Remediation Plan: Water Leak Insurance Claim

This is the prioritized plan to lift the original live bluebot.com page from its pre-pipeline baseline of 54 / 66 available (82%, PASS, GOOD) toward the 90%+ EXCELLENT band. The page already passes; this plan separates the genuine content gaps a manual edit can close from the layers only the full pipeline can unlock.

Content type: How-To Article (validated under the geoarticle profile, STANDALONE)
Source URL: https://www.bluebot.com/how-to-make-a-successful-water-leak-insurance-claim/
Current standalone score: 54 / 66 available, 82% normalized (PASS, GOOD)
Target: 90%+ (EXCELLENT) · Estimated edits: ~6 changes

This page already PASSES. The remediation below targets the EXCELLENT band and flags which deductions are real content gaps versus mode / profile artifacts. The downstream pipeline should focus on the genuine gaps (tables, citations, readability) and on the layers that standalone mode left unscored (schema, entity map, blueprint cross-reference). The actual post-pipeline rewrite scored 98/100 (PASS, EXCELLENT).

Unlike a failing baseline, the work here is additive, not corrective. The original page is a strong, publish-quality how-to with zero critical issues. The pipeline’s job is to capture the points it left on the table: chiefly the missing machine-readable layer (no JSON-LD schema), unverified entity coverage, and uncited headline statistics.

Priority Fixes: Next ~5 Points (ranked by points × effort)

#	Fix	Pts	Effort	Layer
1	Convert 2–3 key bulleted lists into comparison tables	+2	1 (low)	L1 Structure
2	Add inline source citations to headline statistics	+1.5	2	L7 / L8
3	Lower readability grade on the densest insurance-jargon sentences	+1.5	2	L2 Readability
4	Trim / tighten toward target word band (optional; profile artifact)	+1	2	L1 Structure

1. Add data tables (+2 pts), Effort 1

Convert these existing lists into HTML tables so AI engines extract them cleanly:

“Sudden vs. Gradual” coverage: two-column table (Covered / Excluded) with examples (burst pipe, appliance failure, water-heater rupture vs. slow seepage, deferred maintenance).
Adjuster estimate gaps: table of common omissions (drying / dehumidification, mold remediation, undervalued property, excluded ALE) with a “What to do” column.
Coverage categories: Dwelling / Personal Property / ALE rows with one-line definitions.

2. Inline-cite statistics (+1.5 pts), Effort 2

Attribute the headline figures to named, linkable sources:

“$13 billion/year in water-damage claims” → cite the Insurance Information Institute (III) or equivalent.
“$11,000 average payout” → cite source + year.
“5–10% premium discount for leak detection” → cite an insurer or industry survey.
The “43,200 data points/day” figure is first-party Bluebot data and is already adequately attributed, so leave as-is but label it as first-party.

3. Lower readability grade (+1.5 pts), Effort 2

Target FK grade 5–6 (currently ~7–8). Add a one-line plain-language gloss the first time each jargon term appears:

“duty to mitigate (your responsibility to limit further damage)”
“trace and access (coverage to find and reach a hidden leak)”
“additional living expenses / ALE (hotel and meal costs while you can’t live at home)”

Break the longest compound sentences in Steps 2, 5, and 6 into two.

4. Word-count trim (+1 pt, optional), Effort 2

~3,500 words exceeds the geoarticle band. This is mostly a profile-mismatch artifact: a how-to legitimately runs long. Only trim if standardizing to the geoarticle target; otherwise leave the depth intact (it benefits Claude / Perplexity readiness).

Layers That Need Pipeline Inputs (not fixable standalone)

These were skipped or capped only because standalone mode lacked structured inputs. They are where the optimization pipeline adds the most measurable lift, and account for the bulk of the climb from 82% to 98%.

Layer	Standalone	Full-mode ceiling	What unlocks it
5. Schema & SEO	0 (skipped)	10	Add `HowTo` + `FAQPage` + `SpeakableSpecification` JSON-LD; set a 55–60 char title and 150–160 char meta description
3. Entity Optimization	6 / 8	15	Supply the Entity Map → verify surface-form coverage, rotation, and disambiguation
8. AI Citation Readiness	6 / 7	15	Schema-gated signals (speakable, confirmed `@type`)
6. E-E-A-T	5 / 6	10	Blueprint cross-reference of author credentials + cited facts

Recommended schema additions for this page

HowTo root with a one-to-one step array across the 7 numbered steps (strong rich-result fit).
Author Person (Matthew Olin, jobTitle Founder & CEO, affiliation Bluebot), datePublished / dateModified (2026-04-04, ISO 8601), publisher (Organization: Bluebot / Lookout Lab, Inc.).
FAQPage with all 6 existing FAQ Q&A pairs.
SpeakableSpecification targeting the H1 and the opening definition paragraph.
Three Product mentions (Bluebot WiFi Smart Water Meter, Bluebot Mini, Bluebot EcoLink) within the single @graph.

Section-by-Section Fix Guide

Introduction

Keep the strong definitional opening (good for AI Overview). Inline-cite the $13B and $11,000 statistics.

“What Is a Water Leak Insurance Claim?”

Convert the sudden-vs-gradual distinction into a 2-column table. Gloss “sudden and accidental” vs. “gradual” in plain language.

Step 2 (Review Policy) & Step 5 (Adjuster / Estimate)

Add a coverage-category table (Step 2) and an adjuster-gap table (Step 5). Break long sentences. Gloss ALE and trace-and-access.

Step 7 (Future Prep) / Bluebot mention

Keep, but clearly label first-party data; ensure the product mention reads as evidence-collection guidance, not a sales pitch.

FAQ

Already 6 questions, which is ideal. Ensure each answer is self-contained (40–80 words) for snippet extraction. Map directly into FAQPage schema.

Compliance Note (YMYL Financial)

No outcome guarantees and good qualifier usage were observed. Maintain the balanced framing (process / probability, not promised payouts) when editing. Independent review of insurance-claims guidance is recommended before publication.

Forecast

If the page adds the recommended tables and inline citations, lowers the densest readability passages, and the pipeline publishes a full HowTo + FAQPage + SpeakableSpecification JSON-LD merger with a verified Entity Map cross-reference, the post-pipeline composite should land in the high 90s. Standalone manual recovery alone can reach roughly 87%, short of EXCELLENT, because the Schema (0→10), Entity Optimization, AI-Citation, and E-E-A-T ceilings only open with pipeline inputs. The actual achieved post-pipeline score was 98 / 100 (PASS, EXCELLENT).