For agencies that grow D2C brands
How do you write SEO articles and schema that search and AI answers pick up?
SEO articles and schema that search and AI answers pick up start from one real buyer question, answer it in the first lines, back it with steps or a checklist, and carry markup naming the author, the dates and what the page covers. Check each draft against the brand's thesis, and publish only after your team approves.
344k impressions in Google's AI results in three months · Moolwan · led by two ai-dm-written how-to guidesGoogle Search Console · Generative AI features · 25 Jun–23 Sep 2026 · real data
What you'll see in 10 minutes
- Your first brand's report, built from its domain alone.
- The questions buyers ask in the brand's category, with the ones rivals already answer.
- A first article brief with its outline and schema, ready for your team to review.
Why do so many brand articles never get found?
Month three of the content retainer. A dozen blog posts are live, each a long piece on a keyword someone picked from a tool. Search Console shows a trickle of impressions across all of them, and the brand asks what the retainer is buying.
Articles written for keywords rarely answer anyone. The answer sits under a long introduction, the page has no markup, and nothing in it is something only this brand could say. Search engines and AI answers pass it over, and the retainer starts to look like busywork. These are the differences that decide it.
| Part of the article | Gets picked up | Gets skipped |
|---|---|---|
| The question | One real buyer question per page | A keyword stretched into a broad topic |
| The answer | In the first lines, then steps or a checklist | After a long introduction |
| Proof | Facts only the brand can state, with sources | Claims any brand could make |
| Structure | Headings that are the next questions, tables and lists | Long unbroken text |
| Markup | Article schema with author and dates; product schema on product pages | None, or markup the page doesn't support |
| Byline | A named author with a short bio | No author, or "admin" |
How do you write an article that gets picked up?
Question first, markup last, approval always. The order keeps the article honest: it answers something real, proves it, and describes itself accurately. Work through these six steps for every article the brand publishes.
Start from a real question
Pick one question buyers actually ask, from People Also Ask, autocomplete or the brand's own customer messages. One question per article keeps the answer sharp.
Answer it in the first lines
Give the direct answer in the opening 40 to 60 words, then the detail. Readers, search engines and AI answers all look for the answer first.
Back it with steps, a checklist or a table
Turn the how-to into numbered steps, a checklist or a comparison table. Structured parts are easy to read and easy to quote.
Add proof only the brand can give
Use facts the brand can stand behind: the maker, the material, its own experience or data. Link the source of any figure the brand didn't produce.
Mark it up honestly
Add Article schema with the headline, author, publish and update dates and an image, plus product, local business or FAQ markup only where the page really shows that content.
Check it against the thesis, then approve
Read the draft against the brand's thesis for claims it never makes, then your team approves it before it's published.
Google says Article markup has no required properties and recommends the author, publish and update dates, headline and image (Google Search Central, Article structured data). It stopped showing FAQ rich results in May 2026 (Google Search Central, FAQ rich results). To pick questions worth writing, find the brand's keyword and content gaps first, and keep every draft inside the brand's thesis.
What schema should a brand's article carry?
Generate it here, for any article. Fill in the article's details and, if the page shows questions and answers, add them. The generator writes the JSON-LD, flags anything missing or inconsistent, and gives you the block to paste into the page.
Generate Article and FAQ schema for a page
Every field mirrors something the page itself shows. Leave a field empty rather than guess.
- Headlineokmatches the article's title
- Author nameoka named person, not "admin"
- Author page URLrecommendedrecommended: link to a page about the author
- Published and updated datesokvalid, and the update is on or after publishing
- Imageoka full https address
- Page URLoka full https address
- Questions and answersok2 complete pairs, each shown on the page
<script type="application/ld+json">
{
"@context": "https://schema.org",
"@graph": [
{
"@type": "Article",
"headline": "How to brew pour-over coffee at home",
"description": "A seven-step pour-over method for home brewers, with the grind, ratio and water temperature for each roast.",
"image": [
"https://terrabrew.example/images/pour-over.jpg"
],
"datePublished": "2026-10-12",
"dateModified": "2026-10-16",
"author": {
"@type": "Person",
"name": "Asha Rao"
},
"publisher": {
"@type": "Organization",
"name": "Terra Brew Coffee"
},
"mainEntityOfPage": "https://terrabrew.example/guides/pour-over-at-home"
},
{
"@type": "FAQPage",
"mainEntity": [
{
"@type": "Question",
"name": "What grind should I use for pour-over?",
"acceptedAnswer": {
"@type": "Answer",
"text": "Medium-fine, like table salt. Go a little coarser if the brew takes longer than three and a half minutes."
}
},
{
"@type": "Question",
"name": "What coffee-to-water ratio works best?",
"acceptedAnswer": {
"@type": "Answer",
"text": "Start at 1:16, such as 15 g of coffee to 240 g of water, then adjust to taste."
}
}
]
}
]
}
</script>Free, no login. The starting article is an illustrative example. Test the result in Google's Rich Results Test before publishing.
(login required) · ai-dm writes the schema with every article, for every brand you manage.
How does ai-dm write a brand's articles and schema?
From the question to a page ready to publish. ai-dm finds the questions buyers ask in the brand's category, writes the article from the brand's thesis with the answer up front, generates the images, the meta title and description, and the schema, and checks every line against what the brand never says. The finished article waits for your team's approval, ready to publish.
1 article ready · 1 line flagged · waiting for your team's approval
In this illustrative example, ai-dm finds a buyer question for Terra Brew Coffee, writes a draft with the answer in its first 52 words and a seven-step checklist, generates three images and valid Article schema, and flags one line that calls the beans the cheapest, which the brand's thesis rules out. The article then waits for the team's approval.
After it's published, ai-dm watches what it earns. Impressions in search and in AI answers show which articles work, so the next brief builds on them; you can track the brand's AI search visibility from the same workspace. Nothing is published without your team's approval, and you can check how brand data is protected.
On our own brand. Moolwan's pages were shown 344,000 times in Google's AI results between 25 June and 23 September 2026 (Search Console · Generative AI features). The two leading pages are ai-dm-written how-to guides: TV placement by Vastu with 8,479 impressions and feng shui colours with 7,853. The Vastu guide appears in the same results as guides from Nerolac, NoBroker and HouseGyan. On the product side, 1,400 products are ready for Google's product rich results, with 24 of 1,424 items flagged to fix (Search Console · Product snippets · 26 Sep 2026). You can watch ai-dm run Moolwan, with the source of every number.
Related pages
Questions agencies ask about SEO articles and schema
What makes an SEO article likely to show up in AI answers?
There's no guarantee, and Google doesn't say why it picks a page. The pages that do show up tend to answer one real question early and clearly, back the answer with steps or a checklist, carry proof only the brand can give, and use schema that matches the page. Moolwan's two leading guides follow that pattern.
Does schema markup improve a page's rankings?
Schema doesn't guarantee a ranking or a rich result. It tells search engines and other systems what the page is, who wrote it and when, so they can read it correctly. The markup must match what the page shows; markup that describes content the page doesn't have does more harm than good.
Is FAQ schema still worth adding?
Google stopped showing FAQ rich results in its search results in May 2026, so FAQ markup won't change how a page looks in Google Search. Add FAQPage markup only where the page really shows those questions and answers, as part of describing the page accurately. The generator on this page makes it optional.
How long should an SEO article be?
As long as the question needs and no longer. Put the answer first, then the steps, proof and related questions. A 600-word page that answers one question completely usually serves readers better than 2,000 words that circle a broad keyword.
Does ai-dm publish articles to the brand's website?
ai-dm writes the article, its images, the meta title and description, and the schema, ready to publish. Nothing is published until someone on your team approves it, and every approved article is recorded in the decision ledger.
Can ai-dm write product and local business schema too?
Yes. ai-dm generates JSON-LD for articles, products, local businesses and FAQs, and checks that each block matches what the page shows. Product markup on product pages is how a catalogue becomes eligible for Google's product rich results.
What does article planning in ai-dm cost?
There's no plan to pick. Sign up, pay as you go, and start from one brand's domain. Billed in USD and INR. Per-brand plans or pay as you go. A brand's data is deletable on request.
No plan to pick. Billed in USD and INR. A brand's data is deletable on request.