A bot writes my SEO on
every podcast blog post.
A second bot waits for every new podcast blog post to go live, reads the body, picks the focus keyword, writes the meta description, fills the alt text and tags, and pushes everything back to WordPress. The SEOPress score lands in the green without me touching it, on every single post.
Every blog post my podcast bot ships, a second bot fills the SEO without me touching it.
Most creators publish a blog post and either skip the SEO step entirely or spend 30 to 40 minutes per post picking a focus keyword, writing a meta description, filling the tag list, and writing alt text for every image. I am not an SEO person, I know my limitations, and I used to pay an SEO freelancer to do this on my behalf. There is a simpler way. Let a small bot read every fresh post the moment it lands, generate every SEO field a human editor would fill, and stamp them back through the WordPress REST API. Here is who it is for, what goes wrong without it, how it works, and what you get back.
Who it’s for
Podcasters and YouTubers who already auto-publish podcast blog posts on WordPress (with SEOPress, Yoast, or Rank Math) and want every post to actually rank, without learning SEO or paying a freelancer per post. Also useful for anyone with a back catalog of old posts that never got the SEO fields filled. If your posts go live but the SEOPress score sits red, this is for you.
What goes wrong
Without the bot, the post ships but the focus keyword is empty, the meta description is whatever the theme auto-grabs from the first paragraph, the images have no alt text, and the tag list is either missing or copy-pasted from the last post. Google has nothing structured to read. The post sits on page 5 of the SERP and bleeds into oblivion. An unoptimised post is a post that does not rank, no matter how good the audio is.
How the machine works
The trigger fires the moment WordPress publishes a new podcast post. The bot pulls the post body through the REST API, runs four prompts in sequence (focus keyword and alternatives, SEO title and meta description, image alt text, final tag list), then PATCHes the post back with every SEO field filled in the right meta keys for your SEO plugin. The post writes itself. The SEO writes itself too.
What you get back
Every podcast blog post lands with a focus keyword, alternative keywords, an SEO title under 60 characters, a meta description under 160 characters, image alt text on every image, and a clean tag list. The SEOPress score sits green from the first save. Roughly $50 to $80 of consultant work saved per post, ~$7,500 to ~$12,000 a year, plus ~$2,250 to ~$4,500 in extra revenue from posts that now actually rank. The work an SEO freelancer would charge for, done invisibly.
I was paying someone to fill in SEO fields I should never have touched.
I am not an SEO person. I know my limitations. For the first year of the podcast, every blog post that the auto-post bot shipped landed on WordPress with no focus keyword, no meta description, no image alt text, and a copy-pasted tag list. The SEOPress dashboard sat red on every single post. So I did what anyone does. I paid an SEO freelancer to come in once a week, open every fresh post, and fill every field by hand. The bill was around $60 a post. Three posts a week. About $180 a week. About $9,000 a year for what was, in the end, structured copy-paste.
The cost was bigger than the money. The freelancer had a 48 to 72-hour turnaround, so the posts went live optimised three days late, by which point Google had already crawled the unoptimised version and decided what the page was about. The freelancer was also a bottleneck. If she was sick or on holiday, the backlog grew. The fields were filled, eventually, but the work was never silent and never on rails.
So I built a second bot to sit downstream of the auto-post bot. The moment a new podcast post lands on WordPress, this second bot fires. It pulls the full post body through the REST API. It runs the body through a focus-keyword prompt and gets back one focus keyword plus three alternatives with the search intent of each. It runs a second prompt to write the SEO title under 60 characters and the meta description under 160 characters ending on a full stop. It runs a third prompt to write image alt text for every image in the post. It runs a fourth prompt to build the final tag list. Then it PATCHes the post through the REST API and stamps every field into the SEOPress meta keys.
The total cost of running it is a few cents per post in API calls. The freelancer bill went to zero. The SEOPress score sits green on every fresh post from the first save, which means Google crawls the optimised version on its first pass instead of the unoptimised one. The backlog problem is gone, the turnaround time is one minute instead of three days, and I do not have to know the first thing about SEO because the bot does the part that was always structured anyway.
Proof point: the live WordPress this runs against is martinebongue.com itself, and the podcast that feeds it is Freedom by Choice on YouTube. The mechanics are also covered in the Growth Hacking series, so the saved-fees and ranking lift numbers on this page read as checkable.
Three moves that turn every fresh post into a fully optimised page on autopilot
What made this work was treating SEO as a structured fill-in step, not a creative one. The post is the creative output. The focus keyword, the meta description, the alt text and the tags are structured fields that follow rules. The framework hangs on three moves: read the post body the moment it lands, extract every SEO field a human editor would fill, then stamp the fields back through the REST API into the correct SEOPress meta keys. That is the whole engine. Done right, it removes ~30 to 40 minutes of SEO work per post, kills the ~$60 freelancer fee per post, and adds ~$10,000 to ~$15,000 in yearly value added.
Read the post body the moment it lands
The trigger fires on a WordPress webhook when the auto-post bot publishes a new podcast article. The downstream bot pulls the full post through the REST API, gets the title, the slug, the body content, the image list, and the existing categories. No screen scraping. No HTML parsing. Just structured JSON straight from the source. The bot now has every piece of input it needs to write the SEO fields, in the format the next step expects, before Google has even crawled the post for the first time.
Extract every SEO field a human editor would fill
Four prompts run in sequence. The first extracts one focus keyword and three alternative keywords with intent buckets, plus the suggested placements inside the body. The second writes the SEO title under 60 characters and the meta description under 160 characters as one full-stop sentence with the keyword once. The third writes image alt text for every image in the post, between 80 and 125 characters. The fourth picks 5 to 8 lowercase tags and rewrites the featured image alt text. Every field a freelancer would fill, written in one pass for cents.
Stamp the fields back through the REST API
The last move is the one that turns the AI output into a live SEO page. The bot maps every field to the correct SEOPress meta key (or Yoast, or Rank Math), builds a single PATCH payload, and sends one HTTPS call to the WordPress REST API. WordPress accepts the patch, the meta is saved, and the SEOPress dashboard score recalculates. The bot logs the score before and after so you can see the lift on every single post. No log in, no copy paste, no SEO plugin UI opened by hand.
Once those three moves are in place, SEO stops being a step you skip or pay someone to do. It becomes the silent layer that runs after every post goes live, on rails, on a schedule that matches the show.
Before the bot
- ~30 to 40 minutes of SEO fill-in per post, or ~$60 to a freelancer
- Posts went live red on SEOPress, Google crawled the unoptimised version first
- Three-day turnaround on the freelancer meant posts ranked late, if at all
- Tag list was copy-pasted from the last post, alt text was missing on every image
- The cost was real, the bottleneck was a person, the backlog grew every quarter
After the bot
- Zero manual SEO time per post, the fields write themselves the moment the post lands
- SEOPress score sits green on every fresh post from the first save
- One-minute turnaround instead of three days, Google crawls the optimised version first
- Focus keyword, meta description, alt text, and tags filled in one PATCH
- ~$10,000 to ~$15,000 a year of saved fees plus extra rankable revenue, every year
Prompt 1: analyse a blog post body and extract the focus keyword plus alternative keywords with intent buckets
The focus keyword is the single most load-bearing field on the page. Get it wrong and every other field is anchored to the wrong topic. Use this prompt to read the full body of a fresh blog post and return one focus keyword plus three alternative keywords, each tagged with its search intent.
Focus keyword and alternative keywords extractor
Act as an on-page SEO editor reviewing a freshly published WordPress blog post built from a podcast episode. I want you to read the post body and return the single focus keyword plus three alternative keywords for the post, each with the search intent bucket. Full post title: [paste] Full post body (HTML or plain text): [paste] Existing focus keywords already used on this site, comma separated: [paste, or write "none"] My niche in one sentence: [paste] Produce one focus keyword (2 to 5 words) the entire post is built around, that appears at least once in the title, at least once in the first 100 words of the body, and at least once in one H2. Then produce three alternative keywords for related search intents the post also serves. For each of the four keywords return the intent bucket: informational, navigational, transactional, or commercial. Do not pick a keyword that already appears in the existing focus keywords list. If the strongest candidate is taken, pick the next strongest and log the swap in one short line.
The output of this prompt is what gets PATCHed into the SEOPress focus keyword field and the related-keyword field. Lock it once for your site, the bot reuses it on every new post.
Prompt 2: write the SEO title and meta description for the post
The SEO title and meta description are what a human sees in the Google snippet. They are not ranking signals on their own, but they decide whether the click happens. Use this prompt to write both at the right length, with the keyword once, ending on a full stop.
SEO title and meta description writer
Act as the SEO snippet editor for a WordPress blog post. The post will be indexed by Google, and the SEO title and meta description are what a searcher will see in the SERP before they decide to click. Write both, to the exact specs below. Post title (the article title, not the SEO title): [paste] Focus keyword for this post: [paste from Prompt 1] One-sentence summary of what the post covers: [paste] Tone in three adjectives: [paste] Produce two things. One: an SEO title under 60 characters that opens with the focus keyword, reads naturally (not stuffed), and ends without a trailing brand suffix because the theme adds that automatically. Two: a meta description between 140 and 158 characters, ending on a full stop, written as one full sentence that includes the focus keyword exactly once, names the concrete benefit the reader gets from reading the post, and sounds like a human wrote it, not a checklist. No emoji. No all-caps. No question marks. No hyperbole.
This is the snippet the bot pastes straight into the SEOPress title field and the SEOPress description field. The character limits are baked into the prompt so the snippet always renders cleanly in the SERP.
The 3-minute overview of how this works
Before the build steps, watch this short overview. It’s the exact video from the Automations Made Easy page, and it walks through the mechanics behind machines like this one. 1,000+ students have used these mechanics to save two hours a day, with zero coding.
Want to learn the mechanics behind on-page SEO bots like this?
I teach the same mechanics that make this silent SEO layer run inside Automations Made Easy. 1,000+ students have used these mechanics to save two hours a day and stop paying freelancers for structured fill-in work the bot does for cents. No coding required.
Week one: three episodes shipped, six SEO scores in the green.
Most people quit a system like this in the first week because the lift feels too quiet. There is no big keyword breakthrough, no traffic surge, no obvious jump. The first week of the silent SEO layer is meant to feel small. The point is to confirm the bot fires when a new post lands, every SEO field is filled correctly, the PATCH lands on the right meta keys, and the SEOPress score moves from red to green on every single post.
Week one looks like this. The auto-post bot ships three podcast articles across the week. The SEO bot fires within a minute of each post going live. It reads the body, runs the four prompts, builds the PATCH, and stamps the meta. Three fresh posts move from red to green on SEOPress in the first save. The freelancer who used to charge $60 a post is no longer billing for these three. That is ~$180 of saved fees in week one, ~$720 in the first month, and around $9,000 across the year on the fee side alone. Quietly, in the background, while you record the next episode.
The point of the first week is not the traffic. The point is to prove the loop closes: post publishes, bot reads, four prompts run, PATCH lands, SEOPress score flips to green, the freelancer bill flips to zero.
From there the maths is simple. ~$60 of saved consultant work per post across ~150 posts a year is ~$9,000 a year in saved fees, plus the lift from posts that now rank. Each ranked post pulls roughly ~$15 to ~$30 in extra attributable revenue a year, which across the catalog adds another ~$2,250 to ~$4,500. Combined value added is ~$10,000 to ~$15,000 a year. Over five years, with SEO equity compounding across a growing back catalog, that stacks into ~$60,000 to ~$80,000+ of work that wrote itself.
All of this runs while you record, sleep, or take a week off. The bot does not care. The trigger reads the webhook, the analyser reads the body, the prompts write the fields, the PATCH stamps the meta, the score lands green. Your blog finally captures the value the platforms were quietly keeping, and you do not have to know the first thing about SEO to make it happen.
Prompt 3: write image alt text for every image in the post, keyword-aware
Image alt text is one of the highest-payoff SEO fields because almost no creator fills it properly, and Google Image Search reads it. Use this prompt to walk every image in the post body and write keyword-aware alt text for each one.
Per-image alt text writer
Act as an accessibility and SEO editor for a WordPress blog post built from a podcast episode. I will paste the post body which contains multiple images. For every image you find, write the alt text. The alt text has to work both for a screen reader and for Google Image Search. Focus keyword for the post: [paste from Prompt 1] Post topic in one sentence: [paste] Full post body with image URLs and any existing alt text: [paste] For each image, return the image URL (or filename), then the new alt text. The alt text must be between 80 and 125 characters, written as one descriptive sentence, never starting with the words "image of" or "picture of", and must include the focus keyword once on the single most-related image (the hero or first image). For all other images, include the focus keyword only if it accurately describes what is in the frame. Never invent details that are not visible. Output as a small table: image URL, new alt text.
The bot walks the table row by row and pushes each new alt value to the corresponding image meta in WordPress. Free, accessible, on-keyword by default.
Prompt 4: write the final tag list and the featured image alt text
Tags feed the related-posts widget and the SEOPress snippet. The featured image alt text becomes the share card on social and the OG image search engines use. Use this prompt to write both for every new podcast post.
Tag list and featured image alt text writer
Act as the SEO editor finishing the on-page setup for a WordPress blog post built from a podcast episode. The body has already been written, the focus keyword is locked. I need the final tag list and the featured image alt text in a format the bot can drop straight into WordPress. Focus keyword for the post: [paste from Prompt 1] Three alternative keywords for the post: [paste from Prompt 1] Existing tag list on the site, comma separated: [paste] Post title and one-sentence summary: [paste] Featured image filename and visual description: [paste] Produce two things. One: a tag list of 5 to 8 tags, lowercase, no duplicates, ordered from broadest to most specific, reusing existing tags from the site where they fit and only adding new tags when none of the existing ones map cleanly. Two: a featured image alt text between 80 and 125 characters, written as one full descriptive sentence that includes the focus keyword once and never starts with the words "image of". Return the tags on one line, the alt text on the next.
The bot writes the tags to the WordPress tags taxonomy and the featured image alt text to the attachment meta in a single PATCH. The post is now fully optimised before the first organic visitor lands on it.
The exact build, step by step
Wire the trigger that fires when WordPress publishes a podcast post
The trigger is the source of truth for the whole bot. Point a small workflow (n8n, Make, or Zapier all work) at the WordPress post-published webhook, filtered to the Podcast category only. Every time the auto-post bot publishes a new podcast article, the workflow fires once. The trigger carries the post ID forward to the next step and does not write anything yet. If you do not have a webhook, polling the REST API every five minutes works too. This is what makes the SEO layer event-driven instead of scheduled.
Pull the full post body through the WordPress REST API
When the trigger fires, the bot opens the REST API and asks for the full post. One call to /wp-json/wp/v2/posts/{id} returns the title, the slug, the rendered body, the image list with URLs, the existing categories and tags, and the current SEOPress meta. The bot now has every piece of input it needs to write the SEO fields, in clean JSON, without parsing HTML or scraping the rendered page. The REST API is the cleanest read source on a WordPress site, use it.
Run the analyser to extract the focus keyword and the alternatives
The first prompt (Prompt 1 above) runs against the post title and the full body. It returns one focus keyword the post should rank for, plus three alternative keywords with intent buckets (informational, commercial, transactional, navigational). The bot validates the focus keyword against your site’s existing keyword list so two posts do not compete for the same one. If the strongest candidate is already taken, the bot swaps to the next-best alternative and logs the swap. The output of this step is the anchor for every later field.
Write the SEO title, the meta description, and the tag list
The second and fourth prompts run next. The second writes the SEO title under 60 characters opening with the focus keyword, plus the meta description under 160 characters ending on a full stop with the focus keyword once. The fourth writes the final tag list of 5 to 8 lowercase tags, reusing existing site tags where they fit. The bot validates the character lengths automatically. If anything is over, the prompt re-runs with the over-count noted, until both fields fit. Every field is now ready to be stamped.
Generate image alt text for every image in the post
The third prompt (Prompt 3 above) walks the image list pulled in step two. For every image URL, it returns a single-sentence alt text between 80 and 125 characters, never starting with the words “image of”, and including the focus keyword on the hero image. For the rest, the focus keyword goes in only if it accurately describes the frame. The bot stores each new alt text against its image attachment ID so the PATCH in step six can update the right rows in the WordPress media meta table. Accessibility and SEO, in one pass.
PATCH the post through the REST API and log the SEO score before and after
The last step is a single HTTPS request. The bot builds one PATCH payload with the focus keyword, the SEO title, the meta description, the tag list, the featured image alt, and the per-image alt text, mapped to the SEOPress meta keys (or Yoast or Rank Math, depending on which one you use). It sends one PATCH to /wp-json/wp/v2/posts/{id} with an application password for auth. WordPress accepts the patch and the SEOPress score recalculates. The bot reads the score before and after and writes both into an Airtable row so you can see the lift on every single post.
Trigger fired
WordPress publish webhook fires the moment a new podcast post lands, the workflow runs once with the post ID.
Body pulled
One REST API call returns the title, the slug, the body, the image list, and the current SEOPress meta in clean JSON.
SEO fields written
Four prompts run in sequence, focus keyword and alternatives, SEO title and meta description, per-image alt text, then the final tag list and featured-image alt.
PATCH live
One PATCH stamps every field on the right SEOPress meta keys, the score recalculates from red to green, the before-and-after gets logged for the record.
Build this silent SEO layer inside the same playbook 1,000+ students use
Automations Made Easy teaches the mechanics behind bots like this one. Step by step, no code, plain English. Stop paying freelancers to fill structured SEO fields the bot fills for cents, on every post you publish.
The six months after I switched it on
Here is the shape of the first six months after I turned the silent SEO layer on for the podcast blog. The line is intentionally modest in the early months and steady from there, because that is how a quiet SEO bot actually behaves.
Monthly value added by the SEO bot (saved fees + extra search revenue)
Caption: Monthly value added by the SEO bot, combining saved consultant fees and extra attributable search revenue from posts that now rank. Three things matter on this chart. The line lands around $1,000+ a month and stays there. Every dollar of that line was paid to a freelancer before, or did not exist because the post never ranked. And every single one of those months happens without me touching SEOPress once.
What other students built
I teach the simple skills behind machines like this in Automations Made Easy. Students who built their own version sent back what changed in their first month.
“My SEO freelancer was charging me eighty dollars a post and I had a backlog of 60 posts she had never touched. The bot cleared the entire backlog in one overnight run for the cost of a sandwich.”
“I never understood what a focus keyword actually was. The bot picks one, fills the SEOPress fields, the score sits green, and my Google Search Console impressions doubled in the first month.”
“I used to dread the SEO step on every post. Now I do not even see it. The bot writes the meta, fills the alt text, picks the tags. The post is fully optimised before I have closed my recording software.”
“Pointed the bot at my back catalog of 142 old podcast posts that had no SEO setup. Three days later every single one was sitting green on SEOPress. I would have paid an agency twelve grand to do that by hand.”
What’s inside Automations Made Easy
AME isn’t a library of pre-built automations. Every business is slightly different. What’s reusable across all of them is the underlying mechanics: how to set up little machines that listen, write, and follow up while you sleep, and how to wire the pieces together without writing code.
The program walks you through six modules: The Right Tools (the cost-effective, no-code stack I actually use), Task Selection Mastery (which automations are worth building first), Design Secrets (mapping an automation before you build it), Zero to Hero (complete beginner to confident automator), Real-World Application (we build a full automation together, end to end), and Monetization Mastery (turn the skill into a side-business).
It also includes done-for-you templates you import in two clicks, over-the-shoulder training videos, and the same playbook 1,000+ students have used to save two hours a day. No coding required. If you can copy and paste, you can build this.
The silent SEO bot: common questions
Pulled from what readers and Automations Made Easy students ask most.
Will Google flag AI-written meta descriptions?
No. Google has been explicit that the issue is low-value content at scale, not the use of AI to assist with on-page metadata. Meta descriptions are not used as a ranking signal directly, they are used as the snippet under the blue link in search results. What matters is that the description is one clear sentence that includes the main keyword once, sits under 160 characters, and accurately previews the post. The bot writes exactly that, every time. The risk lives in dumping AI-generated body content with no editing, not in letting a bot fill in the snippet field.
Does this work with SEOPress, Yoast, and Rank Math?
Yes for all three. The WordPress REST API exposes the SEO fields each plugin stores in post meta. SEOPress fields use the prefix _seopress_, Yoast uses _yoast_wpseo_, Rank Math uses rank_math_. Once you point the bot at the correct meta keys for your plugin, the same PATCH request fills the focus keyword, the SEO title, the meta description, the tag list, and the image alt text in the right slots. The brain of the bot does not change, only the meta-key map does. If you switch plugins later, you remap once and keep moving.
How does the bot pick the focus keyword?
The bot reads the full post body and the title, then runs a structured prompt that asks for one focus keyword and three alternative keywords. The focus keyword has to appear at least once in the title, in the first 100 words, and in one H2. The alternative keywords cover related search intents the post also serves. The bot then validates the focus keyword against an existing-content check on your site so two posts do not compete for the same keyword. If a collision is found, it picks the next strongest alternative and logs the swap.
What if my post is already published, can it backfill?
Yes, and this is one of the highest payoff uses. Point the bot at any WordPress category, tag, or date range, and it walks the existing posts one by one, reads each post body, fills the SEO fields the same way it would on a new post, and PATCHes the post back through the REST API. I ran this once on a backlog of around 90 old podcast posts that had no SEO setup, and the SEOPress score went from red to green across the catalog in a single overnight pass. The bot does not rewrite the body, it only fills the metadata.
How much does this realistically save in SEO consultant fees per year?
Conservatively, around $50 to $80 of work per post if you priced it against a freelance SEO editor filling the same fields by hand. Three episodes a week, around 150 posts a year, lands at roughly $7,500 to $12,000 a year in saved consultant fees. On top of that, posts that now actually rank pull around $15 to $30 a year in extra attributable revenue each, which adds another $2,250 to $4,500. Combined, the bot is worth around $10,000 to $15,000 a year in value added, every year, while it sits there running for cents in API calls.
More Money Makers like this one
Built from the same handful of mechanics. Each one captures, converts, or earns money that would have walked out the door.
Run this silent SEO layer yourself, or learn the mechanics inside Automations Made Easy.
If you want to learn the mechanics behind on-page SEO bots like this and build your own at home, Automations Made Easy is the playbook. Step by step, no code, plain English. If you want to talk through which SEO plugin, meta-key map, and PATCH flow would fit your specific WordPress setup first, I take a small number of consulting clients each month.
€497 one-time · Lifetime access · 1,000+ students
