All case studies
Project

One draft, every channel: a social workspace that rewrites each post for its platform

A marketing team running accounts on six platforms was writing every post six times. We built one workspace where a post is written once, rewritten for each platform from the team’s own past posts, scheduled, and measured in the same place.

At a glance

  • Connects Instagram, Facebook, LinkedIn, X, Threads, YouTube and Naver Blog accounts in one workspace
  • Rewrites one draft into each platform’s length, tone and hashtag habits, grounded in the team’s own past posts
  • Schedules and publishes in batches, then reports performance for every channel on one dashboard
  • Designed, built and shipped in six weeks, from architecture to release
The conversion screen: one post rewritten side by side as a Facebook post and a Threads post, each ready to upload
One draft, rewritten for Facebook and for Threads, side by side before upload.

The problem

Social media marketers, creators and the people who run a company’s accounts all do the same thing many times over. A post that works on Instagram is too long for X, too casual for LinkedIn and missing the structure a Naver Blog reader expects. So the same idea gets rewritten by hand for every channel, published from every channel’s own app, and measured in every channel’s own analytics, which never line up.

The work that matters, deciding what to say, is a small part of the day. The rest is copying, reformatting and checking six dashboards to find out what worked.

What we built

A single web workspace that sits over every account the team runs:

Connected accounts
Every platform signs in once through OAuth 2.0, and the workspace posts and reads through each platform’s own API.
Content conversion
A draft is rewritten for each platform by a retrieval-augmented model that looks up the team’s own past posts on that platform first, so the result sounds like them rather than like a template.
Scheduling
Posts are queued for each channel at its own time, or published to all of them in one batch.
Insights dashboard
Reach and engagement for every channel on one screen, with recommendations for what to post next.
Trend monitoring
A live view of what is rising on each platform, next to the draft being written.

Before a line of code, we mapped the whole service as one flow: sign-in, connecting accounts, writing, converting, reviewing, publishing and reporting, with every branch where a person decides something. The build followed that map.

A service flow diagram from sign-in through account linking, drafting, conversion, review, publishing and reporting (opens full size in a new tab)
The service flow, drawn before the build: every step from sign-in to the report, and every point where a person decides.

How it was built

Six weeks, end to end. The first week went to the system design and architecture; the second and third to the FastAPI backend and the React front end; the fourth to the conversion engine and the platform integrations; the fifth to the dashboard and live analytics; the sixth to testing and release. The interface was designed in Figma alongside the build and works on any screen size.

The architecture was built to grow: a new platform is a new integration and a new set of past posts to retrieve from, not a change to the engine.

More reading

Case study

How a franchise network let buyers search 100,000 products with a photo

Buyers on a promotional-products franchise network knew what they wanted but not what it was called, and keyword search across more than 100,000 products could not help them. An AI search now finds products from a photo, a phrase, or both, on the head-office site and every franchise storefront.