher-return order is the reverse, and it is where this guide starts: write the bottom of the funnel first.

Write the bottom of the funnel first

The strongest move in content strategy is the one most programs skip: publish your bottom-of-funnel content before your top-of-funnel content. Write the pages that serve a buyer who has largely decided — pricing, proof, comparisons, the objection procurement raises — before the explainers that introduce the problem to a stranger.

This cuts against the standard instinct, which starts at the top because awareness content is easier to write and shows the largest traffic numbers. But traffic is the input, not the goal; influenced revenue is. Bottom-of-funnel pages serve the buyers closest to a decision, so they convert where a top-of-funnel explainer cannot, and they are the pages a sales team can send into a live deal tomorrow. Start there, prove that content moves revenue, then build the awareness layer — which now has finished pages to point buyers toward instead of dead-ending them.

Why the funnel is still the right organizing unit

Bottom and top mean nothing without a defined funnel, so name the three stages you are sequencing against. The buyer's journey has three broad stages, and each asks a different kind of question:

• TOFU — top of funnel. The buyer has a symptom and is researching the problem. They do not yet know solutions exist. Their question starts with what and why.
• MOFU — middle of funnel. The buyer knows the problem is solvable and is comparing approaches and vendors. Their question starts with how and versus.
• BOFU — bottom of funnel. The buyer has largely chosen and needs to justify the purchase internally. Their question is about price, proof, and risk.

Organizing by stage matters because the stages are not interchangeable. A pricing page cannot do an explainer's job, and an explainer cannot close. When you sort your topics into these three buckets, two things become visible immediately: where you have no coverage, and where you have five pieces competing for one job.

Stage	Buyer's question	Content that fits	Job
TOFU	"What is this problem?"	Explainers, guides, definitions	Earn qualified traffic and trust
MOFU	"How do I solve it, and with whom?"	Comparisons, frameworks, worked examples	Capture intent, assist pipeline
BOFU	"Why you, and at what cost?"	Pricing, proof, case studies, ROI tools	Justify the decision

Map content to audience × value, not just funnel stage

Funnel stage is one axis. On its own it is not enough, because two buyers at the same stage can want opposite things. The second axis is audience × value: which persona you are writing for, and which value that persona is trying to get.

Every piece of content should map to one (audience) × (value) pair before it earns a slot. The audience is a persona in the buying committee — the economic buyer, the end user, the technical evaluator. The value is the outcome that persona is measured on, stated in their terms. It is not a feature, and it is not a generic claim like "saves time." A page that cannot name its persona and that persona's value is a page with no reader.

Build the map as a grid. List your personas down one side and each persona's two or three values across the top. Each cell is a content slot — the questions that persona asks about that value. The same subject fills different cells for different readers: an "onboarding" topic is a retention story to a buyer and a time-to-value story to an end user, and those are two pages, not one.

Persona (by buying influence)	Their top value	A slot that fits
Economic buyer	The business outcome they own	"What [problem] costs you" — a BOFU justification
End user	The daily job made easier	"How to [do the task]" — a MOFU how-to
Technical evaluator	Fit, risk, integration	"How [product] handles [requirement]" — a BOFU proof page

Two rules decide how much you write where:

• Weight by buying influence, not seniority. The persona who most shapes the decision gets the most coverage, even when their title is junior.
• Weight within a persona by value rank. A persona's first value earns more pieces than their second, and their second more than their third.

If your team struggles to fill the grid, that is not a creativity problem. It is a signal you are too far from your customers to know what they actually want.

Content is how your positioning travels

A content library that is not tied to your positioning produces pages that rank and still sell nothing. Content is the layer where your position reaches a stranger. It has to carry the position, not wander from it.

Think in three layers. Your why is the change in the world you exist for. Your how is the value you deliver differently from everyone else. Your what is the product and the content that describes it. Most programs write only at the what layer — features, how-tos, definitions — and never connect a page back to the value and the position above it. That content informs and never persuades.

So give every piece a second job beyond its funnel job: it should express one value from your messaging, in the buyer's own words. When the whole library ladders up to the same handful of values, a reader who lands on any single page absorbs part of your position. When it does not, you have a pile of pages that each argue for a slightly different company.

This is content-market fit: the point where the topics you own, the words you use, and the values you claim all line up with what your buyer already searches for and believes. You reach it by writing for the buyer's search, not the brand's ego. The page answers the question the buyer typed, not the one your product team wishes they had typed.

How to build the plan from real questions

The reliable way to build the plan is to start from questions your buyers actually ask, not from a keyword tool in isolation. Your sales calls, support tickets, and search data are the raw material. Every recurring question is a piece of content waiting to be written.

Collect the questions before you cluster them

List the questions across the whole journey — from the first symptom a buyer notices to the final objection a procurement team raises. Do not edit yet. A messy inventory of real questions is more useful than a clean list of keywords, because the questions carry the intent that tells you which stage they belong to.

Tag each question by intent, not by topic

Sort each question into TOFU, MOFU, or BOFU by reading the intent behind it. The phrasing usually gives it away: what is and why does are TOFU; how do I and X versus Y are MOFU; pricing, alternatives to, and is X worth it are BOFU. The same subject can appear at all three stages — and often should, as a set of linked pages rather than one page trying to serve everyone.

Write in the buyer's words, not yours

The questions you collected are worth more than a keyword tool's output because they carry the buyer's own language. Keep that language. The fastest content improvement available to most teams is to mirror the exact words prospects use and stop translating them into internal vocabulary.

Watch two traps. The first is jargon — describing the product through your own technical pride when the buyer describes their problem in plainer words. The second is the assumed-shared term. Words like "analytics," "platform," or "automation" mean different things in different rooms, and a buyer who does not share your definition reads straight past the page. Define the term the first time you use it, or use the word the buyer already uses. Write for the search the buyer runs, not the one you would prefer they ran.

Cluster tagged questions into topics

A topic owns a cluster of related questions, not a single keyword. Ten questions about the same problem become one thorough page that answers all ten, internally linked to the narrower pages around it. This is how depth beats volume: one page that genuinely owns a topic outranks and outlives ten thin pages that each graze it.

Give every piece a metric

The publishing order is already set — bottom of the funnel first, then the layers that feed it. What remains is accountability. Give every piece one accountable metric before it ships:

• TOFU answers to qualified traffic and returning visitors, not raw pageviews.
• MOFU answers to email capture and assisted pipeline.
• BOFU answers to influenced revenue.

Judging a TOFU explainer by direct conversions punishes it for doing the job you assigned it. Each stage is measured against its own job, and the plan is healthy when all three jobs are staffed.

Depth over volume, and why gating is a trap

The instinct under a content quota is to publish more. The better instinct is to publish less and go deeper. A thorough page earns rankings, gets cited, and keeps returning value for years; a thin one decays the week it ships. Twenty pages you can make the most thorough on their topic beat one hundred you cannot.

Gating is where many programs undo their own work. Gate the deliverable, not the teaching. A template, a calculator, or a worksheet is a fair trade for an email — the buyer gets a tool. Gating the explanation itself hides it from search engines and AI crawlers, which is exactly where B2B discovery now begins. The article stays open; the artifact behind it can ask for something in return.

Depth also compounds through structure. A set of related pages that link to each other — a thorough hub with narrower pieces branching off it — reads to both a human and a crawler as a body of work on a subject, not a scattering of unrelated posts. The internal links do two jobs: they route a reader to the next relevant piece, and they signal which pages are central to a topic you mean to own. This is why the plan thinks in topics rather than keywords. A topic is a cluster you can build a linked structure around; a keyword is a single page with nowhere to point and nothing pointing back.

Build a pillar page and cluster around it

Depth is easier to earn when pages are organized, not scattered. The structure that does this is the pillar and cluster: one broad page that covers a subject end to end, surrounded by narrower pages that each go deep on one part of it, all linked back to the pillar and out to each other.

The pillar page answers the big question — the topic a buyer would name if you asked what they are trying to solve. The cluster pages answer the specific questions underneath it, one per page. The pillar links down to each cluster page; each cluster page links up to the pillar and across to its siblings. To a reader, this routes them to the next relevant piece. To a crawler and to an AI model, the linked set reads as a body of work on a subject — which is what earns the pillar its standing on the topic.

This is why the plan thinks in topics, not keywords. A topic is a cluster you can build a structure around. A keyword is a single page with nothing pointing to it and nowhere to point. Build one pillar per subject you mean to own, fill its cluster over time, and keep the internal links current as you publish. The structure is what turns twenty separate pages into one asset that ranks and gets cited.

The three kinds of content you can make

Everything above is about one kind of content. There are three, and the companies that come to own a category produce all of them:

1. Industry expertise — content about your product, your service, and the problem you solve: guides, whitepapers, research, case studies, docs. This is what "content marketing" usually means, and it targets customers, prospects, and the influencers around them. It is also the one most companies fail even at, which is why the whole plan above is about doing it well.
2. Corporate branding — content about how your company operates: how you decide, how you organize, what you believe. It has little to do with your product. Basecamp's Signal v. Noise is the archetype — writing about how they work, not what they sell — and it built a following that a feature page never could. This layer makes a brand relatable.
3. Personal branding — content from your people. A CEO's or a VP's personal brand is part of how the market sees the company: a VP of sales posting weekly on cold outreach, an engineer writing about a hard technical call. Jason Fried carries Basecamp; David Cancel's podcast carried Drift's culture.

The reason to do all three is reach and trust. A company's employees collectively hold far more connections than the company account has followers, and people trust a person more than a logo. So make personal branding part of the job rather than a hobby, and treat corporate and personal content as surfaces the industry content travels on — not distractions from it. Most companies produce only the first kind; good ones add the second; the ones that come to define a category produce all three.

Distribution is the other half of the job

Publishing is not finishing. A page nobody sees does the same work as a page that does not exist. Production has stopped being the bottleneck — once a model drafts faster than any team can edit, the scarce work moves to selection and distribution. Plan how a piece reaches its reader with the same rigor you plan the piece itself.

Treat your own surfaces as a channel you own outright: the site, the newsletter, and the product. The product is the most underused of these. In-product education — the guide that appears where the buyer is stuck, the explainer that ships next to the feature — reaches people who are already using you, and it draws on something no competitor can copy: how your customers actually behave inside the product. Thought leadership grounded in that real behavior lands because it is true, not because it is loud.

One long asset should not stay one asset. A thorough pillar page is raw material for a set of derivative pieces — a short explainer, a checklist, a post, an email. Write the deep thing once, then distribute it in the formats your buyer already reads. Match the channel to the audience × value pair, the same grid you planned against: the value a buyer cares about decides which surface reaches them, and the persona decides the format. The plan is not done when the piece is published. It is done when the piece is in front of the person it was written for.

What the plan looks like when it is done

The finished deliverable is a one-page TOFU/MOFU/BOFU plan: the three columns, the topics that fill each, the gaps that do not yet have coverage, and the next five pieces to ship in order. It fits on a page on purpose. A plan the team cannot hold in its head is a plan the team will not follow, and a content strategy nobody follows is a calendar with extra steps.