Case 43. Dealer of Chinese special equipment: from 54 to 568 visits/month with 1–4 eternal links per month (Tier1.shop)

Владислав Бобров (Mdevostator)Article updated: 3 September 2026

Founder of tier1.shop · PBN Specialist, in SEO since 2009

★ 5.0 on Kwork4 912 orders completed1 197 reviews receivedAbout the author →

Telegram channelCases and PBN breakdownsSubscribe →1 802 followersTier1.Shop

Case from the Tier1.shop collection · 2026 · ← all cases

Result in brief: 📈 the start of placements coincided with the site's lowest point — 54 visits/month in August 2023 (before that, organic traffic had been sliding from 134 for eight months). After three years of drip placements — 568 visits/month (peak 584), a 10-fold growth from the starting point and nearly 4 times the best "pre-link" level. Keywords in top-3 became 42 instead of 3–4, and the site set a record (43) already AFTER the March 2026 core update.

Starting point

The site is a dealer of Chinese special equipment: a catalog consisting of categories (excavators, loaders, cranes, rollers, graders) and cards for specific models. The niche is competitive: for general queries like "buy front-end loader," both federal equipment marketplaces and dozens of dealers compete.

By August 2023, the site's organic traffic was not just stagnant — it was decreasing: from 134 visits/month in January to 54 in August (Semrush, Russia). There were 3–4 keywords in top-3 and about twenty in top-4–10. In other words, the links were not entering a "growing site" (a common objection to cases), but a site in a prolonged decline — this is important for the honesty of the conclusions.

What was done

From August 25, 2023 — regular placements of permanent links from PBN home pages, 1–4 per month. This mode we call later in the text drip — by analogy with drip irrigation: a few links every month, for years, without large one-time batches (the opposite being a link sprint from case #42, where 550 links were placed over 7 months). The largest monthly batch here was 4 links. Target pages — almost the entire catalog: 39 links to equipment categories, 7 to model cards, 1 to the home page. Anchor texts — category and model-based with the manufacturer brand ("excavators [brand]", "wheel loader [model]"); 37 unique formulations for 47 links.

There were two pauses in the placement schedule. The spring pause (March–May 2024) passed without a trace — traffic in those months continued to grow on the accumulated mass: 118 → 190 → 263. The winter pause (October 2024 — January 2025) coincided with a double Google core update — this episode is analyzed in the graph below.

Dynamics by Semrush (monthly data, Russia)

What is visible on the graph:

• Start at the bottom. The placement dotted line sits exactly at the bottom point of the series — 54 visits/mo. The first months after the start showed a typical lag for links: a noticeable break began in the spring of 2024 (118 → 190 → 263), i.e., after 7–8 months.
• Growth steps. 2024 — acceleration up to 300–435 by autumn. End of 2024 (first red bar — double Google core update in November–December, coinciding with the placement pause): a rollback to 276, returning to 387 by January. Spring 2025 — a plateau of 250–320, with a new acceleration in the second half of the year: 436 in August, 549 in December, peak 584 in January–February 2026.
• March 2026 Core update (second red bar): a drop from 584 to 487 (−17%) — and recovery to 568 over three months while placements continued. In total, over the three-year window of the case, Google rolled out about a dozen core updates; two left a noticeable mark in the series (both drops rolled back in 1–3 months), the site passed the rest without losses.

This growth cannot be explained by seasonality: July 2024 — 369, July 2025 — 319, July 2026 — 568; December 2023 — 139, December 2024 — 276, December 2025 — 549. The same months year-over-year are consistently higher.

Keywords in tops: top-3 grew from 3–4 to 42, top-4–10 — from 18–23 to 65. During the core update, top-3 dipped from 36 to 30 — and by June it updated its historical maximum (43). That is, the update cost the site two months of position rollback, after which growth continued with an increase.

Ahrefs dynamics (daily data, world)

Ahrefs is consistent with Semrush across the entire "link" part of the history: steady growth after the start of placements (1.7k → peak 6.6k estimated visits/mo), a drop during the core update (to 4.5k, −32%) and recovery by July. The agreement on positions is the same: keywords in top-3 (world) became 227 versus 40 at the start of placements, in top-4–10 — 292 versus 114.

One honest caveat (gray bar on the graph): in June 2023, two months BEFORE the first link, the Ahrefs estimate jumped from ~50 to ~1000 in three weeks. Such a step is a recalculation of the estimate by the service (expansion of the keyword database/re-indexing after site edits), not a real organic surge: Semrush showed a drop in the same months. We do NOT attribute this jump to ourselves — we calculate "fold" growth by Semrush (Russia), and use Ahrefs as independent confirmation of the curve's shape after the start.

Breakdown by target pages

We ran all 27 acceptor pages through Semrush (Russia, monthly series): comparing the traffic of each page before its first link and after. The home page does not participate in the slice — for the site root, Semrush provides a series for the entire domain. A yellow mark such as ×7 for 31 mo — anchor duplicate: the same wording was used 7 times over 31 months. Result for 26 pages — honestly, without selecting convenient ones:

"Before" in the traffic column is the maximum page views for the period before its first link; sometimes this is a one-time spike rather than a stable level (this is exactly the case for excavator and roller categories: their peaks of 18 and 8 faded BEFORE the first link to the page).

Yellow mark ×N for M mo — anchor duplicate: the same wording was used N times over M months. All duplicates in the case are repetitions to the same page; the typical interval between repetitions is 3–23 months, the minimum is 1 month (a single case for excavators). There are no cross-duplicates between different pages.

The most striking fragment of the slice is the reach stacker category: one link in January 2024 — and growth from 27 to ~69 visits/month, with the current value being a page record for all time; for its main query, the category is in the top-3, and for the transactional "buy" — in the top-1. And the most sobering is the excavator category: the most links (9 over three years), but almost no own traffic according to Semrush. Why the case is still successful is explained below.

It looks like a paradox: the domain grew 10 times, but more than half of the acceptor pages individually "did not fire." Let's analyze what is visible from the data — and what we can and cannot assert.

Fact 1. Acceptor pages collectively provide ~170 out of 568 visits/month. Approximately two-thirds of the growth is collected by the home page and pages to which no links were placed at all — the effect was distributed across the entire site, rather than concentrated at the link entry points.

Fact 2. Traffic accelerations over time coincide with the growth in the number of pages and keys in the top-3. First wave: from April to September 2024, keys in the top-3 increased to 18 instead of 4 — and in these same months, traffic broke away for the first time (118 → 369). Second wave: by March 2026, the top-3 grew from 26 to 36 — and traffic reached a peak of 584.

How to interpret this — to be honest: the mechanism inside Google cannot be proven from the outside. But our working observation on client projects — and this case is consistent with it — sounds like this: the more site pages are established in the top-1–3, even for queries with low frequency, the more confidently the entire site begins to grow. That is exactly the case here: categories that entered the top-1–3 for their queries (reach stackers, recyclers, wheel loaders) are small in terms of traffic, but their establishment in the top went hand in hand with the growth of the entire domain. Practical conclusion for link planning: links to internal pages strengthen the site as a whole, and the return may manifest not where the link was placed.

Disclaimer on accuracy: for pages with traffic in single-digit visits, Semrush only sees keys from its own database; part of the actual long tail does not fall into such rows.

The role of anchor text

All anchors are commercial category and model-based ("excavators [brand]", "truck cranes [brand]", "wheel loader [model]"). There are moderate repetitions: the most frequent anchor appears 6 times — but stretched over three years, meaning fewer than two repetitions per year. In a drip-feed mode (1–4 links/month), such repetitions look natural and did not hinder growth — this is a useful guideline for those who fear anchor "over-optimization": the issue is not the repetition itself, but the concentration over time.

We separately tested the opposite hypothesis — "did Google penalize the pages for repeating anchors?". Let's immediately establish the framework: our conclusion applies ONLY to the repetition mode of this case — 2–7 repetitions per page, stretched over 7–31 months, with a typical interval of 3–23 months between repetitions (minimum — one month, in a single instance; all quantities and terms are in the table above). At such a density, there are no signs of penalization in the data. Firstly, duplicates exist even for the best pages of the slice: the wheel loader category with a repetition of ×3 grew the most. Secondly, for the "suspected" pages, the timing is reversed: for the excavator and roller categories, page traffic faded BEFORE the first link to them, and for the excavator category, the top-10 appeared FOR THE FIRST TIME precisely during the densest series of repetitions: before October 2025, the page had not a single key in the top-10 (there was nothing to lose), by July 2026 — two, and it stands in the top-10 for one of the category queries. This is the dynamics of a slow climb in a competitive cluster, not a filter. The same anchor pattern produced both the best and worst pages — meaning, at this density, it is not the cause.

Important warning from our practice: this does not mean that anchor duplicates are safe in principle. On other projects, we have seen penalization for duplicates — when identical anchors were placed frequently and in clusters. Judging by the combination of observations, it is not the repetition itself that is critical, but its concentration in time: a repetition once every few months looks natural, a series of identical anchors over a short period — like a templated purchase. When repeating the strategy of this case, repeat its pace as well.

Conclusions from the case

• The drip strategy works on a long horizon. 47 links over 35 months — averaging fewer than two per month — provided a 10x growth from the starting point. For comparison: in case #42, a similar multiple was achieved by a sprint of 550 links over 7 months. A small monthly budget is not a sentence if you work regularly and for a long time.
• Links reversed the falling trend. The site had been losing organic traffic for eight months and started from the lowest point. This removes the main objection that "the site would have grown on its own": before the links, it was moving in the opposite direction.
• The effect arrives with a lag of 7–8 months. Someone who had canceled placements after three months "because it doesn't work" would not have waited for the reversal. We see the same lag with the same client on a second project — case #44.
• A Core update is not a catastrophe with a live link profile. A drop of −17% rolled back in three months, and keywords in the top-3 broke records immediately after the update. This is already the fourth case we have analyzed in 2026 with the same pattern (#40, #41, #42 — and this one).
• Pages that reached the top-3 pull the entire site up with them. A page-by-page slice showed: some acceptor pages established themselves in the top-1–3 for their queries, and the growth waves of the entire domain coincided in time exactly with this establishment — despite the fact that two-thirds of the growth was collected by pages to which no links were placed. The link budget works for the site as a whole: links bring pages into the top, and pages in the top pull the rest of the site.
• Pauses do not reset the result. During two placement pauses, traffic did not fall — permanent links continued to work. This effect is analyzed in more detail in case #44 — there, for the same client, the pause lasted 9 months, and traffic grew 2.2 times during it.

Verifiability

The client's domain is hidden due to the privacy policy, so instead of a link to the site — screenshots from analytics services (traffic estimates in interfaces and on our graphs may differ: services have different keyword databases and regions; however, the shape of the curve coincides).

Semrush Domain Overview (2-year window, world): organic traffic and keywords in top-3 / top-4–10 — upward trend, April drop and recovery, as on our graphs.

Ahrefs, organic traffic for all time (estimate of monthly visits, weekly grid): plateau until 2023, growth in the placement window, peak in early 2026. White badges Ⓖ — Google algorithm updates, green marks — site content edits (recorded by Ahrefs).

Ahrefs, keywords by position (buckets 1–3, 4–10, and 11–20): growth of all three buckets in the placement window.

How we measured

The number of links, dates, and anchor text — from the Tier1.shop order database (only completed placements with the placement date). Traffic and positions — Semrush (Russia, monthly points) and Ahrefs (world, daily points), API/interface exports in raw form; graphs are built from these exports without smoothing. Attributing all growth solely to links is incorrect — content edits also influenced the site (green marks on the Ahrefs chart); but the start from a multi-month minimum, the effect lag, and the consistency of two independent sources give links the primary role in the reversal.

*Costs are calculated at current Tier1.shop prices; the client was placed at the prices of their time — the order of magnitude is the same.