> For the complete documentation index, see [llms.txt](https://oodlz-ai-studio.gitbook.io/docs/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://oodlz-ai-studio.gitbook.io/docs/reporting-agents/growth.md).

# Growth

Monthly performance data across SEO, web analytics, and 5 social channels — with plain-English AI summaries on every metric, delivered to the workspace dashboard.

> **Performance reporting on autopilot — 7 channels, every month, with AI summaries.**

Most analytics setups give the brand 7 separate dashboards across 7 platforms and ask the human to assemble the picture. The Oodlz Growth producer pulls every channel monthly, calculates month-over-month deltas, generates a plain-English summary per channel, and presents everything in a single workspace dashboard. The brand spends 30 seconds scanning the month, not 3 hours assembling reports.

## What You Get

* 7 channels reported monthly — SEO (Google Search Console), web analytics (GA4), and 5 social (Instagram, Facebook, LinkedIn, TikTok, X).
* Month-over-month deltas calculated and stored — every metric shows the change vs the prior month, with the right colour direction (lower position is better, higher engagement is better).
* AI-generated 3-bullet summary per channel per month — plain English, no jargon, written from the actual delta data.
* 6-month trend chart per channel — instant visual on direction.
* Single workspace dashboard with month selector and channel tabs — no juggling 7 different platform UIs.
* On-demand reruns — fire the workflow manually if needed; the upserts overwrite cleanly without duplicating.

## How It Works

A monthly cron fires on the 2nd of every month at 06:00 UTC — late enough that every platform has finalised the prior month's data. Each channel's workflow runs the same pattern: pull the brand profile, parse the target period, check the connection status for that channel (skip with a clear log entry if the platform is disconnected), call the platform's API for the current month and the prior month, calculate deltas, and upsert one record per channel per month into the workspace.

Once the metric record is written, a sub-workflow runs the metric snapshot through an LLM with low temperature (0.3) and a fixed prompt to produce 3 plain-English bullets per channel — facts and direction only, no recommendations, no invented context. The summary writes to the workspace alongside the metric record.

The dashboard reads from the workspace directly. Month selector + channel tabs + hero stat cards + delta badges + AI summary card + 6-month trend chart, per channel.

## Quality Baked In

* 1 cron at 06:00 UTC on the 2nd of each month — no manual triggering, no missed reports.
* Connection check before every API call — disconnected channels exit with a clear log entry, no broken-token errors mid-run.
* Per-channel null safety — fields the platform doesn't expose render as `—` in the dashboard, never as 0.
* Avg Position delta inverted in colour coding — lower is better for SEO position, the dashboard shows green when the number drops.
* Idempotent upserts — re-running the same period overwrites the existing record cleanly, never produces duplicates.
* AI summary uses temperature 0.3 — low variance, factual tone, no creative invention.
* Input snapshot stored alongside every summary — the LLM-generated text is always traceable to the exact numbers it described.
* 6-month trend chart per channel — instant visual on whether direction is improving, regressing, or flat.

## Inputs

* Brand profile fields — GSC site URL, GA4 property ID, social account IDs per connected platform.
* Connected platform credentials in the n8n credential store — OAuth2 for Google and LinkedIn, Bearer tokens for Meta-graph, TikTok, X.
* Channel Connections records in the workspace — used to skip disconnected platforms cleanly.

## Outputs

* 7 metric records per month (1 per active channel) in the workspace.
* 7 AI summary records per month (1 per active channel) in the workspace.
* 1 workflow run log entry per channel per month — counts, status, error summary if applicable.
* Dashboard pre-populated for the new month within 1 hour of the cron firing.

## Cadence And Volume

Monthly. 06:00 UTC on the 2nd of each month. On-demand reruns possible via webhook with an optional period parameter. Cost per run is dominated by LLM summaries — $0.01–$0.05 per channel per month. APIs are free (GSC, GA4) or part of existing platform access.

## Your First Month

When your workspace is brand new, the report cards start out empty and each channel's summary reads *"No activity recorded."* That is expected, not a fault — your reports fill in as real performance accrues, so the first set of meaningful numbers lands once you have roughly a full month of activity behind you. From there, every card and summary populates on its own, month after month.

## What This Replaces

* The marketing analyst pulling 7 separate platform reports manually each month.
* The BI tool subscription that costs $2,000+ per year and still requires a human to write the summary.
* The dashboard that shows 47 metrics nobody reads — replaced by 4–6 hero metrics per channel and a 3-bullet AI summary.
* The "what changed this month" question that takes a half-day to answer — answered automatically in plain English at the top of the dashboard.

One pipeline, 7 channels, AI-summarised, delta-calculated, dashboard-ready — every month, autonomously.

***

*This was built. No agent ships this out of the box.*