The audit and Action Plan answer "what's broken and what do we fix first." This playbook answers the next question: once the fixes land, what routine turns a 36/100 site into the shop Katy drivers find when they search for an oil change, brakes, a transmission, or a state inspection? The Playbook Action Plan is the line-by-line checklist that tracks every playbook step as Done, Blocked, or Not started.

Northwest Express Auto Care starts with real assets: a shop founded in 2017, a Google profile with 4.5 stars from 268 reviews, a 36-month / 36,000-mile TechNet nationwide warranty, and 87 published pages. Semrush (US, today) shows 42 organic keywords and about 98 visits a month. It's #1 for the shop's own name and #1 for "transmission flush katy." But the big local searches have no ranking page at all: "oil change katy" (320/mo), "auto repair katy" (210/mo), "tire shop katy" (170/mo), "transmission repair katy" (110/mo, $8.04 CPC), and "brake repair katy" (70/mo, $9.57 CPC). Two problems sit above everything else. Google, Bing, and Brave all send "website" clicks to the duplicate site nwautocaretx.com (C2), and the page sitemap crashes, so 0 of 87 pages are in a working sitemap (C3). This playbook applies Wabu Web's AI Search Ranking Playbook v2 to a single-location, family-founded auto repair and tire shop.

How Search Decides Who to Show

Someone in Katy whose brakes are grinding sees three kinds of results, and each one is decided differently. The plan has to cover all three.

SurfaceTypical searchWhat decides itWhere Northwest Express stands today
Google Map Pack (the 3 map results)"auto repair near me" (301,000/mo US), "state inspection near me" (49,500/mo US), "oil change katy" (320/mo)Relevance (category, services, website content), distance to the searcher, and prominence (reviews, rating, mentions around the web). Google documents these three factors.Good profile, wrong link. 4.5 / 268 reviews with owner replies, and Semrush sees the shop in 27 local packs. But the listing is named "Northwest Express Care" and its website button goes to nwautocaretx.com (C2).
Organic results (blue links)"auto repair katy" (210), "oil change katy" (320), "tire shop katy" (170), "transmission repair katy" (110)One clear page per search (service + Katy in the title and H1), useful content, links, a site Google can crawl (sitemap, Search Console), and a page that loads on a phoneWeak on head terms. 8 keywords in the top 10, mostly the brand name and car-make pages. About 10 pairs of duplicate service pages split the signal (H1), the page sitemap crashes (C3), there's no Search Console access (C4), and mobile LCP is 34.6 s (H4).
AI answers (Google AI Overviews / AI Mode, ChatGPT, Gemini, Perplexity, Claude)"best auto repair shop in Katy TX", "how much does a transmission replacement cost", "where can I get a state inspection in Katy"Short, direct passages that answer the question, one consistent business identity across the web, and third-party proof (reviews, directories, Reddit, YouTube). A top-10 rank helps, but it no longer guarantees a citation.Conflicting signals. Bing (ChatGPT's main source) shows the duplicate site and a Yelp 2.5 / 15 rating. Brave (Claude's source) summarizes "third-generation, 45 years" and "Sun 9–1" hours. The hours in the schema don't match the GBP, and there are no sameAs links (H3, H6).

The good news: the content is already server-rendered, so AI crawlers can read the page copy without running JavaScript. Semrush also already sees 5 of the shop's keywords showing an AI Overview. The job is to make every engine agree on one business, one website, and one set of facts, and then give each money search its own strong page.

Strategy Pillars

1. Close the leaks firstWeek 1

Take the public 1.8 GB site backup off the web and rotate every password (C1). Confirm or remove the two unknown admin accounts (H2). Then the biggest traffic leak: point the Google listing's website button back to nweautocare.com and settle the future of nwautocaretx.com with the client (C2). Today every "website" click from Maps, Bing, and Brave goes to a site Wabu doesn't control.

2. Let Google and Bing see the real siteWeek 1

Fix the page sitemap's PHP fatal error (C3). Get owner access to the Search Console property someone already verified by DNS, submit the sitemap, and set up Bing Webmaster Tools and IndexNow (C4). Bing is the back door into ChatGPT, and today it shows the duplicate site first.

3. One page per service, not twoWeeks 1–3

The May 2026 "-katy-tx" pages and the July 2026 pages compete with each other (oil change, brakes, transmission, inspections ×3, alignment, AC, diagnostics, fluid flush, tire shop, module programming). 301 each old page into its live twin so its rankings carry over: "transmission flush katy" #1 and "shocks and struts replacement katy" #9 both sit on old URLs today (H1). Then fix H1s and titles (M2) and link the orphaned Coupons, Financing, Warranty, and Vehicles pages (M1).

4. One business, everywhereWeeks 2–6

One name, one set of hours (GBP says Mon–Fri 8–6; schema, Yelp, and YellowPages say 8–7; Brave says Sunday 9–1), one founding story, and one phone number. Nextdoor lists 832-683-4478. Fix the schema (H3), then the directories (H6). Add sameAs links to 8 or more real profiles. Claim Yelp and start answering reviews there (M8).

5. Answers and proof only this shop hasMonths 1–6

Answer-first FAQ blocks, specials written as text, and a transmission replacement cost and financing page tied to the Facebook transmission campaign (M4). Name the people: David the service manager, the techs, and real ASE certifications instead of "NWE-certified" (M6). Feature the TechNet warranty on every service page, and stop the reviews widget from showing a 1-star review sitewide (H5).

6. A phone-first site, measuredMonths 1–2, then ongoing

Mobile Lighthouse is 16, with 2.1 MB of inline CSS on every page, a 1.1 MB PNG, and two GA4 tags (H4). The phone and Book Service buttons disappear on phones (H7). Then comes one GA4 property with events for call clicks, Shopmonkey bookings, and financing clicks, plus an AI-assistant channel (M5). That way every step above can be judged by booked repairs, not guesses.

The Differentiator to Build Around

A family shop that put its home on the line in 2017, and backs its work for 3 years
The Katy results for "auto repair katy" and "oil change katy" are crowded with chains: Take 5, Valvoline, Jiffy Lube, Midas, Goodyear, Express Oil. Northwest Express Auto Care's About page tells a story no chain can copy: founded in 2017 after "witnessing a beloved local repair shop lose its personal touch under corporate ownership," it was financed by "putting our home on the line." It also offers a 36-month / 36,000-mile TechNet nationwide warranty, and its reviewers name the people: "David there service manager the office staff are all great people," and one customer has used the shop "about 8 years." That's the "information gain" the playbook says competitors can't copy. Every rebuilt page should carry it: the shop's name in the first sentence, the warranty in a dedicated block, the named team with photos, and review quotes that match the service. The details need confirming first: owner names, ASE certifications, and which experience story is accurate ("since 2017" vs. Nextdoor's "30 years, second generation" vs. Brave's "third-generation, 45 years"). See open questions 5 and 6.

Progress Log

2026-10-03Work in flight on audit day
  • Privacy Policy page published at /privacy-policy/ (WordPress page 3, set as the site's privacy page). Needed for the Facebook lead form, and good hygiene for a site that links to financing applications.
  • Facebook instant form "Transmission Replacement Quote + Financing - Oct 2026" published. The Leads campaign is next. The transmission cluster on the Keywords page is timed to support it.
  • Facebook ad budget set to $16.13/day (about $500/mo).
  • Pending from the client: new shop photos (they replace stock images in M3/M6), and Shopmonkey credentials so we can see whether the scheduler can be integrated more deeply.
  • Not yet started: no audit fixes have been applied. C1 (public backup) and C2 (GBP website link) need approval or the client's input first.

Add a dated card here each time a batch of Action Plan items ships, and update the statuses on the Playbook Action Plan in the same session.

Baseline → Targets

MetricBaseline (2026-10-03)90-Day Target (end Dec 2026)180-Day Target (end Mar 2027)
Organic keywords (Semrush, US)4280+150+
Est. organic traffic/mo~98175+350+
Keywords in the organic top 10815+30+
"oil change katy" (320) / "oil change katy tx" (210)Not rankingTop 30Top 10
"auto repair katy" (210) / "auto repair katy tx" (140)Not rankingTop 30Top 10
"transmission repair katy" (110, $8.04 CPC)Not ranking (#1 for "transmission flush katy" on an old URL)Top 30, flush #1 kept after the 301Top 10
"brake repair katy" (70, $9.57 CPC)Not rankingTop 30Top 10
"tire shop katy" (170) / "tires katy tx" (90)Not rankingTop 30Top 15
Make pages in the top 10 (Lincoln, Cadillac, Jeep, GMC…)2 (Lincoln #9, Cadillac #9)58+
Google reviews268 (4.5★)300+, every one answered340+
Yelp rating2.5★ (15), unclaimed status unknownClaimed, every review answered20+ reviews, trending up
GBP / Bing / Brave website linknwautocaretx.comnweautocare.com everywhereDuplicate site retired or 301'd
Pages in a working sitemap0 of 87 (fatal error)All live pages, duplicates removedMonitored monthly in GSC
Search Console / Bing accessNone for WabuBoth verified, sitemap submittedMonthly reporting
Leads tracked (GA4)Two GA4 tags, no events confirmedOne property; calls, bookings, and financing trackedMonthly leads by channel
AI answer presence (quarterly check)Not yet measured; Bing/Brave cite the wrong siteBaseline recorded, correct site citedNamed in 3+ of 8 test prompts
Mobile Lighthouse performance (home)16 (LCP 34.6 s, CLS 0.754)50+70+ (LCP < 4 s)

Targets are directional, not guarantees. The head terms are contested by national chains and long-established local shops: katyautocare.com has 223 keywords and about 1,168 visits a month, and dillonsautomotive.com has 638 keywords and about 1,245 visits. The "near me" searches are decided in the Map Pack, which Semrush doesn't track by position. Judge progress by booked repairs first (calls, Shopmonkey bookings, and financing applications in GA4), then by Search Console impressions, then by these rank numbers.

The 6-Month Arc

Month 1 (October): Foundation
Every Critical item, plus the duplicate cleanup
  • Remove the public backup; rotate DB password, salts, and admin passwords (C1); confirm or remove unknown admins (H2)
  • GBP website → nweautocare.com; client call about nwautocaretx.com (C2)
  • Fix the page sitemap (C3); Search Console owner access, Bing Webmaster Tools, and IndexNow (C4)
  • 301 the duplicate and legacy pages; delete Hello World (H1); filter the reviews widget to 4–5 stars (H5)
  • Schema hours/sameAs/geo fixes (H3); one GA4 property with events (M5)
Months 2-3 (Nov–Dec): Growth
Money pages, speed, and one identity
  • Rebuild /transmission-service/ and /oil-change/ to the Service Page Template; transmission cost and financing content for the Facebook campaign (M4)
  • Mobile speed and layout: the inline CSS, WebP images, LCP, CLS, and a visible tap-to-call (H4, H7)
  • New H1s/titles on the main service pages (M2); link the orphaned money pages (M1)
  • Directory cleanup: Yelp, YellowPages, Nextdoor phone typo, Facebook, Bing Places, Apple Business Connect (H6, M8)
  • Alt text on the nav-linked pages first (M3)
Months 4-6 (Jan–Mar): Authority
Proof, depth, and the first re-audit
  • Named team, ASE proof, and the 2017 story as a factual About page (M6)
  • Brake, tire shop, and state inspection pages rebuilt; make pages that already rank improved (M7)
  • AI-crawler mu-plugin + llms.txt (L1); clarify NW Mobile Auto Care (L3)
  • First Reddit/community and YouTube steps (L4)
  • January 2027 quarterly re-audit: fresh Semrush pull and the AI citation check against the baseline above
Ongoing: Sustain
The monthly routine
  • Two posts a month from the content calendar (L2)
  • A review request after every repair; a reply to every Google and Yelp review (M8)
  • A GBP post for each monthly special, linking to the matching page
  • The monthly review (checklist below); a quarterly AI citation check (see Schema & GEO Tactics)

Monthly Review Checklist

  • Leads first: on a phone, tap the phone number and the Book Service button and confirm that both work and show up as GA4 events. A broken booking link is the most expensive failure this site can have.
  • Google listing: search "Northwest Express Auto Care Katy". Check that the website link is still nweautocare.com, and check the name, the hours (Mon–Fri 8–6, Sat 8–4), the rating, the newest review date, and that every new review has an owner reply.
  • Search Console: Page Indexing (new "not indexed" reasons? are old duplicate URLs dropping out?), Performance queries (oil change, brake, transmission, tire, and inspection terms gaining impressions?), Manual Actions / Security Issues, and the submitted-sitemaps list (anything you didn't add? The server had a compromise in September 2026).
  • Sitemap: open /page-sitemap.xml. It should load in seconds and list the live service pages, with no PHP error.
  • GA4: sessions by channel (including an "AI assistants" channel), call clicks, Shopmonkey booking clicks, financing clicks, and top landing pages
  • Semrush: organic keyword count, positions for the terms on the Keywords page, and nwautocaretx.com's keyword count (it should fall once C2 is done)
  • Site hygiene: after any plugin or Divi update, re-test the sitemap and a service page on a phone, run wp core verify-checksums, and confirm that nothing new has appeared in /wp-content/ai1wm-backups/ and that no new admin users exist

How the Playbook Pages Fit Together