What a LinkedIn Ghostwriter Costs (and the AI Alternative)

By Jack Stovell · 2026-09-12 · Guides

Here's the thing about ghostwriter pricing: nobody publishes it, and everyone pretends that's normal.

Ask around and you'll get vague gestures instead of numbers. "It depends." "Depends on volume." "Depends who you ask." Fair enough, it does depend, but the vagueness isn't really about complexity. It's about the fact that the whole industry runs on bespoke quotes, and bespoke quotes are murkier by design.

So let's talk about what you're actually paying for, because once you see the line items, the AI question gets a lot easier to answer honestly.

The market, roughly

At the low end you've got junior ghostwriters and generalists, often working with several clients at once, charging what amounts to a part-time retainer. Fine for founders who want consistent posting and don't need much strategic input. In the middle sits the bulk of the market: experienced writers who'll do proper interviews, build out a content calendar, iterate on drafts, and generally act like an extension of your comms function. That tier costs real money, month after month, and most serious LinkedIn creators who use ghostwriters sit somewhere in this band.

At the top end are the executive specialists. C-suite clients, high volume, sometimes a full content strategy bolted on. Those retainers run into territory that makes most founders wince, and honestly, for the right person, it's worth it.

Notice what actually drives the price. It's not the writing. It's the interview time, the revision rounds, and the sheer relentlessness of doing this every week without the voice drifting.

What you're really paying for

Strip the invoice down and four things sit inside it.

First, interview time: someone sitting with you, extracting your opinions, your war stories, the phrasing you reach for without noticing. Second, drafting itself, which is the part everyone assumes is the expensive bit and usually isn't. Third, voice-matching, done by feel, built up over weeks of trial and error and "no, that's not quite how I'd say it." Fourth, revisions: the back-and-forth where you correct tone until it lands.

That third one, voice-matching by feel, is the real cost centre. It's slow because it's subjective. There's no shared reference point between you and the writer for what "sounds like you" actually means, so you end up doing round trips. Draft, feedback, redraft, feedback again. Weeks sometimes, just to lock in a tone that then has to be maintained forever after.

Where AI tools usually fall down

Here's where most people's AI experiment goes wrong. They open a chatbot, paste in a topic, and out comes something fluent, competently structured, and generic in a way that's hard to name but easy to feel. It reads like nobody in particular. Confident, polished, forgettable.

That's not a model problem. It's a measurement problem. The tool was never told what "you" sounds like in any specific, checkable sense. It's guessing at a general register of "professional LinkedIn voice," and that register belongs to everyone and no one.

This is exactly the gap ScriptGrain was built to close. Instead of asking a model to improvise your voice from a vague prompt, it measures it: 45 separate attributes pulled from your actual writing, sentence length variance, contraction rate, how you open paragraphs, where you place your strongest sentence, and dozens more. That measurement becomes a profile, and the profile is what gets enforced when new drafts are generated.

What changes when voice becomes a measurement

The practical difference is this: the "does this sound like me" question stops being a feeling and starts being a score. You're not eyeballing a draft wondering if it's close enough. You're checking it against a profile built from your own writing, which either matches or doesn't.

That collapses the slow part of the ghostwriting cycle. Drafts come back instantly instead of after a days-long turnaround. Revisions shrink, because the starting point is already closer to your actual register, not a generic approximation of "thought leader voice." You can test this yourself for free, no card needed, by building a profile and running one analysis, or just running a sample through the free style tool at scriptgrain.com/tools/writing-style-analysis to see what gets measured.

Where a human ghostwriter is still the right call

None of this makes ghostwriters obsolete, and I'd be lying if I said otherwise.

Strategy is still a human job. Deciding what to say, which battles to pick publicly, how a post fits into a wider narrative, that's judgement, not pattern-matching. Interviews matter too: getting an opinion out of someone who's never articulated it before takes a real conversation, not a form. And if you're a genuinely zero-time executive who can't spare twenty minutes to review a profile or approve a draft, you need a person managing the whole pipeline, not a tool.

The honest split is this: AI handles the voice-matching and the drafting speed. Humans handle the thinking that comes before either.

Where this gets interesting for ghostwriters themselves

At the end of the day, this isn't just a cost story for founders. Ghostwriters running multiple clients have the same voice-matching problem multiplied by however many people they write for. ScriptGrain's Studio plan, ten profiles for £99, lets one writer hold ten distinct, measured voices instead of ten sets of vague instincts. Plans start from £12 a month for individuals who just want their own voice held consistently.

Cheaper than a junior ghostwriter, and it doesn't guess at what you sound like. It checks.