SE Ranking

SE Ranking exposes domain analysis, backlink intelligence, keyword research, site audit, and AI search visibility for competitive SEO work. This bundle covers the Data API product surface (not the separate Project API scope).

First moves

1. Subscription check — read account subscription metadata before large pulls; confirms remaining credits and plan limits.
2. Regional domain overview — organic vs paid keyword stats and traffic buckets for one country database when the user names a market (us, gb, de, etc.).
3. Worldwide overview — cross-region rollup when the question is global visibility rather than one country.
4. Domain keywords or pages — ranking keywords or top landing pages for a competitor domain in a chosen region.
5. Keyword gap — compare two domains' keyword overlap when the user asks what a competitor ranks for that they do not.
6. Backlink summary — aggregated link counts, referring domains, and anchor mix for a target host or domain.
7. Keyword expansion — related, similar, question, or long-tail suggestions from a seed keyword in a regional database.
8. Site audit status — crawl health, issue breakdown, and affected pages when an audit project already exists.
9. AI search visibility — brand discovery and prompt-level AI engine timeseries when the user asks about LLM or AI-overview presence.

Research workflows

Domain and competitor analysis

Start with a regional overview when the user names a country market; use worldwide overview for global snapshots. Follow with domain keywords, top pages, subdomains, paid ads, or organic competitors depending on whether the question is rankings, content, or competitive set. Keyword comparison between two domains answers gap questions — who ranks for what the other misses.

Historical trend reads help when the user asks how visibility changed over time rather than a point-in-time snapshot.

Backlink research

Choose mode to match scope:

• domain — entire registrable domain including subdomains.
• host — single hostname only (common default for one-site audits).
• url — one page.

Progress from summary and metrics to referring domains, anchors, indexed pages, or history as the user narrows from "how many links" to "who links" or "how anchors changed".

Keyword research

Pick the expansion shape that matches intent:

• Related — adjacent terms with volume and difficulty.
• Similar — close variants of the seed.
• Questions — informational query patterns.
• Long-tail — longer phrases for content planning.

Always pass a source country code for regional databases. Apply limit on suggestion reads — credits often scale per returned row.

Site audit

List existing audits, then read status, report, issues, issue pages, pages, links, or history for the audit id the user cares about. Creating, rechecking, or deleting audits mutates crawl projects — confirm before destructive changes.

AI search

Use brand discovery, prompt lists, and engine timeseries when the user asks how a brand appears in AI-generated answers rather than classic SERP rank.

Operational order

1. Read subscription metadata when credit budget is unknown or the pull may be large.
2. Set source (ISO country code) for regional domain and keyword work; omit only for worldwide overviews.
3. Pick backlink mode before comparing two targets — mixing host-level and domain-level scopes skews counts.
4. Start with summary reads; drill into referring domains or raw link lists only when the user needs detail.
5. Pass extra filters through query_params_json as a JSON object when volume, position, or sort filters are needed.

Gotchas

• Credits — successful responses consume credits; failed requests typically do not. Domain overviews are high per-request cost (~100 credits); keyword suggestions often bill per row.
• Regional vs worldwide — most keyword and domain reports need source; worldwide reads use domain-centric parameters instead.
• Data vs Project API — this bundle is Data API scope; Project API credentials will not authenticate here.
• Pagination — use limit and offset on large keyword suggestion pulls; do not assume a single page holds the full set.
• query_params_json — accepts a JSON object or JSON string for advanced filters (volume ranges, sort fields) without hand-building query strings in chat.