SEO · Content

How do scattered articles become a sales channel?

Genomed already had search traffic: mid-quality articles spread across subdomains and leading nowhere. I gathered them into one journal with nested URLs and a path from every article to a test. Traffic doubled in eight months, with no ad budget.

Genomed · 2022 to 2024 ·
My role
Senior Product Manager, owner of the web channel
Company
Genomed, the largest genetic testing lab in the CIS, selling tests online
Timeframe
Journal launched October 2022
The constraint
No budget at all: no ads, no SEO agency, not even paid stock images. Search had to bring almost everyone.

If you only read one thing

The site already had articles that ranked, scattered across subdomains with no structure. Instead of starting a new blog, I used them as the flagships of one journal: nested URLs, sections, markup, and a test linked from every article. New visitors went from 79K to 167K a month in eight months, and the journal came to drive around 10% of monthly sales.

The situation

Nobody had worked on Genomed's site systematically before. There were subdomains, standalone pages and a pile of mid-quality SEO articles, each sitting on its own. Some of those articles already pulled search traffic, but they were not connected to each other or to anything the lab sold.

There was no ad budget, so search was the only channel. The question was whether to build traffic from scratch or put to work the traffic that was already arriving.

The fork

Option A

Start a new blog

A clean content strategy from zero, and leave the old articles where they were. Tidy, and months before anything ranks.

Option B · chosen

Make the old articles the flagships

Gather everything into one journal on the main site, with nested URLs, so the pages that already ranked lift the new ones around them.

The articles that already had traffic were the most valuable thing on the site, and nobody was using them. Throwing them away meant starting from zero; putting them at the top of a structure meant every new article started next to a page search engines already trusted.

What shipped

  • One journal, nested URLs. Articles moved to /journal/section/article on the main site, grouped into sections, with redirects from the old addresses and structured data markup.
  • Articles that feed each other. Pages with traffic link to their neighbours in the same section, so new articles get found through old ones.
  • A path to a test from every article. Each article carries a snippet for the relevant test, leading to its landing page and then to checkout.
  • Writing to demand. New articles were planned from Yandex Wordstat: what people search for, how often, in what words. Soon after GPT-3.5 came out, I paid for it out of my own pocket: the model wrote the drafts and doctors checked them before publication.
A section page of the Genomed journal, 'Genetic diseases', with the breadcrumb Genomed › Journal › Genetic diseases above a list of articles.
A section page today. The breadcrumb follows the URL: Genomed › Journal › Genetic diseases.

What happened

79K → 167Knew visitors a month, July 2022 to March 2023
$0ad budget
~10%of monthly sales came through the journal, my estimate

New visitors to genomed.ru, by month thousands · amber = after the journal launch

0
50K
100K
150K
200K
JulAugSepOctNovDecJanFebMar
Same data as a table
MonthNew visitors
Jul 202279,209
Aug 202285,517
Sep 202287,059
Oct 202296,707
Nov 2022105,127
Dec 2022116,744
Jan 2023140,844
Feb 2023146,030
Mar 2023166,989
From my monthly tracking at Genomed. No paid traffic.

Traffic grew every month after the launch. A year and a half later the journal was still carrying the site: in January to November 2024, five of its articles were among the ten most visited pages on genomed.ru, the top one with 78,000 visitors.

The front page of the Genomed journal in October 2026: a featured article and a list of new articles, each tagged with its section.
The journal's front page in October 2026, still publishing into the same sections.

The takeaway

Before writing new content, look at what already ranks. Structure is cheaper than content, and the traffic you already have is the fastest traffic you can grow.

All cases