Reporting that narrates itself.
Pull Google Analytics, Search Console, Discover, Ads, and BigQuery into one workflow, let a Data Analyzer and LLM turn the numbers into a narrative — executive summary, key wins, concerns, and recommendations — and ship it to Google Sheets on a schedule.
From raw numbers to narrative.
Every Google source
Analytics 4, Search Console, Discover, Ads, and BigQuery — read into one workflow.
Plain-language analysis
The Data Analyzer answers questions about your data without SQL or code.
Narrated reports
An LLM turns KPIs into an executive summary, wins, concerns, and next steps.
Week-over-week
Pass last period's data alongside this period's to compute and explain the deltas.
Composable
Works with the rest of the agentic platform out of the box.
Observable
Every run is logged, traceable, and resumable with retries and quotas.
Every Google data source, wired into one workflow.
Read Google Analytics 4 by account and property, pull Search Console queries and pages, surface Discover traffic, summarize Google Ads campaigns, or run SQL against BigQuery. Each node returns clean JSON ready for the next step.
From raw JSON to the rows that matter.
Source nodes return JSON strings — use the JSON Path Extractor to pull the exact fields you need, Merge to combine sources, and Loop to iterate. No intermediate exports, no manual cleanup.
Ask your data questions in plain language.
Point the Data Analyzer at structured data — CSV, XLSX, JSON, or XML — and ask in plain English. It summarizes, filters, detects patterns and anomalies, and extracts insight with the LLM provider you choose. No SQL, no code.
A report that explains itself.
Executive summary
The LLM opens with a tight read on how the period went, in your team's words.
Key wins & concerns
Standout gains and the metrics slipping the wrong way, called out explicitly.
Recommendations
Concrete next steps grounded in the numbers, not generic advice.
Week-over-week
Structured wow_changes computed from this period against the last.
A weekly cross-channel report, end to end.
Feed the LLM the current period and the previous one; it returns structured JSON — executive_summary, key_wins, concerns, recommendations, and wow_changes — ready to write to a sheet on a schedule.
How a report writes itself.
Configure sources
Add Google Analytics, Search Console, Ads, or BigQuery nodes visually in Studio — no code required.
Transform & analyze
Extract the fields you need, then ask the Data Analyzer for trends, anomalies, and summaries.
Narrate with the LLM
Generate a structured report — summary, wins, concerns, recommendations, and deltas — as JSON.
Publish & schedule
Write the report to Google Sheets and run it on a cadence, with retries and full run logging.
What teams ask before they commit.
What data sources can Draft & Goal pull into a marketing report?
Draft & Goal reads Google Analytics 4, Search Console, Google Discover, Google Ads, and BigQuery into a single workflow. Each source node returns clean JSON ready for the next step, so you can shape, merge, and analyze cross-channel data without intermediate exports, then publish the finished report to Google Sheets.
How does AI-narrated reporting work in Draft & Goal?
An LLM step turns your KPIs into a structured narrative: an executive summary, key wins, concerns, concrete recommendations, and week-over-week changes. The output is structured JSON grounded in the numbers from your connected sources, ready to write into Google Sheets as a shareable report rather than a wall of metrics.
Can I analyze marketing data without SQL or code?
Yes. The Data Analyzer node in Draft & Goal accepts structured data such as CSV, XLSX, JSON, or XML and answers questions asked in plain English. It summarizes, filters, detects patterns and anomalies, and extracts insight with the LLM provider you choose, with no SQL and no code required.
Can Draft & Goal run reports automatically on a schedule?
Yes. A reporting workflow can run on a cadence, such as a weekly cross-channel report, and write its output to Google Sheets automatically. Every run is logged, traceable, and resumable, with retries and quota handling, so recurring reports keep arriving without someone triggering them manually.
How does Draft & Goal compute week-over-week changes?
You pass last period's data alongside the current period's within the same workflow, and the LLM computes and explains the deltas. The report returns structured wow_changes together with the executive summary, key wins, concerns, and recommendations, so stakeholders see what moved and why, not just raw numbers.
Show us the workflow.
We'll show you the 10x.
Bring the marketing workflow that eats your week. We'll build it live, with your data and your models, in 30 minutes.