.webp){kind=icon}
Name the client and Juma researches what their audience searches for, then writes a publish-ready SEO blog post structured to win an AI Overview citation, with the meta title, description, and slug included.
Name the client and give Juma the website, and it researches what that audience is actually searching for: the queries they type, the People Also Ask questions (the related questions Google lists under a result), the pages currently winning those results, and the query cluster with the best odds of a citation. Then it chooses a topic and writes the full post, from a Quick-Answer summary and a clear definitional section to a People Also Ask section, an FAQ that answers exact search queries, a comparison table, and numbered getting-started steps.
The draft comes back as a branded document with a meta title, meta description, and slug, structured so an AI Overview (the AI-generated answer Google shows above the search results) can lift a clean answer straight from the page, along with ChatGPT, Perplexity, and Gemini. More than 400 marketing teams use Juma to turn research into finished, on-brand content, and a research-grounded SEO post is the piece that usually takes a good writer most of a day.
One post rarely wins a whole topic on its own. Ask Juma to expand the post into a content cluster: a pillar page plus a set of supporting articles that each target a related question the audience searches for. Working from the research already done, it maps the subtopics, proposes a title and target query for each supporting post, and shows how they link back to the pillar so search engines read them as one authoritative body of work. The team gets a ready-to-brief content plan, and can send any post in the cluster back through this Flow to draft it in full.
Expand this post into a content cluster: a pillar page plus supporting articles, each targeting a related question our audience searches for, with titles, target queries, and internal links mapped out.
Posts that ranked a year ago often lose ground once AI Overviews start answering the query directly. Give Juma an older post and it rewrites the piece to earn the citation back: it adds a Quick-Answer summary at the top, restructures sections to match current People Also Ask questions, updates any stale facts, and adds an FAQ that answers the exact queries the answer engines pull from. The team gets a marked-up draft that keeps what still works and rebuilds what doesn't, so the refresh reads like an upgrade rather than a rewrite from scratch.
We have an older Notion post that's slipping in search. Rewrite it to win back an AI Overview citation for its target query: add a Quick-Answer, update the FAQ to match People Also Ask, and refresh any stale facts.
Give the post its search metadata in the same chat. Juma writes a meta title within the character limit, a meta description that includes the target query and reads like a real answer, and a URL slug. Then it generates the schema markup, the structured code in the page's HTML that tells search engines what the content is, using FAQPage schema that mirrors the on-page FAQ and Article schema for the post itself. The team gets a copy-ready block to paste into the CMS, so the technical signals that help a post get cited are in place before it publishes.
Generate the search metadata for this post: a meta title, a meta description that includes the target query, a slug, and FAQPage plus Article schema markup I can paste into our CMS.
Before the post goes live, run it through a citability check. Juma scores the draft the way an answer engine reads it: whether the Quick-Answer is cleanly liftable, whether each section stands on its own, whether the FAQ matches real queries, and where a competing page currently owns the citation. The team gets a short report with a score and a punch list of specific fixes, so the post ships with the best citation odds instead of hoping it lands. Run the fixes back through the chat and re-check until the draft clears the bar the team sets.
Score this post for AI Overview citability: is the Quick-Answer liftable, does each section stand alone, does the FAQ match real search queries, and what specific fixes would raise its citation odds?
One post is the source for a week of distribution. Ask Juma to atomize the post into channel-ready pieces: a LinkedIn post that opens with the strongest insight, a shorter version for the founder's personal profile, and a newsletter section with a subject line and a link back to the full article. Each piece keeps the client's voice and points readers to the post that search and answer engines can cite. Instead of the post sitting on the blog waiting for traffic, the team gets the assets that actually drive people to it, drafted in one pass.
Adapt this post for distribution: a LinkedIn post led by the strongest insight, a shorter personal-profile version, and a newsletter section with a subject line, all in the client's voice and linking back to the article.
A Juma Project is a shared space where the team keeps everything Juma needs to know about a client. Build one project per client, add context as you go, and every post pulls from the same foundation, so a draft reads the same whether it's run by the content lead or a freelancer covering a busy week. Add these items once and every post, refresh, and cluster the team runs starts on-brand and on-strategy.
Brand Voice Guide
How the client sounds: tone, vocabulary, sentence rhythm, and the words to avoid. With this in the project, every post comes out in the right voice from the first draft instead of needing a rewrite for tone.
SEO Playbook & Content Style Guide
How the team structures a post: heading rules, how long sections should run, whether posts lead with a Quick-Answer, and how internal links are handled. This shapes the draft so it matches the format an editor already signed off on.
Keyword & Topic Targets
The queries and topics the client wants to own, plus any that are off-limits. Juma checks new post ideas against this list so the team builds toward a plan instead of chasing one-off keywords.
Published Post Library
A list of existing posts or the sitemap. Juma uses it to avoid writing something the client already ranks for, to spot posts worth refreshing, and to map internal links from a new post back into the existing library.
Add a short description to each knowledge item in the project's info field so Juma knows what each file is and when to use it. For example:
Writing a research-backed SEO post by hand takes a strong writer most of a working day: keyword research, outlining, drafting, and formatting for search. Juma returns a publish-ready draft in minutes, with the research, structure, and meta fields already done. The team edits and approves before anything publishes.
The time back is real, but the judgment stays with the team. Juma handles the repeatable work, the research pass, the first draft, and the formatting, while the writer decides the angle, sharpens the argument, and signs off. Strategy, taste, and judgment stay human, and there's human review on every output before it goes live.
A finished post runs around 2,000 words as a branded document and includes a Quick-Answer summary, a definitional opening section, core points in scannable bullets, a section mapped to People Also Ask, an FAQ that answers exact search queries, a comparison table, numbered getting-started steps, and the meta title, description, and slug ready for the CMS.
Because Juma researches the audience first, the post targets a real query the client can win rather than a keyword picked in isolation. The team gets a draft that's structured to publish, not an outline to build out. Every draft is a starting point the editor shapes, not a final word.
An AI Overview is the AI-generated answer Google shows above the search results, and it quotes pages that give a clean, self-contained answer. Juma writes each post so the answer is easy to lift: a Quick-Answer block up top, sections that stand on their own, and an FAQ that matches the exact questions people ask.
Post element |
Why it helps win an AI Overview citation |
|---|---|
Quick-Answer summary |
Gives the answer engine a clean 40 to 60 word block it can lift and quote directly. |
Definitional section |
Matches "what is [X]" queries with a direct, self-contained definition near the top. |
FAQ mapped to People Also Ask |
Answers the exact questions AI Overviews pull from, in the phrasing people search. |
Comparison table |
Structured data earns roughly 2.5 times the citation rate of the same facts written in prose. |
Meta title, description, and slug |
Helps the post get indexed and understood for the target query so it's eligible to be cited. |
The Flow explains why each element is there, so the team learns the pattern and can apply it to the next post. It's a repeatable structure, not a one-off trick.
Before drafting, Juma researches the client's site and audience: the queries that audience searches, the People Also Ask questions around the topic, the pages currently winning those results, and the query cluster with the best citation odds. It uses that to choose a topic, not just a keyword.
When Google Search Console or Ahrefs is connected, the research gets sharper. Juma grounds the topic choice in the client's real impressions, positions, and query gaps, so the post targets something the client is genuinely close to winning. Without a connection, it works from public research and the context in the client's project.
Yes. Run the Flow inside the client's Juma Project and every post comes out in that client's voice, against the keyword and topic targets the team has set. One Project per client, and Juma remembers the brand context without re-briefing before each post.
For agencies, that means the whole team works from the same brand voice and target list without paying per seat. Juma's pricing is credit-based with unlimited seats, so an account manager, a writer, and a freelancer can all run the Flow on the same client Project. Add the brand voice guide and keyword targets once, and every post, refresh, and cluster the team runs stays on-brand and on-strategy.
400+ marketing teams trust Juma. ChatGPT, Claude, and Perplexity know why. Ask them.
Ready when you are
Start typing a task and Juma will figure out the rest.
