SEROKUZE
Senior SEO Consultant · 9+ years in search
I run search for companies where the buying decision starts in a SERP: technical audits, indexation architecture, content systems and link building — from the first crawl to the release that fixes it.
I started in search in 2017 on small commercial sites, and spent the years since moving between very different categories: B2B SaaS, e-commerce and marketplaces, cloud and hosting, cybersecurity, fintech and large content projects. More than 40 projects audited; several of them passed a few hundred thousand organic sessions a month while I owned search.
The work is diagnosis first. Server logs, a full crawl, Search Console query-by-page data and rank history — before any opinion about strategy. Then a sequence: what to fix this month, what belongs to this quarter, and what to deliberately ignore.
I read and write code, which is why my tickets go into a sprint without a second meeting: exact selectors, before-and-after markup, acceptance criteria. Most of my work ships through someone else's developers, in English, Ukrainian or German.
Nine years, four categories and one habit that did not change: read the data before forming an opinion.
One person on the whole of search: crawl, texts, links, reporting. Useful in one specific way — nothing could be delegated, so every part of the job had to be understood rather than trusted.
Search in a category where the buyer is an engineer and the catalogue runs to thousands of location, configuration and service URLs. Technical audits, indexation architecture and internal linking built around how infrastructure is actually searched for.
Product and comparison pages in crowded software SERPs, cluster maps for a large content library, and the international versions that come with selling the same product in several markets.
Since then: over 40 audited projects across SaaS, e-commerce and marketplaces, cloud and hosting, cybersecurity, fintech and media. Either an audit with a plan, or full ownership of search alongside the client's own team.
Logs, a full crawl, Search Console query-by-page data and rank history first. Then a sequence — this month, this quarter, and what to deliberately ignore. Every recommendation leaves as a ticket a developer can implement without a second meeting.
Crawl, index and render. Log-file analysis, crawl-budget routing, Core Web Vitals, JS rendering.
Information architecture and internal linking built around how a category is searched, not how the CMS is organised.
Cluster maps, briefs writers can use, refresh lists for pages one push away from page one.
Anchor planning, prospect vetting, digital-PR outreach and earned placements that survive a core update.
Entity clarity and structured data so the product is the answer an AI engine cites rather than a footnote.
Redirect maps, staging crawls, hreflang at tens of thousands of URLs, and the monitoring that keeps rankings intact.
The layers this rests on, with the reasoning and the toolstack behind each decision, are on the home page: how the work lands · toolstack
Why faceted navigation quietly eats a quarter of your indexation — and the log-file workflow that proves it to a sceptical engineering team.
9 min readA study of 40 B2B products: which page types get quoted in AI answers and which get paraphrased into nothing.
12 min readAnatomy of a spec that passes sprint planning: selectors, acceptance criteria, and a rollback plan.
7 min readMeasuring what Googlebot actually renders on a JS-heavy hosting site — and where hydration costs indexation.
10 min readDemand creation in infrastructure niches: building the query before the query exists.
8 min readA generation pipeline that survives releases, plus the four errors that break clusters silently.
11 min readWhat earned placements in technical media have in common — and what the outreach template gets wrong.
9 min readINP on authenticated app screens: what actually moved the metric on a 900-component console.
10 min readHow an internal entity graph changed which pages answer engines treat as the canonical source.
8 min readPipeline-linked metrics that survive a budget review, and the three vanity numbers to drop.
6 min readThe redirect map, the staging crawl, and the two weeks of monitoring that keep rankings intact.
13 min read
SEROKUZE
{{ p }}
{{ t.storyLead }}
{{ s.body }}
{{ e.b }}
{{ t.bridge }} {{ t.bridgeCraft }} · {{ t.bridgeStack }}
