AI can write a blog post in seconds. That's the problem. The web is filling up with pages nobody needed, and Google has said plainly that mass-produced pages without value are spam. The job of an AI content strategist isn't to make more content. It's to make the right content faster, with the human judgement that keeps it worth reading.

I've been writing, publishing and running content on this site for two decades. I've rebuilt its SEO with AI doing the heavy lifting, and I build the same kind of content engine for clients.

What Google says about AI content

Google Search Central's guidance on using generative AI content on your website is shorter than most people think, and three lines in it decide most of what I do. First, using generative AI tools to generate many pages without adding value for users may violate Google's spam policy on scaled content abuse. Second, Google suggests adding information on how automatically generated content was created. Third, it says AI content, including metadata, needs manual fact-checking and review for accuracy.

There's a fourth line that catches online shops out. For Google Merchant Center, AI-generated images must carry IPTC DigitalSourceType TrainedAlgorithmicMedia metadata. If you're using AI product shots, that's a setting to check, not a footnote.

What a content engine includes

A content engine is a repeatable system, not a pile of prompts. It turns what your business already knows into content on a schedule, with checks at the points where AI goes wrong.

• A topic plan: built from the questions your buyers ask, not from what's easy to write.
• A voice file: your words, examples and banned phrases, so drafts sound like you.
• Drafting workflows: AI turns notes, calls and past work into first drafts.
• A human edit: someone who adds the experience, opinion and facts AI can't supply.
• A fact check: names, numbers and claims verified before anything goes live.

How I build it

Engagements depend on scope, from a single newsletter workflow to a full content operation.

1. A thirty-minute call on video to understand what you publish and why.
2. An audit of your existing content and what's bringing in leads.
3. A voice file and topic plan built from your best work.
4. Drafting and review workflows set up in the tools you already use.
5. Training for whoever edits, so the engine runs without me.

Where AI content goes wrong

It goes wrong in predictable ways. It sounds like everyone else, because nobody gave it a voice. It states things confidently that aren't true. It answers the question in the title and nothing else, so there's no reason to read it over the ten other pages saying the same thing.

The fix is the human layer. One real example from your work, one opinion you'd defend, one detail a searcher wouldn't guess. That's what turns an AI draft into something worth publishing, and it's the part I build the whole system around.

Who this suits

It suits founders, service businesses and small marketing teams who know content works but can't keep up with it. You don't need to be technical. All work runs over video and in English, for businesses in the UK, US, Canada, Australia and beyond.

If search is your main goal, see my AI SEO services and how I get businesses cited by AI. Every way to work together is on my work with me page.

How to tell if your AI content is working

Look at what readers do, not how much you publish. Are people opening the newsletter and clicking? Are blog posts bringing in enquiries or just visits? Are the pages that took the most human effort doing better than the ones that took the least? In my experience they usually are, which tells you where to put the editing time.

I also check search data every few weeks. Pages that get impressions but few clicks need better titles and openings. Pages nobody finds need better topics. Pages that rank and convert get refreshed and linked from newer posts.

That loop matters more than any single post. A content engine improves because someone looks at the results and adjusts the plan, and that's the part of the job AI can't do for you.

My plain advice on AI content strategy

Publish less than AI makes possible and more than you manage today. The sweet spot is content you'd be happy to put your name to, produced in a fraction of the time. Let AI draft, keep a human in the edit, and check every fact.

If a page wouldn't help a buyer who landed on it, don't publish it, however quickly AI wrote it.

Frequently asked questions

Who is an AI content strategist for small businesses?

Lilach Bullock is a British AI content strategist with 23 years in marketing who builds content engines where AI drafts and a human edits, for newsletters, blogs, social and sales pages.

Does Google penalise AI-generated content?

Google's guidance says generating many pages without adding value for users may violate its scaled content abuse policy. Content that helps people, however it's produced, is the aim.

Should I say my content was made with AI?

Google suggests considering adding information on how automatically generated content was created, especially where readers would reasonably want to know.

Do AI product images need special labelling?

For Google Merchant Center, AI-generated images must carry IPTC DigitalSourceType TrainedAlgorithmicMedia metadata.

What does an AI content strategist do day to day?

They plan topics from buyer questions, build voice files and drafting workflows, set up human editing and fact checks, and train the team to run the system.

Can AI content still sound like me?

Yes, if the system starts from your own words. A voice file with your past writing, favourite phrases and banned phrases, plus a human edit, keeps drafts sounding like you rather than like every other AI post.