How to Use Claude for Content Strategy

Picture this: you're a content lead at a 40-person fintech startup, it's Monday morning, and your quarterly content calendar is still a blank spreadsheet. You open Claude, type something like "help me plan a content strategy for my blog," and get back a list of 15 topic ideas so generic they could belong to any company in any industry. That's the most common way people learn how to use Claude for content strategy — and it's also the fastest way to end up with a calendar full of posts nobody searches for.

Before: The Weak Prompt

Here's roughly what most people start with:

Help me create a content strategy for my company's blog. We're a 
fintech startup that helps small businesses manage cash flow. Give 
me some blog post ideas for the next quarter.

What this does: it gives Claude your industry and product, but no information about your audience's actual questions, your competitors, or what's already been covered. Claude fills the gap with safe, broad topics — "5 Tips for Managing Cash Flow," "Why Cash Flow Matters for Small Business" — the same ideas any generic search would surface.

Why It Fails

This prompt fails because it asks for a strategy but only supplies inputs for a topic list. A real content strategy needs a point of view: who you're writing for, what stage of awareness they're at, what your competitors have already covered, and what you want the reader to do next. Without those constraints, Claude has to guess — and it guesses toward the middle, which is exactly where content strategy shouldn't live.

⚠️ Common mistake: asking for "a content strategy" when what you actually want is a list of blog topics. These are different requests. A strategy includes audience segments, funnel stage, and distribution — a topic list is just the output of a strategy, not the strategy itself.

After: The Improved Prompt

I run content for a fintech startup that helps small businesses (5-50 
employees) manage cash flow forecasting. Our audience is mostly 
bookkeepers and small business owners who are already aware they have 
a cash flow problem but haven't picked a tool yet (consideration 
stage, not awareness stage).

Our two biggest competitors are Float and Pulse. Float's content 
leans heavily on templates and spreadsheets. Pulse focuses on 
integrations with accounting software.

Build a quarterly content plan (12 posts) that:
1. Targets consideration-stage search terms our competitors haven't 
   fully covered
2. Includes a mix of comparison content, workflow guides, and one 
   data-driven report we could produce from anonymized customer data
3. For each topic, note the search intent and what a reader should do 
   after reading (subscribe, book demo, download template)

Output as a table: Topic | Search Intent | Funnel Stage | CTA

What this does: it gives Claude the competitive picture, the funnel stage, and the desired output format, so the resulting plan reflects an actual strategic gap instead of a generic industry overview.

⚡ Pro tip: naming your specific competitors and what they already cover is the single most useful addition you can make to a content strategy prompt. Claude can't identify a content gap it doesn't know exists.

Breaking Down Each Element

Each part of the rewritten prompt is doing a specific job. The audience description ("bookkeepers and small business owners, consideration stage") stops Claude from writing awareness-stage "what is cash flow" content you don't need. The competitor breakdown gives it something to differentiate against, rather than writing in a vacuum. And the output format — a table with funnel stage and CTA columns — forces every topic to justify itself instead of just sounding plausible.

The request for one data-driven report is worth calling out separately. A marketing manager at a B2B logistics company added a nearly identical line to her prompt ("suggest one report we could build from data we already have") and it surfaced an idea — a quarterly "shipping delay index" — that ended up being her highest-performing piece of content that year, earning links from three industry newsletters. Claude is actually good at spotting these opportunities, but only when you tell it what data you have access to.

⚡ Pro tip: if you have any proprietary data — usage stats, survey results, transaction patterns — mention it explicitly. Data-driven content is harder for competitors to copy and Claude will often suggest angles you hadn't considered.

Variations for Different Contexts

A solo freelance copywriter building content for a client can use a scaled-down version — swap "12 posts" for "6 posts" and drop the competitor analysis if the client hasn't given you access to competitor research. A content team at a larger company (say, a 200-person SaaS company with a dedicated SEO analyst) can add a line requesting search volume estimates and keyword difficulty ranges, though Claude's estimates here should be treated as directional, not authoritative — pair them with actual keyword tool data before finalizing.

For e-commerce brands, swap the funnel-stage framing for "pre-purchase, post-purchase, retention" since that maps better to how shoppers actually move through a buying journey than the B2B awareness/consideration/decision model.

Agencies managing content for multiple clients run into a different problem: reusing the same base prompt across accounts without adjusting the competitive inputs, which produces plans that all sound suspiciously similar. One agency account lead started keeping a short "competitive brief" for each client — three sentences on positioning, two named competitors, one content gap already identified — and pasting that in fresh each time rather than editing an old prompt. It added maybe five minutes per client, but the resulting plans stopped reading like templates with the company name swapped out.

There's also a scale question worth addressing directly: how many topics should you actually ask for in one pass? Twelve works well for a quarter because it's specific enough to force prioritization but not so long that Claude starts padding with filler topics to hit the count. Asking for 30 or 50 topics in a single prompt tends to produce diminishing quality toward the end of the list — the last ten ideas usually feel noticeably thinner than the first ten, since the strategic constraints you gave (competitors, funnel stage) can only differentiate so many distinct angles before repeating itself. If you genuinely need a year's worth of topics, it's actually more reliable to run four separate 12-topic prompts, one per quarter, than one 48-topic prompt in a single pass.

One more variable worth adjusting deliberately: how much you tell Claude about past performance. If you have data on which of your existing posts actually drove signups versus just traffic, include a short summary of that before asking for new topics. A newsletter-driven B2B company found that adding one sentence — "our comparison posts convert at 3x our how-to posts, weight the plan accordingly" — shifted the resulting topic mix meaningfully toward comparison content, without needing to explain why comparison content converts better in the first place. Claude doesn't need the reasoning, just the constraint.

Save and Reuse This

The structure here — audience, competitors, constraints, output format — works for almost any content strategy request, not just quarterly planning. Swap in a single competitor and a narrower topic and you can use the same skeleton to plan a single pillar page and its supporting cluster.

Once you've got a prompt structure that reliably produces plans you actually use, don't rebuild it from memory every quarter. Save it in something like PromptABCD so the next planning session starts from a prompt that already works, instead of a blank page and a generic first draft.