v1 manual scan; full DataForSEO + Firecrawl run available as Option A scoping. Every estimate row below converts to a measured number in that run.
Technical SEO
A manual look at the public site, its structure, and the multi-campus footprint. Items are marked as either Observed (seen directly on the site) or Estimate (a reasonable inference pending a crawl).
| Item | Read | Type | Note |
|---|---|---|---|
| Multi-campus structure | Fragmented | Observed | Vienna, Berlin, Milan, Paris, Ljubljana, Linz. Each campus needs its own entity-clean, geo-targeted footprint. A single shared structure dilutes local signal per city. |
| Multilingual setup (DE/EN) | Partial | Observed | German and English present. hreflang completeness and per-locale parity look like an open question. Likely gaps across program pages. |
| Schema markup depth | Likely thin | Estimate | No obvious rich-result footprint for an institution of this size. Course, EducationalOrganization, and FAQ schema are very likely missing or minimal. |
| Course / Organization schema | Opportunity | Estimate | Each program is a citable entity. Structured Course and CollegeOrUniversity markup is the single highest-leverage technical add for AI answer eligibility. |
| Page-speed | Mid | Estimate | University sites of this vintage typically score mid-range on Core Web Vitals. Worth a real Lighthouse pass per template before any claim. Estimate only. |
| Crawlability / sitemap | Unknown | Estimate | Sitemap hygiene and internal linking across campuses not yet verified. A full crawl confirms whether deep program pages are reachable and indexed. |
Every "estimate" row above converts to a measured number in the Option A scoping run. Nothing here is presented as a precise measured metric.
AI readiness: the 6-signal gap
The six signals that decide whether an AI answer engine cites SFU when a student asks about its programs. Rated High / Med / Low by the size of the opportunity - High means a big, fixable gap.
| Signal | Current read | Opportunity | Why it matters |
|---|---|---|---|
| Schema / structured data | Likely thin | High | AI engines lean on structured Course and Organization data to understand and cite programs cleanly. |
| Inbound links / citations | Academic, uneven | Med | Strong academic citations exist, but commercial and program-level citation density is likely under-built for AEO. |
| Knowledge graph / Wikidata entity | Partial | High | A clean, well-connected entity is what lets an AI map SFU to "private university Vienna, psychotherapy science." Likely incomplete. |
| Citable answer-content | Prose-heavy | High | University copy is written for humans, not for extraction. Direct, answer-shaped content is what gets quoted by AI. |
| NAP / entity consistency across campuses | Likely inconsistent | Med | Six campuses across countries means six chances for name, address, and entity drift. Consistency feeds trust. |
| E-E-A-T | Strength | Low | Real strength. Faculty credentials, founding in 2005, accredited degrees, and a "comprehensive health university" mandate are genuine authority signals to surface, not build. |
E-E-A-T is marked Low opportunity because it is already strong. The work there is surfacing existing credibility into citable, structured form, not creating it.
Fix sprint: 8 moves ordered by lift over effort
The sequence a first engagement would run, front-loaded with the highest-return, lowest-effort moves.
| # | Move | Lift | Effort | What it does |
|---|---|---|---|---|
| 1 | Course / EducationalOrganization schema | High | Low | Makes every program machine-readable and AI-citable. Fastest technical win. |
| 2 | AI-readable program pages | High | Low | Restructure each program page so the key facts (degree, language, duration, outcome) are extractable, not buried in prose. |
| 3 | FAQ / answer content per program | High | Med | Direct question-and-answer blocks per program are what AI engines quote. Maps to real student queries. |
| 4 | Faculty entity pages | High | Med | Turn SFU's real faculty credibility into structured, linkable entity pages that feed E-E-A-T and citations. |
| 5 | Wikidata entity build | Med | Low | Clean, connected knowledge-graph entry so AI can place SFU and its programs accurately. |
| 6 | Internal linking across campuses | Med | Low | Tie the six campuses together with deliberate internal links so authority and crawl depth flow. |
| 7 | Multilingual hreflang cleanup | Med | Med | Correct per-locale targeting so DE and EN pages do not compete or leak signal across markets. |
| 8 | Citable program outcome data | Med | Med | Surface real outcome and accreditation data in citable form, the kind of proof AI engines reward. |
All ratings above are a manual v1 estimate. The Option A scoping run replaces every estimate with measured DataForSEO and Firecrawl data, and confirms schema, speed, hreflang, and citation gaps against live crawl evidence.