How it works

Data in, decisions out.

Nothing here is a black box. Below is exactly where we get the data, how it's processed, what a pack contains, and how fast it reaches you.

Step 01 — Data sources

We only read what stores already publish.

No private accounts, no logins, no buying someone's database. Just the public product data that storefronts make available on purpose.

Shopify public data

Shopify storefronts expose a public products endpoint that lists products, variants, prices, compare-at prices and availability. We read that, the same way a browser does.

products / variants price compare-at price availability published date

Woo Store API

Stores running WooCommerce expose a public Store API with product listings, prices and stock status. We read that too, so the set isn't limited to one platform.

product listings price stock status categories

We stay inside what's public. If a store hides its data behind a login or blocks automated reads, we don't force our way in — we leave it out and tell you which stores made the cut.

Step 02 — The pipeline

Four moves from raw pages to a conclusion.

Deterministic where it must be (the numbers), AI where it helps (the reading and the writing).

STEP 01

Collect

We pull product and stock records for the stores in your category from their public endpoints.

STEP 02

Structure

Records are normalised into one schema: store, product, variant, price, discount, stock, date.

STEP 03

Cluster & analyse

Products are grouped into price bands and formats; stock movements and new listings are flagged.

STEP 04

Conclude

The AI turns the analysis into plain-language findings and positioning suggestions — figures unchanged.

Step 03 — The report

What's in the pack.

Five blocks, in a fixed order, so you can skim in five minutes or dig into the numbers for an hour.

01

Competitive price bands

List price, original price and discount depth per competitor, grouped into bands — premium, core, value — with your position marked.

02

Stock-out & scarcity signals

Variants that dropped to sold-out or low-stock since the last snapshot, so you can catch demand while it's live.

03

New-arrival tracking

Freshly published products, formats and sizes across the set, with the launch price and first-seen date.

04

Whitespace positioning

Where the market is crowded, where it's thin, and a plain-language read on the angle worth testing.

05

Raw data table (CSV)

Every row behind every claim: one row per product/variant, one column per field, timestamped. Yours to sort and pivot.

+

Cover note

A one-page summary: which stores were scanned, the capture date, and the three findings worth acting on first.

Step 04 — Delivery

Paid, then in your inbox in about 30 seconds.

Checkout finishes, and the download link goes out automatically. No waiting for a human, no "we'll be in touch", no dashboard to log into.

Three formats, one link

A PDF you can read or forward, a CSV you can open in a spreadsheet, and a web report you can view in a browser.

Automatic, not manual

The link is generated the moment payment clears. For subscriptions, refreshed packs arrive weekly without you asking.

Yes, a human is reachable

If a link fails or something's off, reply or write to [email protected]. A person answers.

your-inbox~30s
Payment received

Your PixHarvest pack is ready

Category: creatine · 14 stores scanned · captured 2026-05-12 09:14 UTC

report.pdf data.csv web report

Illustrative email. Real packs contain real scraped data.

See it for yourself

The fastest way to judge this is to read one.

Open the free sample and look at the format, the bands, the signals. Then decide if your category is worth $19.