Sample Deliverable
Sample deliverable — Factory Investigator self-audit
Not a client case study.
This is a self-audit of Factory Investigator’s own public website. It shows the assignment, evidence, findings, priorities, implemented status, open verification and limitations a manufacturer should expect to see—without presenting internal work as client proof.
Assignment / Scope
Inspect the public buyer and publication system end to end.
The assignment covered the homepage, global navigation, five service journeys, the Manufacturer Website Investigation form, representative article templates, the industry directory and representative category, News and industry archives.
- Homepage proposition, Investigator continuity and CTA hierarchy.
- Masthead, six evidence desks, service navigation and five service-detail shells.
- Investigation form trust, validation, success-state contract, consent and lead-event behavior.
- Article headings, dates, rail, policy links and intent continuity.
- Industries structure and depth, plus responsive media, mobile behavior and self-hosted fonts.
Method / Public Evidence
Observe first, verify the renderer and form contract, then separate fact from inference.
The independent pass on 28 July reviewed public pages at desktop, tablet and mobile sizes, exercised visible interactions, ran a bounded crawl and lab performance checks, and then compared the public result with the documented renderer and release state. The four captures below were made anonymously from public pages on 29 July 2026 after the verified renderer, form and service-image releases. They contain no WordPress administration interface, personal information or private analytics.
The Rubric
Twenty checks, each tied to visible evidence or a bounded contract test.
- PropositionCan a manufacturer understand the publication’s operating purpose and the commercial boundary?
- Investigator continuityDoes the recurring character and black/brass/ivory evidence world remain coherent across journeys?
- CTA hierarchyAre Library, Industries and Website Investigation actions distinct without competing for the same decision?
- Masthead and taxonomyDoes navigation expose exactly the six current evidence desks without retired categories?
- Services panelIs the horizontal panel usable with mouse, keyboard, Escape and focus return?
- Five service shellsDo the pages use consistent framing, current links, scope, boundaries and next actions?
- Five service imagesAre scenes distinct, relevant, responsive and framed without invented clients, text or outcomes?
- Form trustAre purpose, use of submitted data, public-page scope and optional paid implementation clear?
- Form error contractDo invalid fields restore safe values, expose linked errors, move focus and prevent a false lead?
- Form success contractCan success appear only after a valid, signed, one-use result rather than a public query string?
- Consent and analytics behaviorAre lead events consent-aware, deduplicated and separated from administrator-only test paths?
- Skip links and keyboardIs there one skip link, visible on focus, with keyboard-safe global navigation?
- HeadingsDoes each checked page keep one H1 and a logical hierarchy without renderer-created competition?
- Link contrastAre Editorial Policy and contextual links clearly legible, underlined and focus-visible?
- Date metadataDoes the template show Published and show Updated only when the calendar dates differ?
- Right railAre reading progress, sharing, TOC, evidence signals and service paths legible without changing the article body?
- Industries structureDoes the directory retain meaningful H2 groups and honest routes without changing search controls?
- Industries depthDoes each exposed sector have sufficient evidence, or is its current thin/empty state stated honestly?
- Mobile and mediaDo pages avoid horizontal overflow, retain useful crops and serve responsive WebP derivatives?
- FontsDo Fraunces, Source Serif 4, DM Sans and IBM Plex Mono remain self-hosted and role-disciplined?
28 July Findings · Point in Time
A strong identity and strong individual articles were carrying visible template and proof gaps.
- The independent overall score was 7.5/10.
- RFQ conversion assurance scored 4.5/10 because success, mail-result handling and lead measurement were not then trustworthy.
- There were 24 industry archives, all then non-indexable; 15 were empty, six held one post, and most of the remainder were thin. That finding does not imply thin archives should be indexed.
- The strongest assets were the distinctive Investigator identity, information design, policy discipline and substantial individual articles.
- The main gaps were a truthful form contract, one coherent rendering system, inspectable commercial proof, thin industry journeys and generic article-to-next-step continuity.
Three Priorities
Fix reliability and inspectability before adding bigger claims.
Truthful enquiry states
Success, error and analytics must describe what the server actually accepted, with replay and duplicate paths rejected.
One publication/rendering system
Masthead, six desks, archive framing, dates, article furniture and service journeys should come from one current renderer family.
Inspectable proof
Show the assignment, evidence, priorities, implemented state and limitations instead of substituting invented credibility.
Status at 29 July 2026
Implemented, partial and unverified are kept separate.
Implemented / publicly verified
- One skip link and six current desks.
- Current archive renderer, headings, dates, rail spacing and legible policy links.
- Signed one-use RFQ result contract with accessible invalid-email restoration and spoof/replay rejection.
- Five distinct responsive service images and consistent service shells.
Prepared / partial
- Industries directory structure is preserved, but sector depth remains uneven.
- Commercial proof begins with this self-audit specimen; it is not client outcome evidence.
- Article next steps are intent-aware at renderer level, while the existing rail and article bodies remain unchanged.
Still unverified
- Production inbox receipt and spam placement.
- Client outcomes, revenue impact, ranking lift or qualified-lead lift.
- Field Core Web Vitals and operating-system reduced-motion behavior.
- Whether every thin industry will later justify a substantive public hub.
Output
What a bounded first review records.
- Pages and buyer journeys inspected.
- Dated, safe public-page evidence captures.
- Finding severity and confidence, separated from assumptions.
- Three priorities that explain what should happen first.
- Immediate corrections versus controlled-verification items.
- Implemented, partial and unverified status.
- Search, content and measurement boundaries.
- Limitations and explicit non-claims.
Limitations / Non-Claims
This document demonstrates the form of the work, not a client result.
- This is a Factory Investigator self-audit and not a client case study.
- It makes no claim of client revenue, ranking, enquiry, qualified-lead or conversion lift.
- It does not claim production inbox receipt or email deliverability.
- It does not claim field Core Web Vitals; the dated audit used lab measurements with disclosed variance.
- It does not imply that indexing empty or thin industry archives is desirable.
- It does not promise delivery time, service outcome or a paid engagement.
Your Public-Site Review
Request your $0 first review
Paid implementation is optional.