If I Were Running Paul Davis Restoration’s SEO, Here’s What I’d Do Differently

I’m about to do something that most agency owners would never do: tell you exactly what went wrong with one of restoration’s most strategic franchises.

Not conspiracy theories. Not guesses. The actual data that explains why Paul Davis Restoration — a $2+ billion company with 600+ franchises across North America — lost half its organic keyword portfolio between November and December 2025.

Why? Because I pulled their SpyFu data this morning, and what I found was different from the 911 Restoration story I told three weeks ago. This isn’t a domain in freefall. This is a franchise that was actually winning — growing their keyword portfolio from 39K to 50K through most of 2025 — and then tripped on the finish line.

That’s not a systemic failure. That’s a fixable problem. And the recovery opportunity is enormous.

The SpyFu Data: A Franchise That Peaked, Then Stumbled

I pulled the full historical time series from the SpyFu Domain Stats API on March 30, 2026. Here’s what pauldavis.com looks like over the last 12 months:

Period	Organic Keywords	Monthly Organic Clicks	SEO Value ($/mo)	PPC Spend ($/mo)	Domain Strength
Mar 2025	38,980	10,260	$370,100	$20,950	51
Apr 2025	39,220	7,638	$387,500	$24,300	51
May 2025	41,620	11,420	$431,000	$27,380	49
Jun 2025	42,620	11,830	$450,200	$31,940	49
Jul 2025	45,220	12,990	$482,800	$35,990	49
Aug 2025	48,420	14,670	$532,800	$37,940	50
Sep 2025	49,470	15,430	$491,200	$57,140	52
Oct 2025	50,339	14,490	$484,200	$49,000	52
Nov 2025	49,400	14,420	$484,300	$665,600	53
Dec 2025	23,250	12,620	$372,400	$258,500	51
Jan 2026	22,490	12,930	$365,100	$213,000	51
Feb 2026	22,190	13,590	$952,800	$206,100	54

Look at the trend. From March to October 2025, Paul Davis did exactly what every restoration company should be doing: they grew. 39K keywords → 50K keywords. $370K/month SEO value → $532K/month. That’s not a fluke. That’s execution. That’s a team running the playbook.

Then November happened. PPC spend spiked to $665,600 — an 18.5x increase from October’s $49K. The same panic pattern I saw with 911 Restoration. And by December? Half the keywords vanished. 50K → 23K. That’s a 54% collapse in a single month.

But here’s the thing that makes Paul Davis different than 911 Restoration: their SEO value per keyword is actually higher. At $43/keyword (based on Feb 2026 data), Paul Davis is ranking for higher-value keywords than most competitors in this space. That tells me they weren’t ranking for junk keywords. They were ranking for money terms — the ones that matter.

Which means the fix isn’t a rebuild. It’s a recovery.

What Actually Happened in Q4 2025: The Diagnostic

Let me be direct about what I think happened. A keyword collapse from 50K to 23K in a single month isn’t gradual content decay. That’s one of three things:

Scenario 1: A location page massacre. Paul Davis has franchises everywhere — across all 50 states. If someone restructured the location page architecture, consolidated pages, or switched hosting/CMS without a clean redirect map, Google would have vaporized thousands of pages from the index overnight. Franchise sites live and die on location pages. Lose those, lose everything.

Scenario 2: A technical issue that broke indexation. A rogue robots.txt rule, an accidental noindex tag at the template level, a CDN misconfiguration returning 404s to Googlebot — any of these can silently deindex thousands of pages while organic traffic is still flowing because cached versions serve users fine. You don’t notice until you check GSC and see “Excluded – currently not indexed” spiked by 50%.

Scenario 3: The November Google Core Update hit harder than anticipated. Google dropped a core update in November 2025. If Paul Davis’s location pages are thin, templated content with minimal local differentiation, the update could have targeted them specifically. Combined with algorithm changes favoring AI-extracted answers and entity authority, thin content gets deprioritized fast.

My money? Scenarios 1 and 3 combined. But I’d verify with data before doing anything permanent.

Step 1: The 72-Hour Diagnostic Audit

Before touching a single page, I need to know what’s actually broken.

Day 1: Crawl and Index Validation

I’d run Screaming Frog against the full pauldavis.com domain — every page, every redirect. For a 600-franchise network, I’m expecting 8,000-15,000+ URLs. I’m specifically looking for:

• Redirect chains longer than 2 hops — These leak PageRank and slow crawl budget.
• Orphaned location pages — Pages that exist but have zero internal links. If city pages aren’t linked from a parent hub, Google treats them as low-priority and deprioritizes crawling.
• Canonicalization issues — A single bad canonical tag at the template level can tell Google to ignore thousands of pages simultaneously. This is the most common cause of sudden deindexation I see.
• JavaScript rendering problems — If Paul Davis uses any client-side rendering for critical location content, I’d compare Screaming Frog’s text extraction vs. what a headless browser sees. Mismatch = indexation risk.
• Soft 404 patterns — Pages returning 200 status code but with “not found” content structure. Googlebot gets confused. Pages don’t index.

Day 2: Google Search Console Analysis

I need 16 months of GSC data — the period before and after the collapse.

Specifically:

• Coverage report trends — Did “Valid” pages spike downward in November/December? Did “Excluded – currently not indexed” spike upward? The answer tells the story.
• Performance by URL pattern — Segment by location pages, service pages, blog content. Which pattern lost the most impressions? If it’s /locations/*, it’s an architecture problem. If it’s /services/*, it’s content quality.
• Exclusion reason breakdown — What’s excluding the pages? “Blocked by robots.txt”? “Crawled – currently not indexed”? “Redirect error”? Each reason points to a different root cause.
• Query data comparison — Export top 5,000 queries from October 2025 vs. February 2026. Which keyword clusters disappeared? If it’s geo-modified queries (“water damage restoration [city]”), location pages are the problem. If it’s service-level queries, the content strategy failed.

Day 3: Competitive Analysis

I’d pull the same SpyFu data for SERVPRO, 911 Restoration, ServiceMaster, and Rainbow International. If all of them declined in November/December, it’s an industry-wide algorithm shift. If Paul Davis uniquely declined, it’s site-specific.

Then I’d audit the top-ranking competitors for Paul Davis’s highest-value lost keywords. What does their architecture look like? How many location pages? What schema are they using? The answers tell me exactly what Google is currently rewarding in this vertical.

The Recovery Strategy: Rebuild What Was Already Working

Here’s the critical insight: Paul Davis doesn’t need a redesign. They need a rescue. They proved they could rank for 50K keywords. Now I need to figure out what broke and fix it, then scale what was already working.

Priority 1: Recover the Indexation Foundation (Days 1-30)

This is the emergency phase.

Canonical tag audit: If there’s a template-level canonical issue, it’s a one-line fix that could immediately un-exclude thousands of pages. I’d verify canonicals across 50+ representative pages from different URL patterns (locations, services, blog) and check GSC’s URL Inspection tool to see what Google actually crawled vs. what we think we served.

Location page linking structure: I’d verify that every location page is explicitly linked from a parent hub page. No links = low crawl priority = Google ignores the page even if it’s technically valid. A simple site map regeneration or parent page update can fix this.

Robots.txt validation: One bad rule and 90% of your site might be blocked from crawling. I’d audit the current robots.txt, compare it against historical versions (via Wayback Machine if needed), and remove any rules that shouldn’t be there.

Redirect map cleanup: Any redirect chains longer than 2 hops get collapsed to 1-hop direct redirects. Every hop loses 10-15% of PageRank. In a franchise network with hundreds of redirects, this can be thousands of dollars in lost equity.

Priority 2: Location Page Architecture Renaissance (Days 30-90)

Now we rebuild what was working.

Paul Davis has 600+ franchises. That’s 600+ locations that could have dedicated SEO landing pages. If they’re structured right, that’s 3,600+ pages (600 locations × 6 core services: water damage, fire damage, mold remediation, storm damage, sewage backup, dry cleaning/contents restoration).

Each page needs:

Locally-specific content that proves expertise. Not “water damage restoration in Houston” templated 500 words. I’m talking about: “Houston’s sub-tropical climate creates unique challenges — the combination of high humidity, frequent thunderstorms, and clay-based soil means water damage in Houston spreads faster than in drier climates. Our Houston team is trained on Gulf Coast moisture dynamics, local building codes, and Houston’s specific insurance requirements.” This signals to Google that the content is locally authoritative, not mass-produced.

LocalBusiness schema with complete NAP + service area. Every location page needs JSON-LD marking up the franchise location with exact coordinates, service area polygon, hours (24/7 for emergency response), and a catalog of specific services with local pricing where available.

Embedded Google Map. A map showing the service area reinforces local relevance and keeps users on-site instead of searching for competitors.

Real project stories. “In March 2025, our Paul Davis team responded to a commercial water intrusion affecting 8,000 sq ft of office space in downtown Houston. Complete water extraction and structural drying completed within 48 hours.” Specificity builds trust with both users and algorithms.

Priority 3: Content Depth Beyond Location Pages (Days 60-120)

Now I add the layers that Google currently rewards.

Crisis-moment content (targets the 2 AM searcher):
– “What To Do When Your Basement Floods: A Step-by-Step Emergency Checklist”
– “I Smell Mold In My House Right Now — What Should I Do First?”
– “Fire Damage: What To Do In the First 24 Hours”

These need HowTo schema, numbered steps, and definition boxes at the top for AI Overviews to extract. They capture intent before the decision to hire a pro is made.

Decision-stage content (targets the insurance call):
– “Water Damage Restoration Cost in 2026: A Regional Breakdown”
– “Homeowners Insurance and Water Damage: What’s Covered and What Isn’t”
– “Mold Remediation Timeline: Expectations From Day 1 to Completion”

These need comparison tables, cost breakdowns, FAQPage schema. This is where Paul Davis wins against SERVPRO.

Authority-building content (earns backlinks, builds topical authority):
– “The Complete Guide to IICRC Certification Standards: S500, S520, and What They Mean”
– “Understanding FEMA Flood Zones: How to Check Your Risk and What It Means for Insurance”
– “Water Damage vs. Water Intrusion: Why the Distinction Matters (and What Your Insurance Company Cares About)”

These earn backlinks from IICRC, FEMA, RIA, insurance publications, and local news outlets. Those links flow authority to location pages through internal linking.

Priority 4: Schema Markup at Scale (Days 45-90)

For a 600-franchise network, schema markup scales multiplicatively.

Every location page needs:

{
  "@context": "https://schema.org",
  "@type": "LocalBusiness",
  "name": "Paul Davis Restoration of [City]",
  "telephone": "+1-XXX-XXX-XXXX",
  "address": {
    "@type": "PostalAddress",
    "streetAddress": "[Street Address]",
    "addressLocality": "[City]",
    "addressRegion": "[State]",
    "postalCode": "[ZIP]"
  },
  "geo": {
    "@type": "GeoCoordinates",
    "latitude": "[LAT]",
    "longitude": "[LONG]"
  },
  "openingHoursSpecification": {
    "dayOfWeek": ["Monday", "Tuesday", "Wednesday", "Thursday", "Friday", "Saturday", "Sunday"],
    "opens": "00:00",
    "closes": "23:59"
  },
  "areaServed": {
    "@type": "City",
    "name": "[City], [State]"
  },
  "hasOfferCatalog": {
    "@type": "OfferCatalog",
    "itemListElement": [
      {
        "@type": "Offer",
        "@id": "https://pauldavis.com/[city]/water-damage-restoration/",
        "itemOffered": {
          "@type": "Service",
          "name": "Water Damage Restoration"
        }
      },
      {
        "@type": "Offer",
        "@id": "https://pauldavis.com/[city]/fire-damage-restoration/",
        "itemOffered": {
          "@type": "Service",
          "name": "Fire Damage Restoration"
        }
      }
    ]
  }
}

Service pages need Article + Service + FAQPage + HowTo (when applicable).

When you implement this at scale across 3,600+ pages with consistent, accurate data, you’re giving Google a machine-readable map of every franchise location and every service offering. That’s how you dominate Local Pack results and organic search simultaneously.

Priority 5: Google Business Profile Velocity (Ongoing)

The Local Pack wins happen here.

For every franchise location:

• Weekly GBP posts — Real posts, not automated junk. Project summaries with before/after photos, seasonal preparedness tips, team spotlights. Google’s algorithm visibly rewards active, engaged profiles.
• Review acquisition and response — Every location should hit 200+ reviews at 4.8+ stars within 12 months. SMS review request 2 hours post-completion, email 24 hours later. Respond to every review within 24 hours. This is the #1 Local Pack ranking factor after proximity.
• Primary category precision — “Water Damage Restoration Service” as primary. Secondary categories should reflect the strongest service mix for that region.
• Photo pipeline — 50+ geotagged photos per location updated monthly. Team, equipment, completed projects, office, vehicles. Google prioritizes profiles with fresh, diverse visual content.

Priority 6: Answer Engine Optimization for the AI Age (Days 60-120)

Google AI Overviews now dominate informational restoration queries. If your content isn’t structured to be cited, you’re invisible.

Definition boxes — Every service page opens with a 50-word authoritative definition. “Water damage restoration is the professional process of returning a property to its pre-loss condition following water intrusion from flooding, burst pipes, or precipitation. It encompasses emergency water extraction, structural assessment and documentation, industrial-grade dehumidification, antimicrobial treatment, and full restoration of affected materials.”

Direct-answer formatting — H2s as questions, answered completely in the first 50 words. “How much does water damage restoration cost? The average cost ranges from $2,000 for minor localized damage to $25,000+ for significant structural involvement, with most homeowners paying $5,000-$15,000. Your final cost depends on the square footage affected, severity of damage, materials involved, and necessary structural repairs.”

Comparison tables — “Water Mitigation vs. Water Restoration: Key Differences.” Side-by-side comparison of timeline, cost, scope, and outcomes.

Numbered process lists — “The 5 Stages of Water Damage Restoration: 1. Emergency Response and Assessment, 2. Water Extraction and Removal, 3. Drying and Dehumidification, 4. Cleaning, Sanitizing, and Antimicrobial Treatment, 5. Restoration and Reconstruction.” This format wins HowTo rich results and AI Overview citations.

Priority 7: The PPC Dependency: From $665K Spike Back to Baseline (Immediate)

The November 2025 PPC spike to $665,600/month tells a clear story: organic pipeline broke, paid ads compensated.

Here’s the math:

• October 2025: $484,200/month organic value, $49K PPC spend. Healthy ratio.
• November 2025: $484,300/month organic value, $665,600 PPC spend. Panic mode — the algorithms changed mid-month and they flooded with paid to keep revenue up.
• Current: $952,800/month organic value (February 2026), $206,100 PPC spend. Recovery mode, but still elevated PPC.

The strategic move isn’t to cut PPC cold turkey. It’s to systematically shift budget back to organic as rankings recover:

• Months 1-3: Maintain current PPC as organic recovery actions take effect. Target high-intent paid keywords that should be ranking organically but aren’t.
• Months 4-6: As location pages recover and start ranking, reduce PPC spend by 20-30% on those keywords and reinvest savings into content creation.
• Months 6-12: If organic recovery hits 60%+ of the pre-November level, reduce PPC spend by another 50%.

The goal: In 12 months, get back to a $50K-75K/month PPC baseline (for new market testing and seasonal peaks) while organic carries the core demand.

That $206K/month in current PPC spend? Reinvested in organic SEO gives you a 8-12 month payoff at which point that traffic is free for the next 5 years.

Why Paul Davis’s Recovery is Easier Than 911 Restoration’s Rebuild

Here’s the critical difference:

911 Restoration peaked at 4,466 keywords in July 2024. By March 2025 when we wrote the playbook, they were down to 3,306. Now (February 2026) they’re at 816. They’ve been declining for 20+ months. The recovery path is long.

Paul Davis peaked at 50,339 keywords in October 2025 — last year. They were still growing in September. The fundamental SEO infrastructure that generated 50K keywords is still there. The content is still there. The domain authority is still there (54, up from 51 in March).

The problem is fixable because the foundation is recent and sound. It’s not a rebuild. It’s a bounce-back.

With the 7-step strategy above, here’s what I’d expect:

• Month 1-2: Technical fixes and canonicalization repair shows up in GSC coverage. Expect 500-1,000 re-indexed pages.
• Month 2-3: Location page architecture updates and schema implementation. Expect rankings to improve on the most valuable pages first.
• Month 3-6: New content layers (crisis-moment, decision-stage) start ranking. Keywords begin recovering. Conservative estimate: 35,000-40,000 keywords by June.
• Month 6-12: Full content architecture matures. Location pages reinforce each other through internal linking. Authority content earns backlinks. Expect 45,000-50,000 keywords recovered.

That trajectory puts Paul Davis back to $450K+/month organic value within 12 months, which means cutting PPC spend from $206K to $50-75K and freeing up $150K+/month in marketing budget that can be reinvested in growth.

The Playbook Works Because Paul Davis Proved It Works

The reason I’m confident in this recovery isn’t theory. It’s data. Paul Davis demonstrated they could execute SEO at scale — they grew from 39K to 50K keywords over eight months. That’s not luck. That’s a team running a good playbook.

The November collapse wasn’t a signal that the playbook failed. It was a signal that something broke in execution — a technical issue, a structural change, an algorithm shift.

But the foundation is there. The domain authority is there. The franchise network is there. All that’s missing is the diagnostic (days 1-3), the fix (days 4-30), and then doubling down on what already works (months 2-12).

I’ve built the systems to execute this at franchise scale — the AI-powered content pipelines, the schema automation, the GEO optimization frameworks. And honestly? Watching a company that was actually winning bounce back is far more satisfying than watching a company rebuild from 800 keywords.

Frequently Asked Questions

What caused Paul Davis Restoration’s 54% keyword drop in December 2025?

Based on the data pattern — a collapse from 50K to 23K keywords in a single month, combined with a spike in PPC spending — the most likely causes are a location page architectural change without proper redirects, a technical indexation issue (robots.txt, noindex tag, or CDN misconfiguration), or the November 2025 Google Core Update hitting thin location pages specifically. The best way to confirm is through a 72-hour audit of GSC coverage data (checking when “Excluded – currently not indexed” spiked) and a URL crawl to identify redirect errors, orphaned pages, or canonicalization issues.

Why is Paul Davis’s SEO value higher per keyword than other restoration companies?

Paul Davis has an estimated SEO value of $43/keyword ($952,800 ÷ 22,190 keywords in February 2026), compared to SERVPRO’s $33/keyword. This suggests Paul Davis is ranking for higher-value, higher-intent keywords — likely more commercial terms and geo-modified queries rather than informational content. It’s a quality-over-quantity advantage: fewer keywords, but more profitable ones. This is actually the ideal position for recovery, since restoring 5,000 high-value keywords is more profitable than restoring 20,000 low-value ones.

How should Paul Davis balance PPC spending during SEO recovery?

Don’t cut PPC immediately — that leaves money on the table and risks losing customers to competitors during the recovery window. Instead, maintain current PPC baseline (around $206K/month) during the first 60-90 days of recovery actions, then systematically shift budget to organic as rankings improve. A realistic timeline: reduce PPC by 20-30% by month 6 (when organic is recovering), then by another 50% by month 12 (when organic has achieved 60%+ recovery). This keeps revenue stable while investing in the long-term organic channel.

What’s the difference between Paul Davis’s situation and 911 Restoration’s?

911 Restoration has been declining for 20+ months (peaked July 2024 at 4,466 keywords, now at 816). It’s a comprehensive, systemic failure requiring a full rebuild. Paul Davis peaked in October 2025 (50,339 keywords) and collapsed sharply in November/December — suggesting a fixable technical or structural issue rather than a fundamental SEO failure. Paul Davis’s recovery is faster and more straightforward because the foundation (domain authority, content corpus, franchise network) is recent and proven to work. It’s a bounce-back, not a rebuild.

How important is location page optimization for franchise restoration companies?

It’s the engine of the entire strategy. If Paul Davis has 600 franchises across 6 core services, that’s 3,600+ location-service pages. A well-optimized location page can rank for 15-40 related keywords through local modifiers, long-tail variants, and service-specific searches. The math: 3,600 pages × 15 keywords average = 54,000 potential ranked keywords. Paul Davis currently has 22,190, meaning they have capacity for 32,000+ additional keyword rankings just by optimizing what exists. Location pages are where restoration companies win.

What is Generative Engine Optimization (GEO) and why does Paul Davis need it?

GEO is optimizing content so that AI systems — ChatGPT, Claude, Gemini, Google AI Overviews, Perplexity — cite and recommend your business by name. For restoration, GEO involves entity saturation (consistent brand-attribute associations across the web), factual density (specific claims about IICRC certification, response times, service areas), authoritative citations (EPA, FEMA, IICRC standards), and implementing LLMS.txt to guide AI crawlers. As AI-generated answers increasingly replace traditional search results, GEO becomes as important as traditional SEO. Paul Davis needs GEO to win when someone asks an AI system “who should I call for water damage in Houston?”

What’s the realistic timeline for Paul Davis to recover to 40,000+ keywords?

Based on the severity of the collapse (54% in one month) but the strength of the foundation (recent peak, high domain authority, proven content infrastructure), I’d estimate:

• Month 1-2: Technical fixes and indexation recovery (expect 1,000-2,000 page re-indexing)
• Month 3-6: Location page optimization and new content layers take effect (expect climb from 22K to 35,000-40K keywords)
• Month 6-12: Full architecture maturity and authority building (expect 45,000-50,000 keywords)

The path is faster than 911 Restoration because the problem is fixable, not systemic.

---

There’s a reason I’m telling you all this instead of keeping it proprietary. Paul Davis Restoration was doing it right through most of 2025. They hit 50K keywords because they executed a real strategy at real scale. Then something broke. But broken things can be fixed.

We’re Tygart Media. We build the systems that execute this playbook for restoration companies at franchise scale. We’ve already figured out the location page architecture, the schema automation, the content velocity pipeline, the GEO optimization. And honestly? Helping a company that knows how to execute bounce back is exactly the kind of project we live for.

The data is public. The opportunity is real. And the timeline for recovery is tight — every month without action is another month where competitors gain ground.

Reach out here if you want to have the conversation. Or don’t. But at least you’ll know what’s possible.

(And hey, if you actually do have a water damage emergency while you’re thinking about this, we can recommend a Paul Davis location. We probably know a guy. Actually, at this point, we’ve worked with enough franchises that we definitely know a guy.)