Every content team eventually hits the same wall: the calendar has more open slots than the writers have hours. AutoSchedulePost’s AI Content Generation tools exist to close that gap without turning your blog into obvious, interchangeable AI filler. This guide is the map — how the three generation modes work, what happens to your draft the moment you write it, how tone settings keep everything on-brand, and where each companion article in this series picks up the detail. If you’re new to the generation side of the product, start here and branch out.
Three Ways In: Generate, Paste-Generate, and Search-Scrape
Most AI writing tools give you one box: type a topic, get an article. AutoSchedulePost gives you three distinct entry points, because “I need content” means different things depending on what you’re starting with.
Generate from a keyword
This is the classic flow — you hand the tool a seed keyword or topic, pick a tone, and it produces a full article built around that keyword. It’s the right choice when you’re working from a keyword list, a content calendar slot, or a gap identified by keyword research and you don’t yet have any source material of your own.
Paste-Generate from raw notes
Sometimes you already have the substance — a call transcript, a bullet-point outline, a half-finished draft, a customer email thread full of useful phrasing — but no time to shape it into a publishable article. Paste-Generate takes that raw, unstructured input and turns it into a coherent piece, preserving the facts and ideas you supplied instead of inventing new ones from a bare keyword. Our companion piece on turning raw notes into a finished article walks through this mode in depth, and how Paste-Generate handles messy, unstructured input covers what happens when your notes are genuinely disorganized.
Search-Scrape from live search results
The third mode starts from what’s already ranking. You give it a keyword, it pulls back live search results, and it works from that structure — headings, the topics competitors cover, the shape of what’s already answering the query — to build an original draft informed by the current search landscape rather than written in isolation. This is covered in detail in Search-Scrape workflows and how to turn a scraped competitor page into an original draft, including the ethics of building from other people’s rankings, which we address directly in scraping ethics.
Each mode writes to its own separate space. A draft you’re building from a keyword doesn’t overwrite or get confused with a draft you’re building from pasted notes or from a scrape — you can have work in progress in all three at once, and each remembers its own state independently. That separation matters more than it sounds like it should once you’re juggling several pieces of content at different stages, a topic covered in managing Generate, Paste, and Scrape separately.
The Problem Draft Persistence Actually Solves
Anyone who has used a browser-based AI writer has lived through this: you generate a solid draft, get pulled into something else, come back twenty minutes later — or the next morning after logging out — and the tool has forgotten everything. You’re staring at an empty box again, and the only options are to regenerate (burning time and possibly landing on a worse result) or dig through browser history hoping a tab survived.
AutoSchedulePost closes that gap by saving the generated draft the moment generation succeeds, not just in the browser session but against your account. Navigate away, close the tab, log out entirely — when you come back to that generation screen, your draft is sitting there waiting, exactly as it was. This isn’t a browser cache trick that breaks the moment you clear cookies; it survives logout and login because it’s tied to your account rather than your session. We go deep on why this exists and what it fixes in why your generated draft should never disappear after logout, and on the team angle in why draft persistence matters for teams working across sessions.
Each of the three modes keeps its own persisted draft, so switching from your Generate screen to your Paste-Generate screen doesn’t clobber either one — you can genuinely have three separate works-in-progress alive simultaneously, each restored independently the next time you land on that screen.
Start Fresh vs. Clear: Two Buttons That Do Different Things
Once persistence entered the picture, a new question came up immediately: what if you don’t want the old draft back? That’s where the distinction between clearing your current view and genuinely starting fresh matters. “Start Fresh” doesn’t just blank the text box in front of you — it discards the saved draft itself, so the next time you open that screen you get a clean slate instead of your old draft reappearing. A plain “Clear,” by contrast, only touches what you’re looking at right now.
The Search-Scrape mode adds one more wrinkle worth knowing: starting fresh there discards the generated article but keeps the pages you already scraped, since re-scraping the same search results a second time wastes a step for no benefit. The full mechanics, including why this distinction exists and when to reach for each one, are in why “Clear” and “Start Fresh” are different actions and how to recover a draft after an accidental navigation away.
Tone: Set It Once, Apply It Everywhere
Every account has a default tone — Informative, Persuasive, Conversational, and other options sit in your settings — and generation honors it automatically rather than making you pick a tone on every single article. Set it once for your brand voice and every subsequent generation, across all three modes, starts from that baseline. You can still override it per piece when a specific article calls for something different, but you’re never stuck re-selecting your house style every time you open the generator.
This same tone default extends into AutoSchedulePost’s pillar and cluster content tools, so a full topic cluster generated in bulk carries the same voice as a single article you generate by hand — no silent fallback to a generic default buried in the background. The full breakdown of tone options and when each one fits is in tone settings explained, and setting a default tone for every piece of AI-generated content covers the setup itself.
What Generation Actually Produces
A generated draft isn’t just body copy. Depending on your settings, generation can also produce a hero image for the top of the article and embed images under section headings, sized responsively so they never blow out the layout on mobile or shrink to nothing on desktop. If you use structured data on your site, generated content can carry through to schema markup for article type and FAQ sections — see from scrape to schema for how that connects. And if your workflow includes FAQ blocks, generating FAQ sections that actually answer real questions covers how to get those right rather than generic.
Word count is another lever worth understanding early — longer targets change pacing and depth, and how word count targets shape AI-generated article quality covers the tradeoffs.
From One Article to a Full Pipeline
Generation doesn’t have to be a one-off action you repeat manually for every post. It’s the engine underneath AutoSchedulePost’s larger content pipeline — feeding pillar pages, topic clusters, and scheduled batches. From keyword to published post: the generation pipeline end to end traces that full path, and batch generating a week of content in a single session covers how to use generation at volume without every article sounding the same — a real risk covered honestly in generating content in bulk without sounding repetitive.
What This Series Covers
The articles that sit alongside this one go deep on individual pieces of the system: how to write better seed keywords, how to brief the AI the way you’d brief a freelance writer, how to layer human editing into a generation workflow, what guardrails the AI won’t cross, and how to build a review checklist before anything moves on to Auto Scheduler. None of it requires you to understand PHP, databases, or how any of this is built under the hood — the point of this pillar is the opposite. You should be able to open Generate, Paste-Generate, or Search-Scrape, trust that your tone is applied and your work won’t vanish, and spend your remaining attention on making the draft actually good.
Where to Start
If you’re choosing a first move: pick Generate if you’re starting from a keyword with nothing else to go on, Paste-Generate if you already have notes or a transcript worth preserving, and Search-Scrape if you want your draft informed by what’s currently ranking. Set your default tone once in settings before you generate anything, so every draft — from your first test article to your five-hundredth — starts from the voice you actually want. Everything else in this pillar builds from that starting point.
