How One Brief Becomes Eight Native Posts

A brief is not a post. In SocialPower, one brief plus your Brand Brain produces eight separate posts — each rendered at that platform's native size and rewritten to that platform's character limit, hashtag norms and CTA style. Nothing is cropped down from a master image. Nothing is truncated from a master caption. Each post is generated for its own feed.

That distinction is the whole point of this article. Most scheduling tools take one asset and one caption and copy them eight times. That is distribution, not publishing. Below is what actually breaks when you do it, the real numbers behind the breakage, and the three passes SocialPower runs instead.

The default workflow in most small marketing teams looks like this. Someone designs a 1080 × 1080 square in Canva. Someone writes a 900-character caption with 22 hashtags. Both get pasted into a scheduler and pushed to every connected account. The post looks correct in exactly one place — usually Instagram, because that is where it was designed.

Everywhere else, one of four things goes wrong:

• **The image is cropped or padded.** A square pushed to TikTok gets letterboxed with dead bars top and bottom. A 16:9 banner pushed to Instagram sits as a thin strip in a 4:5 slot and gives up half the screen area you were entitled to.
• **The caption is truncated.** A 900-character caption is fine on Instagram and impossible on X. On Facebook and LinkedIn it does not get rejected — it gets folded behind "See more" after roughly the first 125 to 140 characters, which means your hook is whatever words happened to land there.
• **The hashtags read wrong.** Fourteen hashtags is normal on Instagram and reads as spam on LinkedIn. Pinterest wants keywords in the description, not a hashtag block. X rarely rewards more than two.
• **The CTA points at nothing.** "Link in bio" is meaningless on LinkedIn, where you can put the link in the post. "Swipe up" is meaningless everywhere it is not a story. "Comment below" is a weak ask on Pinterest, where the action is a save.

None of these is catastrophic on its own. Together they are the reason a brand's LinkedIn presence looks like an Instagram account someone forgot about.

Here are the numbers SocialPower renders and writes against. Platforms change these without much notice, so treat the table as the shape of the problem rather than a permanent contract — the platform-by-platform breakdown carries the values the product currently enforces.

Two rows deserve a flag. TikTok and YouTube Shorts are video-first surfaces, and SocialPower does not render video yet — the video agent is on the roadmap and labelled SOON. For those two, the system generates the copy, the hashtags and the still image, and you supply the video file. Everything in this article about text and scheduling still applies. Nothing about it should be read as "SocialPower makes your Reels."

The reason the eight outputs are different from each other is that they are produced in three distinct passes, and only the first one is about your brand. The full pipeline is documented in how it works; this is the short version.

Pass one: the brand conditions the draft

Before a word is generated, your Brand Brain is loaded as structure, not as chat history. It carries voice, audience, reading level, target sentence length, always-say facts, banned words, emoji policy, hashtag ceiling, CTA style and claims policy. A generation that violates the banned-word list or the claims policy is wrong at the point of writing, not corrected afterwards by you in a review queue.

This is also the layer that keeps agencies out of trouble. Each brand carries its own Brand Brain and its own connected accounts. Nothing crosses between them.

Pass two: the platform reshapes it

Now the brief splits. For each target platform the system generates against that platform's constraints as hard inputs — character ceiling, hashtag ceiling, whether a link belongs in the body, whether the first 125 characters have to carry the hook alone, and the native canvas dimensions for the image. The X version is not the Instagram caption with the middle removed. It is written to 280 characters as its own object.

Hashtags are scored on competition rather than pulled from a list, so the ten tags that make sense for a 4,000-follower D2C skincare brand are not the ten that make sense for a national FMCG account chasing the same keyword.

Pass three: the calendar places it

Eight finished posts still need eight slots. Posts are scheduled per platform and per timezone, which matters more in India than the feature list suggests — a LinkedIn post that lands at 9:40 am IST on a Tuesday and an Instagram post that lands at 9:40 pm are the same brief, correctly placed twice.

The reshaping is deliberate, so it helps to be exact about what does not change. Across all eight outputs, these hold:

1. 01The core claim or offer in the brief.
2. 02Every fact on your always-say list.
3. 03The claims policy — if you cannot say "clinically proven", no platform version says it.
4. 04The banned-word list and emoji policy.
5. 05The product, price and link, where they appear at all.

What changes is length, structure, hashtag count, CTA phrasing and image dimensions. Meaning does not change. If a platform version says something the brief did not, that is a defect, not a feature.

Autonomous mode runs this loop on a schedule, with two limits worth understanding precisely.

The **approval gate is on by default**. The worker generates and queues, and nothing publishes until a person releases it. You can turn the gate off. Most teams running client accounts should not.

The **daily post cap is enforced by the worker**. It counts today's AI-generated posts and stops at the ceiling for your plan — 1 a day on Starter, 5 on Pro, 15 on Business. The monthly spend cap is a different thing: it is stored and displayed, but the worker does not enforce it. Treat it as a dashboard number, not a circuit breaker, and keep an eye on your provider console. Plan limits and what each tier includes sit on the pricing page, in rupees, before 18% GST.

Also not built, and not implied anywhere in the product: comment moderation, inbox management and paid ads. Trend detection, SEO, analytics readback and competitor tracking are labelled SOON in the interface because they are on the roadmap, not shipping.

The quality of eight platform posts is capped by the quality of one brief. A brief that says "post about the new serum" produces eight versions of a generic serum post. Here is the shape that actually produces usable output, in the order it matters:

1. 01**The one claim.** A single sentence a reader could repeat to a friend. Not a theme, not a topic — a claim. "The 15 ml bottle now costs ₹399, down from ₹549."
2. 02**Who it is for, narrowly.** "First-time buyers who found us through a review video" beats "skincare audience".
3. 03**The proof.** One fact, number or detail that makes the claim credible. If you have none, say so — the system will write around it rather than invent one.
4. 04**The single action.** Buy, save, reply, register, download. One per brief. The CTA gets rephrased per platform, but only if it is one action.
5. 05**The exclusions.** Anything you cannot say this week — a price that is changing, a stock line that is thin, a competitor you do not name. This is faster than fixing it eight times in review.
6. 06**The asset you have.** Whether you are supplying a photo, want an image generated, or have a video ready for the 9:16 surfaces. That decides which of the eight are ready to publish today and which are waiting on you.

Six fields. Roughly ninety seconds. That is the input that turns into eight correctly sized, correctly written posts — and the reason the two minutes are worth spending is that a vague brief does not fail loudly. It publishes eight times.