READER PILOT / NEWS & TOOLSDated reporting. Explicit assumptions. Assessment intake unavailable.Pilot scope ↗
SUPPLY CHAIN
SAVINGS
MEASUREMENT / A CLEAR COUNTING BASIS

COUNT THE REQUEST.
NOT THE PERSON.

Public page requests are counted automatically, without a page-view checkbox. The count tells us which reviewed pages were requested—not who visited. Optional local tests and action-category sharing remain separate below.

01 / AUTOMATIC PUBLIC PAGE REQUESTS

OPEN A PUBLIC PAGE.
NO CHECKBOX REQUIRED.

Collection began at release on 25 October 2026 UTC, so that first day is partial. Eligible successful HTML GET requests to reviewed, canonical public pages are counted automatically by the Site. This works without JavaScript or a page-view opt-in. A normal public-page reload can count again, including a reload by the owner.

REQUESTS / NOT PEOPLE

Four aggregate fields. No visitor record.

The application stores only the namespace (REAL or TEST), UTC day, reviewed page path and count. It does not create measurement cookies or store IP addresses, user-agents, referrers, accounts, visitor IDs or session IDs in these request counts. Request headers may be inspected transiently to apply exclusions. Query strings, fragments and search terms are not stored in the counters; profiles, lanes, costs and contact details are never included.

These are requests, not unique people, complete visits or an exactly-once traffic census. Repeats and undetected bots can count. Cache or browser-history restores, network failures and exclusions can miss activity. This is not Search Console, search impressions, click-through rate or organic-search attribution.

Clear exclusions. No hidden preference switch.

Requests carrying DNT: 1 or Sec-GPC: 1 are excluded by design, along with obvious bots, prefetches, HEAD requests, private/admin pages, APIs, assets, redirects and errors. Explicit internal/test requests stay in TEST, separate from REAL.

Adding scs_count=off to a URL excludes that request only. It is not a persistent preference, does not apply to other links and cannot undo an earlier request. There is no browser-storage switch pretending to stop a request already received by the server. Open this page with the per-request exclusion ↗

A rolling window, not a deletion timer.

Request counts cover a rolling 90 UTC-day window. Older buckets are pruned on eligible collection or an owner request-report read, not by a scheduled job. Inactive storage and hosting backups can retain data longer. Combined counts cannot be isolated as an individual’s contribution because no person identifier is stored.

The former optional page-view collector is closed. Its historical opted-in views are kept separate from these new request counts. Local test counters and optional action-category sharing below remain independent choices. Standard hosting infrastructure may keep separate operational logs.

READ THE MEASUREMENT PRIVACY BOUNDARY ↗
02 / OPTIONAL BROWSER-LOCAL TEST

TEST ACTIONS IN THIS TAB.

These controls affect only local test counters, not automatic public page requests.

THIS TAB. FOUR COUNTS.

No new activity is counted. Shipping profiles and private inquiries are separate and unchanged.

OFF / NO COLLECTION
01 / LOCAL ACTIVITY0

Article pages opened

One count per document load while enabled. Reloads and repeat visits count again.

02 / LOCAL ACTIVITY0

Calculator pages used

First valid, explicit calculator interaction in a document. Initial examples are not interactions.

03 / LOCAL ACTIVITY0

Owner-test assessment reviews

Administrative category only; assessment intake is unavailable during the reader pilot. This is not a submitted inquiry.

04 / LOCAL ACTIVITY0

Owner-test save confirmations

Administrative category only; no public assessment save is available in the reader pilot. Receipt replays and email delivery are not counted.

Nothing counted in this tab. Zero is the correct starting point—not missing visitor data. Enable a local test if useful, then use an article or its calculator. Assessment review and save are not public reader-pilot features. There is no sample activity to inflate these numbers.

Open the article index ↗

These counters cannot establish a funnel: different pages may be opened repeatedly and categories are not linked into a person’s journey. One category counts at most once per document while enabled; reloads can count again. An inquiry save is a private database confirmation, not an email, a qualified lead or a completed sale.

03 / OPTIONAL FIRST-PARTY ACTION CATEGORIES

HELP IMPROVE THE PATH.
WITHOUT SHARING YOUR BRIEF.

Local testing above never authorizes server collection. This separate option can send the same four action categories to this Site’s private aggregate database. It is off until you explicitly choose it.

OFF / NO SERVER SHARING
WHAT GOES OVER THE CONNECTION

A category, not your customer record.

This browser sends one of the four categories, a random one-use event ID to avoid counting the same retry twice, and the current consent version. It sends no article ID, lane, profile, scenario, amount, email, search term, page URL or referrer. No advertising pixel or outside analytics provider is used.

Standard hosting and authentication infrastructure still processes network requests. The application’s metric records do not store IP addresses, user agents, accounts or cookies. No persistent person or session identifier is created.

LIMITED COUNTS / HONEST RETENTION

Test is always separate from public.

Private/staging and explicitly identified internal action activity is separated into TEST by the server. It is never presented as visitors, leads or conversions. The owner-only report shows daily action totals separately from automatic page requests; it cannot connect a person’s journey or attribute organic search.

Reports cover 90 UTC days. Older daily buckets and random receipts older than 24 hours are removed on the next valid aggregate POST—not by a scheduled deletion job. Inactive storage and hosting backups may retain them longer. Previously mixed counts cannot be removed per person because they carry no person identifier.

Safety caps allow at most 60 accepted actions per minute and 5,000 per UTC day across the Site. Each category is attempted at most once per document. Opt-out, connection failures and caps cause undercounting; failed sends are not queued or automatically retried.

OPEN THE OWNER-ONLY AGGREGATE REPORT ↗
LOCAL TEST ONLY / EXCLUDED FIELDS

Your business stays out.

The local test at the top of this page uses no names, email addresses, company selections, search terms, article identifiers, page URLs, shipping lanes, prices, profiles, scenarios or inquiry IDs. That local test makes no network request. Automatic public request counts and optional action-category sharing have their own clearly described boundaries above.

WHAT THESE CONTROLS CHANGE

Only local measurement.

Stop & clear disables counting and clears local test counts as each open tab next runs. It does not delete your shipping profile, comparison shelf, editorial records or private inbox. Those have their own controls.

INTAKE UNAVAILABLE
KEEP THE STREAMS SEPARATE

Requests are not action counts.

Automatic public requests, optional action categories and local tests are separate measurements. No checkbox enables page-request counting, and stopping a local test or optional action sharing does not stop the initial server page request. The legacy opted-in page-view collector is closed.

RETURN TO THE NEWS HUB ↗

Search Console and optional Clarity are separate.

Google Search Console reports Google search impressions, clicks and indexing; no visitor opt-in is needed for those search reports. Microsoft Clarity is optional: use Privacy settings to Accept analytics or Reject analytics. After acceptance it can measure masked interactions on selected public information pages, not articles, tools, account or financial pages. It uses its own pseudonymous cookies and is separate from our counts-only request report. Browser privacy signals and URL/referrer exclusions can keep it off. See the privacy notice for cookies, Microsoft uses, retention and withdrawal. A saved preference is not proof that a recording reached Microsoft.