About the service
Keyword research — the foundation of SEO
The keyword research is the complete set of search queries your target audience uses to look for products or services. By analogy, it’s the blueprint of a house before you dig the foundation. Without a core, SEO turns into guesswork: we write copy for assumptions, build pages for ‘it seems people search this’, and then wonder why traffic isn’t growing.
At heleum.studio we build a core not as a Key Collector export into Excel, but as a ready-to-use document. Every query goes through cleanup, clustering by search-result similarity and classification by intent — so you understand which query belongs on a category page, which on a product page, and which in the blog.
What you get
The final result isn’t a single file but a package of materials the whole team can work with right away: SEO specialist, copywriter, developer, marketer.
- A query table with broad and exact volume, competition, priority and page assignment.
- A sitemap in Miro or FigJam — the structure with landing pages, category hierarchy and filters.
- A content spec — for each page: H1, Title, Description, required subheadings, LSI phrases, recommended length.
- Prioritization — which pages to start with so traffic comes faster and at minimal content cost.
Why clustering is critical
If two queries have very similar search results, Google shows the same page for both. Trying to create two pages for such queries is cannibalization: they compete with each other and neither reaches the top. So we cluster queries automatically by SERP and then always check every cluster manually. The algorithm gets 60–70% right; a human refines the rest — and that’s the difference between a ‘normal’ and a ‘working’ core.
Intent: what the user is really searching for
The same query has different intent in different contexts. ‘Buy iPhone’ is commercial — the person is ready to buy. ‘iPhone review’ is informational — they’re choosing. ‘iPhone official site’ is navigational. If you point an informational query at a category page full of products, you lose to competitors who realized the user wants content, not a product grid.
A core for GEO results (ChatGPT, Perplexity)
AI search engines build answers from long, natural, question-style queries — not short keyphrases. We separately collect ‘long-tail’ queries in question form (‘how to choose’, ‘which is better’, ‘is it suitable’), because that’s what generates GEO traffic. In 2025, ignoring this layer of queries means leaving about 15–25% of potential traffic to competitors.
How we work with international projects
For foreign markets, simply translating a Ukrainian core isn’t enough. We collect local queries from scratch, verify them with native speakers, and account for cultural differences in search behavior. Polish and Czech users phrase queries with the same meaning differently — and the site structure for those markets will differ.
When the core needs updating
The market doesn’t stand still. New product categories appear, seasonal trends shift, new products launch in your niche. We recommend a full core expansion every 6–12 months and partial additions monthly if you’re actively producing content. For large online stores we monitor trends continuously.
You can order keyword research through the form above or by phone. We first look at your site and niche, then give a concrete estimate of scope, timeline and cost — without ‘market-average’ price tags or upselling.
