Source context
An international remote-staffing agency (B2B) that provides virtual assistants and remote talent to businesses. Revenue: sourcing fees, monthly management fees and staffing contracts.
← All case studies · B2B · Remote staffing agency · Global
A 336-page topical map for a remote staffing agency
An international agency that places virtual assistants and remote talent with businesses. Before any content was written, I built the semantic blueprint: the five components, the research, and a map of every page the site needs, in priority order.
- 336Pages in the final map
- 570Candidate topics evaluated
- 26Steps in the process
01The business
A crowded niche full of generic content
- Industry
- B2B · remote staffing and outsourcing
- Market
- Global · English
- Delivered
- 2026
- Scope
- Topical map, knowledge-domain research, URL architecture, then content briefs and writing
Outsourcing is dominated by big agencies and listicles that all say the same thing. To be seen as an authority, the site needed to cover the whole topic, from "what is outsourcing" to choosing a partner, in a structure Google can understand, and to connect every informational page back to the hiring service.
02The 5 components
The foundation every topic hangs on
Before collecting a single topic, the five components fix what the site is about, who it serves and how informational pages lead to the service.
Central entity
Outsourcing, anchored to its Wikipedia entity so the whole map shares one meaning.
Central search intent
Know, find, screen and hire remote talent.
Core section
Every attribute of outsourcing that leads to the service: hiring remote talent, costs, models, countries, roles.
Outer section
Everything else people want to know about outsourcing and virtual assistants: definitions, history, comparisons, risks.
03Research
26 steps from idea to a finished map
Phase 1 · Steps 1–5
The 5 components
Source context, central entity, central search intent, core section and outer section, agreed before any topic research.
Phase 2 · Steps 6–19
Topic research
Buyer personas, Wikipedia and expert sources, LLM entity research, knowledge domains, lexical relations, ontology, frame semantics, query networks, SERP analysis (autocomplete, PAA, knowledge panels, competitor query networks), forums and competitor sitemaps. Then SERP-based clustering to decide which queries deserve their own page.
Phase 3 · Steps 20–26
Processed map
Title tag, URL, image name, alt text and meta description for every page, quality nodes highlighted, and a publishing momentum plan.
- 100+Knowledge-domain terms defined
- 5Researchers collecting raw topics
- 234Topics rejected: no page deserved
04The topical map
Every page, its priority and its place
The final map gives each page a priority, a URL in a clear parent-child hierarchy, a title tag, an image name and alt text. Colours mark the status of each page, from brief ready to published. Topic names are hidden to protect the client.
- 135High-priority pages
- 182Mid-priority pages
- 4 levelsURL depth, from hub to detail
05What came next
From map to content
- Content briefs built page by page from the map, starting with the high-priority hubs.
- Articles written in map order, so every new page links into a hub that already exists.
- An independent fact-check report on each batch before delivery.
Starting a site or a content plan from scratch?
I build the topical map first, so every article you publish has a reason to exist and a place to link.