Brand kit/Search Pulse

Inside your brand kit · SEO Analysis · Scan · Diagnose · Fix

What is search quietly costing your client?

The SEO diagnosis agencies bill for, built into the brand kit. The client's own Search Console data across the window you choose, the live results page photographed and dated — AI Overviews included — every finding diagnosed to one cause with evidence, fixes, and copy-ready rewrites, and a white-label client report. Re-run monthly: the recoveries are the retainer.

The monthly loop

Every pulse makes the last one pay off

Scan, diagnose, fix — then measure again. The difference from a one-off audit is the arc back to the start: the next pulse reads the same site after the fixes, so every recommendation returns as a number the client can see. Recoveries, on the record, month after month.

OWN
THE SERP
Scantheir Search Console · the live results
Diagnoseone cause · evidence
Fixrewrites · briefs
Re-measurethe next pulse

Most SEO tools stop at the audit. The re-measure arc is the retainer — every fix returns as a number the client can see.

The diagnosis, end to end

One click connects the account's Search Console read-only, one property maps to each client kit — then every pulse runs the whole arc with the evidence attached.

The window

From-date to-date, your call

SEO analysis is a time span, not a snapshot. Pick the window — it defaults to “since the last pulse” — and every number measures that span against the equal span before it. Every stat on the page carries its dates.

You pick the datesWindow-over-windowDated stats

The two lanes

Click gaps and quiet decay

A click gap ranks fine but loses the click — position healthy, click-through far under what that position should earn. Decay is the slow bleed: clicks falling, or the invisible traffic killer — click-through collapsing while Google keeps showing the page.

Click gapDecayThe invisible traffic killer

The live results page

Photographed, dated, honest

For every finding, the pulse photographs the live mobile results page — competitors, People-Also-Ask, and the AI Overview when Google shows one. Failed photos retry; when one still fails, the newest good photo is reused and clearly dated. Never a confident number the data can't back.

AI Overview captureDate-stampedRetries built in

The diagnosis

One cause, with receipts

No hedging: every finding gets exactly one primary cause from a fixed set of seven, a confidence score, and the evidence cited line by line. Click gaps come with three replacement titles and three descriptions, copy-ready, each with the persuasion lever named.

One cause of sevenEvidence citedCopy-ready rewrites

The proof of work

Show everything that was read

The Scan lists every page and every query–page pair Google reported, green-dotted as read and checked — plus the pulse line day by day, the pages and queries growing, brand versus discovery clicks, and the device split. The work is visible, not claimed.

Full scan ledgerThe pulse lineWinnersBrand vs discovery

The deliverable

A client report built to bill

White-labeled with the agency mark and the client kit's logo: executive verdict, the pulse line, the scoreboard, every finding with its diagnosis and rewrites, the full scan appendix, and a method page. Prints to PDF with an agency-ready filename — and the recommended-actions page detaches.

White-labelPrint to PDFDetachable actions

Run the whole diagnosis from Claude

Measure the window, read the findings and their one diagnosed cause, take the copy-ready rewrites, and forward the client report — the same diagnosis, driven from a conversation.

Works inClaude.ai (web)Claude CoworkClaude Code— no API key needed.

Run the diagnosis

Ask Claude to

“Run a search pulse for Acme — what is search quietly costing them?”

Work the findings

Ask Claude to

“Where is Acme leaking clicks, and what is the one cause behind each finding?”

Deliver it

Ask Claude to

“Generate the Search Pulse client report for Acme and send me the link.”

Where this is heading

The pulse finds and diagnoses; these are the corrective and conversational layers on the way.

The Fix PackCOMING

The chosen rewrites packaged for deployment, a writer-ready brief for every stale page, and schema snippets where the diagnosis calls for them — the deliverable the client deploys.

Compare-to-last delta cardsCOMING

When a pulse has a prior run, the stat cards carry the movement — “since the last pulse: clicks +12%” — the client-progress story on every card.

The depth laneCOMING

The third lane: pages that were never strong enough — compared against the top-ranking competitors for their query, with a content-expansion blueprint.

Moments vs windows — labeled apart

Window numbers come from the client's own Search Console data and are stable. The results-page photos are moments — Google changes that page hour to hour — so every photo carries the time it was taken, and an incomplete sample says so instead of showing a confident zero.

Google hides the rare long tail

Search Console omits rarely-searched queries for privacy, and only shows queries where the site actually appeared. The scan shows everything Google reports — which is honest, and less than everything that exists.

Read-only, always

The connection reads Search Console performance data and nothing else — nothing is ever written to the client's Search Console, and disconnecting removes the stored credential entirely.

A clean bill is a real result

The thresholds flag meaningful loss, not noise. A pulse with zero findings means nothing is measurably leaking at those thresholds — documented in the report as a clean bill, not an error.