Free · Six working files · No sequence, no drip

The operator’s kit

Everything on this site explains what to do and why. These are the six sheets we actually run it on — the difference between reading a framework and operating one.

The six files

These are not lead-magnet PDFs. They are the working artefacts, with the columns, thresholds and QA gates already in them, because a template without the thresholds is just a blank table.

1 · Search Protection Register
The baseline artefact everything else references. One row per protected query: raw funnel counts, the four computed indices, classification, and the paid-coverage column almost every keyword sheet omits — the one that decides whether an organic loss is alarming or expected. Append-only, dated, with the weekly refresh procedure.
2 · Attribute census sheet
The three-state census — valued, placeholder, blank — with the wrong-value scan, the contamination metric, and the 48-hour re-pull log that distinguishes an accepted edit from a silent rejection from a higher-authority revert. This is the sheet that finds problems no dashboard reports.
3 · Listing Issues scorecard
The five-layer audit scored 0–20 per layer, with the severity definitions and the blocker list. The composite is a communication device; the ranked issue list is the deliverable, and the sheet produces it.
4 · Term inventory template
Every term in a listing mapped to its function and its destination, with no blank rows permitted. This is the single control that stops a title migration silently dropping queries you had already earned, and it is the one most teams skip.
5 · Indexation sample sheet
The 20-term standard per hero — eight identity and spec, four highlights, three description-only, five backend-only — with the verbatim result log and the re-verification column. The backend-only rows are the point: nothing else can prove that field saved.
6 · Judge-by calendar
The measurement windows as a working calendar, with rollback triggers written before a change ships rather than argued about under pressure two weeks later.
Plus the changelog

An email when Amazon changes something that invalidates a procedure. Between May and August 2026 that would have fired five times — the 75-character title cap, the Highlights lock, the end of variation review pooling, Featured Offer eligibility, and Rufus becoming Alexa for Shopping. Each one broke advice that was correct the week before. It is not a newsletter and there is no sequence behind it.

What is deliberately not in it

The literal prompt library stays private. The procedures describe what each step produces, what its inputs are and what its QA gate is — enough for a competent operator to run them. The exact prompt wording is what we use on client accounts and it is not being published. If that is the part you wanted, the honest answer is that it is the part you would be hiring us for.

Get the kit

One email with the files. After that, only the changelog, and only when something actually changes. Unsubscribe in one click. We do not sell, share or rent your address. What we do with it.

Not sure whether you need any of this?

Run the scorecard first. Twenty checks, scored 0–100, ranked issue list by severity, no signup to see the result. If it comes back clean, keep your email address.

Score my listings →
ZBD Growth

Full Amazon channel management for established brands. Catalog, advertising, inventory and cases — run by a team you know by name, reported in depth.

© 2026 ZBD Growth. Selling on Amazon since 2012.Run by Hymie Zebede →