Getting Started

How the Mavrtr Pipeline Works

A transparent look at the multi-stage pipeline that turns a store URL into a strategy brief.

When you hit Run Intelligence, the Mavrtr pipeline kicks off an async multi-stage run. Each stage runs sequentially — the output of one feeds the next.

Stage 1 — Catalog

The pipeline pulls the store's public Shopify product catalog using the storefront API. Product titles, types, descriptions, prices, and handles — enough to understand what the brand sells and where your focus product sits.

If the store isn't on Shopify, this stage fails and the brief can't proceed.

Stage 2 — Market reading

This is the heaviest stage. A specialised research agent is handed either a product brief or a brand brief (depending on your research mode) and instructed to read primary sources in this order:

The brand's own reviews (Trustpilot, on-page reviews, Yotpo, ProductReview)
Relevant subreddits and forums for the product's niche
Amazon reviews of close substitute products
Social comments on competitor ads
2–3 direct competitor review pages

The agent synthesises all of this into structured signals: review signals, Reddit signals, competitor signals, and objection signals — every one sourced from URLs the agent actually visited during that run.

The market read is cached per store + product + mode for up to 7 days. Re-running a brief with the same settings within that window uses the cached read, which is faster. Different modes (product vs brand) never share a cache entry.

Stage 3 — Customer intelligence

The market signal is fed to a customer-intelligence analyst pass. It produces 2–3 ranked customer segments — each one describing a meaningfully different type of buyer. For each segment you get:

Demographic + psychographic snapshot
Pain points ranked by severity (1–10)
Desires ranked by intensity (1–10)
Objections with sharp one-liner rebuttals and recommended placements
Verbatim quotes pulled from the read
Awareness stage (Schwartz's 5 levels)
Dream outcome

Stage 4 — Strategy

A senior-media-buyer pass takes the segment intelligence and writes the strategy:

Positioning statement
What's already working / key risks
Ranked ad angles, each with a hook (≤14 words), primary text, CTA, target segment, confidence score (1–10), and rationale
A CBO campaign structure with 3–5 ad sets, budget shares, audience descriptions, and placement recommendations

Stage 5 — Deployment artifacts

The brief is rendered into a branded HTML template using Jinja2 and a PDF is built with WeasyPrint. Meta Ads CSV is built from the angles. Notion / Zapier / webhook integrations fire if configured. You get a notification when the brief is ready — usually within 2–5 minutes.

What happens if a stage fails

If any stage encounters an unrecoverable error, the brief is marked as Failed with an error message. Common causes:

The store isn't on Shopify (Stage 1)
The store has a very small catalog or is heavily JS-gated (Stage 1)
Market reading timed out due to an unusually obscure niche (Stage 2)
The model returned malformed XML after the maximum retry attempts (Stage 3 or 4)

Failed briefs don't count against your monthly cap.