• Commercial SEO · Screaming Frog

Auto Parts Store SEO: When a Buyer Searches for an Article, Not a Word

The promotion of an online auto parts store is based not on texts, but on addresses: a selection by brand, a tree of nodes and a brand page lead to one detail - the search engine sees one product as three pages. Therefore, we start with measuring the catalog, not with semantics. In the B2B catalog of industrial sewing machine parts, we have counted 1,992 addresses in the sitemap: 1,848 cards and 138 categories, no duplicates. And there are zero blocks of structured data for the entire directory. A clean map and invisible goods are easy to get along with.

See how it works
Price
after a free audit
Guarantee
30 days after project sign-off
Measurement of the catalog of spare parts
1,992 addresses: 1,848 product cards and 138 categories
after a free audit
Price
30
Guarantee

days after project sign-off

1,992
Measurement of the catalog of spare parts

addresses: 1,848 product cards and 138 categories

zero blocks of structured data for 1,848 items
Marking on the same directory
Universal Analytics in code, no GA4 - no data collected from July 1, 2023
Analytics
HTTP 200 with empty body, /sitemap_index.xml is 404
Site map in the second measured directory
388.1
The first response from the server

ms at 120.2 KB of markup and 524.8 ms at 295.7 KB

technical part by project, then monthly; 30–120 days
Format
Who it's for

Situations where this service delivers results

Scenario 1 of 5

The catalog is compiled from the prices of suppliers, and you only go to the main one

Items are filled, cards are opened, there are tens of thousands of lines in the site map — and at the same time, the search leads people to two or three categories. The reason is usually mechanical: the page titles have the template "product name - store name" without an article number, and the cards themselves are available from several branches at the same time. The search engine goes around the same thing in a circle.

We'll review your situation in a free audit
Cases

Tasks and results in numbers — all metrics measured by us

B2B catalog of spare parts and components for industrial sewing machines

Task
Understand the state of the directory before working on visibility: how many addresses are actually in it and what the search engine sees from them.
Solution
Traversal of the site map with counting of unique addresses, checking of structured data, language versions, robots.txt and analytics counters.
Result
1,992 addresses in the site map without a single duplicate — 1,848 product cards and 138 categories. First server response 388.1ms, full load 419.5ms with 120.2KB of markup: third lightest document among 16 sites measured. Language versions uk and ru with correct uk-ua, ru-ua and x-default. Defects: zero blocks of structured data on a directory of 1,848 items, no sitemap directive in robots.txt, and Universal Analytics is still connected in the code, disabled by Google on July 1, 2023 - no GA4 property, which means three years of no data collection.

online retail store of heating equipment, catalog of about 3.5 thousand items

Task
Check whether the search engine receives a list of technical product catalog pages with article numbers.
Solution
Calculation of the directory by standard engine search page by page, checking of both standard addresses of the site map, markup and language versions.
Result
3,565 cards in the live catalog — 35 pages of 100 items and 65 for the last one. First response 524.8ms, full load 662.7ms with 295.7KB markup. Defects: /sitemap.xml returns code 200 with an empty body, /sitemap_index.xml — 404, that is, the search engine does not receive a list of pages at all; no block of structured data on the master. What was in order — two language versions with correct hreflang and x-default and connected GA4.
What's included

Complete list of work and what you get as a result

  • We calculate the catalog using two independent methods — a site map and a regular search engine — and fix the starting numbers from which the work is further measured
  • We break down the semantics into three search methods: by article number, by brand-model and by the name of the node - so that each page has its own type, and not the same one
  • We write the template of the page name and h1 with the article and the manufacturer's number for the entire catalog: the number gets into the header automatically with each new product
  • We glue the addresses of one product: one is the main one, the rest of the paths are closed by a canonical link or a 301 redirect - the weight stops being shared between copies
  • We make landing pages based on the connection "node + brand + model" instead of a bare output of the filter: what people really type appears in the search
  • We put the Product markup with sku, mpn, brand, price and availability, and on the way to the card — BreadcrumbList: the page becomes suitable for an extended snippet
  • We write a rule for items removed from sale, so that the bot does not spend traversal on thousands of empty cards, and the product address experienced a pause in delivery
  • We check whether the data is collected at all: without a working counter, no figure in the following reports will be verified
  • Every month we give a report with the dynamics of indexing and a list of what has been done - without interest, which we did not measure
When this service isn't right

What's not included — so there are no surprises at delivery

  • Filling the catalog with goods and feeds prices of suppliers
  • Purchase or connection of the cross-number database
  • Contextual advertising and marketplaces
  • Promises of specific positions for specific requests - positions are managed by the searcher, not the contractor
Process steps

Transparent stages with approval at every step

Total duration:30–120 days

  1. Measurement and analysis of the catalog

    3-5 working days

    We count the addresses using the site map and standard search engine, compare the discrepancy, build a map of the paths to one card. The output is a document with numbers from which everything else is calculated.

  2. Pasting addresses

    7-12 working days

    We choose the main address of the product, close the rest of the paths with a canonical link or redirect, clean the site map and the rules for processing the filter parameters. We warn you in advance: the number of pages in the index will initially drop, and this is an expected course of events, not a failure.

  3. Article in headings and markup

    5-10 working days

    Title and h1 template with OEM number and manufacturer number for the entire catalog, sku, mpn, brand, price and availability fields in the Product markup, BreadcrumbList on the way to the card. We mark only what is really on the page.

  4. Landing under the node and model

    10-20 working days

    Request groups "node + brand + model" receive their own addresses with a separate header and their own text, rather than remaining in the filter state. We start with the demand that is visible in Search Console, and not with all possible combinations - otherwise it is just new takes.

  5. Rule for out-of-stock items

    3-5 working days

    We record what happens to the card when there is no detail: the status is honestly given in the markup, the address is saved, the position is hidden when the catalog is issued. We negotiate with your engine, not with theory.

  6. Monthly cycle

    from 30 days

    Next, on a monthly basis: expansion of semantics, new landings, indexing control, a report with dynamics and a list of what has been done. You can stop at any month.

Free measurement of the spare parts catalog

The parts store is almost always bigger than the owner thinks: there are several ways to each card — through the brand, through the node, through the brand. We count how many addresses you actually have and show which ones are competing with each other.

What we measure

  • The actual size of the directoryWe count in two ways — with a site map and regular search engine — and compare. The second method looks like this: an empty query in the search returns the catalog page by page, and in one of the measured stores it gave 35 pages of 100 items plus 65 for the last one, that is, 3,565 cards.
  • Duplication through selectionHow many addresses does the chain brand → model → year → engine generate and how many of them lead to the same part.
  • The article is in the title and in the markupDoes the OEM number and manufacturer number appear in the title, in the h1 and in the sku field of the structured data. The article is searched for most often - and that is where it is most often not found.
  • Analogues and crossesIs there a list of substitutes on the card and is it available to the search engine as text, not just clickable.
  • Availability statusHow many cards are given "out of stock" and what happens to them next - they stay, hide or disappear along with the address.
  • Does the sitemap work at all?We check not the existence of the file, but its content. In the measured technical product store, /sitemap.xml returns code 200 with an empty body, and /sitemap_index.xml returns 404: the search engine does not receive the list of pages at all.
  • Is data collectedThe B2B parts catalog we measured still has Universal Analytics in the code, turned off by Google on July 1, 2023, and no GA4 property at all — three years of no numbers.

What you get

  • Two independent numbers of catalog size and the difference between them in percentage.
  • List of mechanics that multiply the addresses of one product.
  • Report on the card: what is available from the articles and markings, what is not.
  • Conversation for 30 minutes on the document.

Timeline: 3-5 working days

Why is it free

Because without measuring, we cannot distinguish a catalog with two thousand products from a catalog with two thousand products and forty thousand addresses. This is a different job, and it is not fair to guess it in advance.

What's next

After measuring, we show what gives the effect first: in most spare parts catalogs, it is not texts, but the pasting of addresses and the article number in the page title.

Short form: your contact and site URL

  • Contract, act and 30-day warranty

    Every project gets a written contract: scope, deadlines, amount, acceptance procedure. After delivery — act and invoice, then 30 calendar days of warranty.

  • Sole proprietor & bank transfer

    The contractor is a registered sole proprietor. Payment by invoice with closing documents.

  • Rights & access — yours

    Code, design and materials transfer to you after full payment. Domain, hosting, repository and analytics are registered to you.

  • Client portal instead of email chains

    During the project you get access to a portal: contracts, invoices, acts and project status in one place.

  • European clients

    Among our work — projects for Norway, Bulgaria, Moldova and Spain.

  • Verifiable numbers

    Every case in the portfolio comes with a link to a live site and a technical measurement.

  • Audit first, then pricing

    There is no price list on the site intentionally: the scope of the same work differs multiples between clients.

  • We say "no" when unsure

    If the task isn't ours or the deadline is unrealistic — we tell you upfront.

Did not find your case?

Describe how it works on your side — we will tell you whether “SEO for auto parts store” fits and what it means in your situation. No brief and no call: one question, one answer.

What affects the price

Why two seemingly identical tasks are priced differently

  • How many paths lead to one cardThe main multiplier of the volume of work. A directory sorted by make, model, and year gives more addresses than a directory ten times larger than an unsorted directory. First, we analyze which paths remain independent pages, and which are glued together.
  • Where does the catalog come from?If the supplier's price is updated every week and the product addresses are new every time, the work begins with fixing the addresses. A stable product identifier in the import is a condition without which the rest will be reset to zero by the next update.
  • Are there article numbers in the data?When the OEM number and the manufacturer number are separate fields, the header template and markup are done once for the entire catalog. When the number is sewn inside the name of the product mixed with the description, it must be obtained by parsing the line, and this is a separate volume.
  • Share of items out of stockThousands of empty cards eat up the bypass and pull the decision at the engine level: hide from the issue of the catalog, leave available at the direct address or combine in a group. The larger the share, the more expensive the error.
  • Status of analytics and accessIf the meter is dead or there is no access to Search Console, the first month is spent getting the numbers. We measured a catalog where data had not been collected for three years: the dynamics were calculated from zero.
  • Language versionsThe two languages ​​double the number of addresses and add match checking in the sitemap. In the measured parts catalog, this was done correctly - the exception rather than the norm.
Technologies & integrations

What we build on and what it connects to

Stack

  • Screaming Frog — directory traversal from the side of the bot; on hundreds of thousands of addresses, it takes hours and requires a speed limit
  • Bypassing sitemap.xml with a count of unique addresses — shows the gap between the declared and real directory; is true and incomplete
  • Standard search engine as a second counter - counts what the buyer sees, without hidden items
  • Google Search Console is the source of the truth about indexing; delayed and sampled data do not provide daily dynamics
  • rel=canonical and 301-redirects — gluing of addresses; canonical is just a hint, redirect is tougher, but breaks other people's links
  • JSON-LD Product / Offer / BreadcrumbList — makes price and availability machine-readable; snippet does not guarantee, markup non-existent leads to sanctions
  • Rules of filter parameters — remove the tail of addresses; close the excess - you will lose boarding
  • GA4 — without it, there is nothing to compare the reports with; shows the trend, not the box office

Integrations

  • Google Search Console
  • GA4
  • Google Merchant Center
  • Rozetka
  • Prom
Own cards with articles against trading only through the platform

How this option differs from the alternative

Search by articleworks if the number is in the header of the page and in the sku markup field - this is the main work
Who owns the address?the address is yours: you change the gluing rules, headings and markup
Selling priceone-time work on the catalog, then the card works without a separate payment for the display
Long tail of articlesthousands of cards work simultaneously; the price of the question is a bypass, not a budget
Start speedchanges in the index are noticeable in 4–8 weeks
What we need from you

We can't start without this — best to prepare in advance

  1. Access to Search Console - without it, you can see the structure, but you can't see what got into the index from it.
  2. Access to the CMS or to your developer: some of the rules live in the engine settings, not in the texts.
  3. A decision that is considered the main address of the product, if the selection, node and brand lead to it at the same time.
  4. A person on your side who confirms the conformity of articles and applicability to models. We do not invent technical data about someone else's assortment.
  5. Export the price in the form in which it enters the engine - with the fields of article, brand and availability.

If something is missing — let us know, we'll help you gather it or do it as a separate task.

FAQ

Most frequently asked questions — with concrete answers

What are they looking for in spare parts — by article number or by name?

And yes, and yes, but these are two different scenarios and two different pages. Anyone who knows the number enters it in full and waits for one card. Who does not know - goes through the brand, model and name of the node, and he needs a page with a selection. The mistake of most directories is that the second scenario is closed only by the state of the filter without its own address: it is not found at all in the search.

We have 200 thousand addresses in the site map. is that a lot?

Depends on how many of them are unique, which is why the first number without the second number means nothing. We measured the B2B catalog of spare parts, where the card contains 1,992 addresses and no duplicates: 1,848 cards and 138 categories. And they measured the opposite — a hardware store, where the sitemap gives code 200 with an empty body, and /sitemap_index.xml returns 404: zero declared addresses with 3,565 live cards.

What to do with out-of-stock items?

Do not delete silently. The item is returned to stock after a month, and each removal and subsequent return erases the address history. Working option: the card remains, the availability is honestly given in the Offer markup, and the item is hidden from the issue of the catalog. What to choose depends on your volume, because on hundreds of thousands of cards it directly affects the circulation.

Do you need texts for the category of spare parts?

Much less than commonly thought. First, a header with an article number, markings and glued addresses - this is almost always missing. The 1,848-item parts catalog we measured did not have a single block of structured data, and analytics had been writing to the property for three years, which Google disabled on July 1, 2023. Texts in such a situation are not the first point, but the third.

We have selection by VIN. will it help the promotion?

Directly - no: for the search engine, it is a closed tool inside the site. Indirectly - yes, he keeps the man and reduces errors in the selection. Our task next to it is different: to make the connections "brand - model - node" have their own addresses with their own header, because they are the ones that are typed in the search. VIN is rarely typed.

Are crossovers and analogues also part of you?

We work with what you already have: we display the list of substitutes in the text of the page so that it can be read by the search engine, and we link the cards together. We do not buy or connect the database of cross numbers - this is a separate license and separate money on the part of the data provider. We also do not verify the correspondence of the numbers ourselves: this is confirmed by your technical specialist.

How long to wait for the result?

The technical part is visible in 4-8 weeks in the Search Console reports: the number of indexed addresses and the frequency of crawling change. Warning in advance: after gluing duplicates, the number of pages in the index initially drops - this is the expected result of work, not a loss. The dynamics of orders depends on the season, availability and prices, which are beyond our control. We do not name the period of "coming to the top" because we have no way to ensure it.

Do you work on a monthly or project basis?

The technical part is a project with a fixed list of works. Then, if necessary, monthly work on semantics and landing. The monthly subscription can be canceled in any month: there is no binding for half a year, materials and access remain with you.

Send a link to the catalog and sitemap address.

The answer is two numbers of your catalog, a list of mechanics who multiply the addresses of one product, and which one is made first. If you work here for a few days, you will hear about it in the first email.

From measured cases1,992 addresses in the site map without a single duplicate

View cases
  • Reply within 2 hours
  • No commitment
  • We work under a contract

There is no price list on the site on purpose: the same work differs several times over between two clients, and a “from” figure explains nothing in that case. First a free audit — we count your pages, duplicates and speed — then we name the sum and the deadline and fix both in the contract.