Skip to content
All case studies

Social Playbook, Marketing and content

Competitor activity turned into a weekly content plan

A content-operations SaaS that turns competitor activity into a reviewable strategy and weekly planning workflow.

Built for: Marketing SaaS product for brands and agencies

posts analyzed
14,200+
posts analyzed
social platforms
5
social platforms
competitor capacity
25
competitor capacity

First agent live in one week. A short call. Your first agent, a fixed quote, and the timeline. No obligation.

Social Playbook product

Sound familiar?

  • “We spend hours scrolling competitors and still do not have a plan.”
  • “Our research tools keep charging us for the same scrape twice.”
  • “Reports break when images expire or a job hangs forever.”

How the system works.

  1. Step 1

    Brand onboarding

    The business sets its brand context once, then adds the competitors it wants to track.

  2. Step 2

    Competitor collection across five platforms

    Apify collection jobs gather activity from Instagram, LinkedIn, TikTok, X, and Facebook, running in parallel with duplicate-job prevention.

  3. Step 3

    AI strategy that waits for complete data

    Anthropic Claude generates the playbook, weekly insights, and content calendar only after every competitor finishes loading.

  4. Step 4

    Reliable media and jobs

    Thumbnails are stored permanently in S3 and stuck jobs time out after 15 minutes with a clear message.

  5. Step 5

    A complete subscription product

    Stripe billing, plan limits, competitor slots, and scheduled weekly updates run the product end to end.

What changed.

  • 14,200+ posts analyzed across 5 platforms
  • Up to 25 competitors tracked per account
  • Duplicate scrapes, repeated AI generation, expired images, and endless loading all fixed in production
  • Collection, strategy, weekly updates, and calendar run as one repeatable workflow

The AI inside.

Anthropic Claude

Turns competitor activity into a strategy, weekly insights, and a content calendar, streamed with structured parsing and retries

Apify

Collects competitor posts and activity from five social platforms

The full stack.

Frontend

  • Next.js
  • React

Backend

  • TypeScript
  • Node.js
  • Express
  • Background ingestion workers
  • Weekly workers

Data

  • PostgreSQL
  • S3 media storage

AI and automation

  • Anthropic Claude
  • Apify
  • FFmpeg tooling

Integrations

  • Stripe
  • Resend

This fits you if...

  • Your team researches competitors by hand every week
  • You want a content plan built from real market activity, not guesses
  • You run a recurring AI workflow that keeps breaking on bad data or stuck jobs

Start with a free Agent Audit. First agent live in one week

The AI Operations Assistant brings this to your team.

An AI team member that handles the admin work across your email, documents, CRM, and spreadsheets, and asks before anything important goes out.

Full project write-upRead it

Project overview

Competitor research that produces a weekly content plan

Services: AI Agents ; AI Workflow Automation
Delivery: Competitor-intelligence SaaS, AI strategy, subscriptions, and background processing.

Project Snapshot

FigureWhat it represents
14,200+ postsProcessing volume reported in earlier portfolio records
5 platformsInstagram, LinkedIn, TikTok, X, and Facebook
Up to 25 competitorsDocumented account-plan capacity
15 minutesTimeout introduced for stuck scrape jobs
2 minutesGrace period for correcting an accidental competitor addition

The processing count is a historical reported figure. Plan capacity and timeouts describe the product, not customer growth.

The Problem

The product needed to turn competitor activity into a strategy and content calendar for the user's business. That required more than a model call: competitor data had to arrive, remain usable, and refresh on a schedule.

Delivery exposed practical problems. Adding a competitor could create duplicate scrape jobs and spend collection credits repeatedly. Analysis could start before every competitor finished loading. Social-platform image URLs expired, leaving missing thumbnails. Stuck jobs left the interface loading indefinitely.

Those failures affected both the quality of the plan and the cost of producing it.

What We Delivered

Business onboarding establishes the brand context. Users add competitors, collection jobs gather platform activity, and the backend prepares that material for AI strategy generation.

The application presents the playbook, competitor benchmarks, weekly insights, and a content calendar. Subscription billing and plan limits control access and usage. Scheduled updates repeat the collection and analysis workflow.

The March 27 delivery record documents parallel collection, duplicate-job prevention, and a generation gate: strategy, weekly update, and calendar run once after competitor loading finishes.

It also records persistent thumbnail storage to avoid expired CDN links and a fifteen-minute timeout for stuck jobs.

Engineering Decisions

Make collection completion explicit. Generating a strategy before the inputs are ready risks producing an incomplete comparison and repeating paid model work.

Store media rather than depending on temporary URLs. The user needs to inspect the source post when reviewing the analysis.

Treat competitor slots as a billable resource. Deleting an accidental addition within two minutes refunds the slot. Later deletion does not reset consumption; slots renew with the billing period.

Make failure visible. A timed-out job needs an end state and an actionable message, not an endless spinner.

Results

The product connects ingestion, strategy, weekly updates, calendar generation, billing, and administration. Existing portfolio records report 14,200+ posts processed.

The dated delivery log records fixes for duplicate jobs, repeated generation, expiring images, and indefinite loading. These are concrete operational improvements; the records do not establish a percentage reduction in marketing labor or growth in audience engagement.

Technology stack

Next.js and React; TypeScript, Node.js, Express, PostgreSQL; Apify; Anthropic Claude streaming API with structured parsing and retries; S3 media storage, FFmpeg tooling; Stripe; Resend; background ingestion and weekly workers.

Want this for your team?

Book a short audit call and show us the workflow. You leave with the first agent we would build, a fixed quote, and the timeline.

Get my free Agent Audit