After go-live, we audit the store again.

Where the store stands today

Google Analytics has been live since the switch-over, and its purchase tracking matches Shopify order for order (10 of 10 in the period; the 11th Shopify order landed on the day of this report). These are the first real numbers the business has had, and they are the baseline every change below will be measured against.

Source: GA4 property [Brand], Aug 15 to Sep 10, 2026. Traffic starts on Aug 20, the day the domain moved. Revenue is GA4’s figure; Shopify records $4,208 across 11 orders including Sep 11.

What people land on

Half of all visits start on the homepage. The rest land straight on product and category pages from search, which is exactly the pattern a parts catalog should show. The pages Google sends the most people to are the J-series adapter plates, the L15B7 (Civic and Accord 1.5 turbo) head gasket, the head stud category and the turbo manifold category. Three of the top ten landing pages have engagement below 40%, which usually means the page did not answer the question the searcher had: the L15B7 head gasket page (28% engaged, 67 visits) and the B-series turbo manifold page (37%) are the two to fix first, and both have descriptions under 60 words.

Three things the analytics tell us beyond traffic

• Mobile is two thirds of visits and eight of ten orders. Every copy and layout decision should be checked on a phone first.
• On-site search shows what the catalog is not surfacing. The most typed terms are “s2000”, “J series” and “k20 k24”, which are navigation terms, so the platform filters (still to be switched on in Search & Discovery) matter. People also searched for things the store does not stock: intake manifold gaskets, motor mounts, air intakes, Wiseco pistons, clutch for a 2024 Civic Si. That is a free product-range signal.
• Some “direct” traffic is not human. 76 desktop visits from Singapore and 35 from China, each with exactly one visit per user and no engagement, are crawlers. GA4 can filter them once we set a known-bot rule; until then treat direct traffic as roughly 15% overstated.

How well Google has indexed the store

The migration’s redirects and sitemap did their job. Indexing is not the problem; the way the indexed pages are presented is.

Where the store ranks, and who outranks it

Live Google results for the searches that matter most, checked from the United States on September 14. The pattern is consistent: the store owns the swap-adapter searches it invented, and loses everything else, often to shops reselling the client’s own parts.

Sources: live SERPs via DataForSEO (us, en, mobile), search volumes from Ahrefs Keywords Explorer, United States, September 2026.

Five of the ten searches show a dealer ranking above the manufacturer for a part the manufacturer makes. Those dealers have the same photo and a shorter description; they rank because their page names the chassis and the engine in the title, and because their domains carry more links. Both are fixable: the copy work in the next section, and the manufacturer’s natural advantage of being the source of truth (fitment tables, torque specs, dyno data) that a reseller cannot write.

Authority and links

Source: Ahrefs, September 14, 2026, United States. Ahrefs’ index is shallower than Google Analytics: it credits [domain] with 92 organic visits a month where GA4 records about 1,800. Use the table for relative position, not absolute traffic.

• The link profile has eroded. 1,312 domains have linked to [domain] over its life; 432 still do, and only 92 of those are followed links. Super Street’s 27 links, Bimmerforums’ 137 and Grassroots Motorsports’ 5 were all lost between 2019 and 2022 as those sites pruned old threads. What remains is good: MotoIQ (80 followed links from a project-car series), Project Mu, Tein, Fuelly, Hondata.
• The decline started long before the migration. Ahrefs’ estimate of the old BigCommerce site’s organic traffic fell from about 380 visits a month in June 2025 to under 100 by early 2026 and stayed there, as pages with no fitment copy slipped behind dealers and forums. The new store inherits that position; it did not cause it.
• The migration lost no links. Ahrefs finds zero inbound links pointing at a page that now returns an error, so the 320 redirects are doing their job.
• Ballade Sports is the proof point. At domain rating 10, barely above [Brand], it ranks for 912 keywords because every product page names the car and the engine. Authority is not what separates the store from page one; page copy is.
• A cluster of spam “rank” domains started linking in 2025 and 2026. Google ignores these on its own. No disavow is needed, but do not buy links from anyone offering them.

How much search there is to win

This is a small, specific market. The commercial part searches are 10 to 150 a month each, and almost all have a difficulty score of zero, meaning no strong sites compete for them. The volume is in the swap and platform topics, and that is where the blog and the landing pages should aim.

Source: Ahrefs Keywords Explorer, United States, 12-month average, September 2026. Difficulty is on a 0 to 100 scale; 0 means the top 10 contains pages with almost no links.

Technical tag audit

The usual checklist, run on the homepage, a product page and a category page. Shopify and the Horizon theme get most of it right out of the box; the gaps are the ones a stock theme cannot fill for you.

Speed

Lab figures from Lighthouse 12 on a simulated mid-range phone. Real-user data (Chrome UX Report) needs 28 days of traffic at this volume before Google publishes it; re-check in October.

Product titles and descriptions: what we have, and how to improve them

This is the largest single lever. Google ranks product pages on the words on the page, and the words on the page are thin.

What the audit found

• Fitment is implied, not stated. 92 titles carry an engine code, but only 21 say Honda or Acura, and only 83 descriptions do. Searches are phrased “honda k20 connecting rods”, “s2000 head studs”, “l15b7 head gasket kit” (Google’s own autocomplete for these terms). The pages should say the chassis and the years, not just the engine code.
• The copy was pasted from BigCommerce with its formatting. 162 of 207 descriptions carry inline font-size styling from the old editor, 16 are built with line breaks, and only 2 use a real bullet list. Google reads the text fine, but the pages look uneven and the phone layout suffers.
• The best pages already exist and prove the point. The timing chain tensioner (about 190 words, explains the failure it prevents), the F24 rod set (“proven to 1,000 hp, ARP2000 bolts, rated to 10,000 rpm”) and the head stud sets (material, tensile strength, why not L19) get their own text shown in Google instead of the boilerplate. The rest of the catalog needs that treatment.
• Head gaskets are the biggest block of thin pages. 32 gasket listings, most of them one line: bore, thickness, done. Each is a separate URL competing with its siblings for the same search.
• Structured facts already live in the data. Every product carries an engine-platform tag; 23 have thickness, 21 have a horsepower rating and 9 flag “core required” as metafields. That is the skeleton a good description is built on.
• One flagship has no photo. The K-Series cylinder head package, the product the shop is known for, is the only active listing without an image.

AI-assisted versus manual rewriting

We compared three ways of doing the work. The recommendation is the hybrid: the AI drafts from the structured facts and the owner verifies the fitment line, because fitment is the one thing a machine must not guess.

The template every page would follow: a search-ready title (brand spelled one way, engine, part, key spec, under 60 characters), a two-sentence opener that names the engine and chassis, a “Fits” line, a spec list drawn from the metafields, a short paragraph on why it is built the way it is, and a “What is in the box” line. Two examples from the live catalog:

Collections need the same pass

All 55 category descriptions are one sentence (48 to 141 characters), which is why Google swaps in the footer text for category results. A category page should carry 120 to 200 words that name the engines, the chassis and the sub-categories, and a written SEO description of 150 characters. Example for the K-Series collection: “K20 and K24 engine parts for Civic Si, RSX, TSX, Integra and K-swap builds: cylinder head packages, forged rods, head studs, timing chain tensioners and adapter plates, machined in-house since the late 1990s.” Eight collections and the homepage cover 80% of category traffic; write those by hand first, template the rest.

Quick wins, in the order to do them

Each of these is a settings change or a copy task, not a rebuild. Numbered because the order matters: the first four unlock measurement for everything after them.

1. Give us Search Console access, and connect Bing Webmaster Tools

   the owner · 10 minutes · unlocks every later measurement

   Add the Ruby Digital Google account (we will send the address) as a full user on the existing Search Console property, or let us verify by DNS. Bing Webmaster Tools imports from it in one click, and Bing sent the largest order of the period.

2. Write the homepage title and description

   RDA · 15 minutes · Online Store → Preferences

   Proposed title: “[Brand] | Honda & Acura Race Engine Parts, K-Series, B-Series, S2000”. Description: “Cylinder head packages, forged rods, head studs, adapter plates and turbo hardware for Honda and Acura engines, machined and assembled in-house since the late 1990s.”

3. Fix the Facebook link and the Instagram profile

   the owner · 20 minutes

   The footer’s Facebook link goes to an unrelated personal profile in another country. The real page (12,000 followers, shop street address) has a different URL; send it and we will change the theme setting. On Instagram the account is named the owner’s personal name with no bio; rename it [Brand], add a one-line bio and a tagged store link.

4. Install the three Shopify sales channels: Facebook & Instagram, TikTok, Google & YouTube

   RDA with the owner’s logins · 2 hours · free

   Each channel installs its pixel, syncs the catalog and enables tagged products in posts. This is what makes social traffic and ads measurable, and it puts the catalog in Google’s free Shopping listings.

5. Write SEO descriptions for every product and collection

   RDA · 1 day · published by API

   262 descriptions of 150 characters, each stating the part and the fitment. This alone changes what every search result says under the store’s name.

6. Standardize product titles

   RDA · half a day · the owner approves the list

   One brand spelling (we suggest the spelling on the logo), engine and chassis in the title, under 60 characters, and the 11 titles with double spaces cleaned up. Shopify keeps the URLs, so no redirects are needed.

7. Run the hybrid description rewrite

   RDA 2 days, the owner 3 to 4 hours · start with the 40 most-visited products

   Drafts from the template above, fitment lines confirmed in a spreadsheet, published by API with the BigCommerce styling stripped. The head gasket family and the top landing pages go first.

8. Add a reviews app and ask past customers

   RDA · 1 hour setup · Judge.me free tier

   Star ratings appear in Google results once a product has reviews. A one-time email to the 991 subscribed customers from the migration would seed the first hundred.

9. Lighten the homepage and product hero images

   RDA · 1 hour

   Re-export the two hero slides at 1600 px and under 150 KB, set the first slide to load eagerly, and cap product hero requests at the displayed size. Expected to move the mobile performance score from the 50s into the 70s.

10. Tidy the index

    RDA · 30 minutes

    Unpublish the seven empty collections from the online store, add breadcrumb structured data to the existing breadcrumb block, photograph or render the K-Series head package, and close the BigCommerce store once the last invoice is settled.

11. Publish the first four blog posts, and one K-swap landing page

    RDA drafts, the owner reviews · weekly

    The questions Google shows under these searches today: “What are the symptoms of a bad timing chain tensioner?”, “Which K20 head is best for a K24 block?” (a 250-search-a-month topic on its own), “What is needed for a J-series swap?”, “What years have Honda L15 head gasket problems?”, “Do head studs add horsepower?”. Alongside them, one page that owns “k swap” (1,600 searches a month, no strong competitor) and links to every one of the client’s swap parts. Each post becomes the source for the week’s social posts.

Facebook, Instagram and TikTok: the transition plan

The audience exists. It is not connected to the store, and nothing on the store side can measure it. The plan connects the accounts first, builds a repeatable content routine second, and only then spends on ads.

Why the boosted posts did not work before: they had no pixel behind them, so there was no audience to build, nothing to retarget and no way to see which post produced an order. With the channels installed, every campaign reports orders inside Shopify’s Marketing section, and the cost per order is visible from the first week. Total suggested ad spend for the first quarter is roughly $1,500 to $2,000, all of it in the last 60 days.

What to expect

Estimates, not promises. They assume the quick wins ship in September and the content routine holds. The basis is the keyword volumes above and what comparable shops achieve: Ballade Sports, with the same authority as [Brand], ranks for 912 searches; the store ranks for 6 today. Reaching 300 to 500 ranking searches in six months is the target behind the organic row. We will replace these with Search Console figures as soon as we have access.

Needs your input