AI Search Optimization (AISO)

AI search visibility for modern brands.

Track, analyze, and improve how AI engines cite your brand — across Visibility, Position and Sentiment on ChatGPT, Perplexity, Google AI Overviews and Gemini.

// no account needed — your score in about a minute

Example workspaceIllustrative figures — your own numbers appear after the first daily check

Teams tracking their AI visibility with Radar

MyEventPlanerDatacube.ai
The pipeline

One keyword goes in. A ranked, published page comes out.

Most tools in this category stop at analytics — they tell you that you are losing and leave the work to you. Radar runs the whole loop: it researches, writes, scores and publishes, then watches what happens. Every workspace runs the same four stages on a schedule.

Stage 01

Research

Expand one seed keyword into hundreds of real queries, clustered by meaning — with volume, difficulty and intent attached to every one.

Stage 02

Write

A brief is built from a live read of who ranks right now. Then a full draft is written to match the intent behind the query — title, structure, body.

Stage 03

Score

A three-stage funnel grades relevance, readability and originality, then tells you what to fix. Nothing ships until it clears the number you set.

Stage 04

Publish & track

Approve and push straight to the CMS — WordPress today. From then on, position is checked daily and every move is flagged.

The platform

Six tools that used to be six subscriptions.

One workspace, one bill, one place where the keyword research, the draft, the score, the published page and the citation check all know about each other.

AI visibility tracking

Per-prompt, per-engine citation status across ChatGPT, Perplexity, Google AI Overviews and Gemini — with sentiment and position, not just a mention.

Daily checks

Keyword intelligence

One idea becomes a full topic map: hundreds of real queries clustered by meaning, each with volume, difficulty and intent.

Semantic clustering

Content generation

SERP-grounded briefs and full drafts written to search intent — blog, landing, product and FAQ pages.

Brief → draft

Rank tracking

Daily position checks on every published page, with alerts the moment you climb or slip.

Movement alerts

Technical site audits

Crawl the whole site and get a prioritised punch-list — titles, meta, headings, canonicals, broken links, missing alt text — sorted by severity.

Critical / warning / info

CMS publishing

Approve a page and push it to WordPress with meta title and description intact. No copy-paste, no export.

Draft or publish
AI visibility

Know exactly where you stand in AI answers.

Track the real prompts your buyers ask and see, per engine, whether you are cited, merely mentioned, or absent. No more guessing whether ChatGPT and Perplexity are sending your buyers to a competitor.

  • Per-prompt, per-engine citation status
  • Sentiment and position, not just a mention
  • Daily checks, so a drop shows up the next morning
Recommendations

Get told exactly what to publish next.

Every visibility gap becomes a ranked, sourced to-do — tied to the specific engine and prompt it would move, and ordered by impact. Actions, not another dashboard to interpret.

  • Each item names the engine and prompt it lifts
  • Impact-ranked, so you start where it counts
  • Sourced from who actually gets cited today
The scoring engine

A verdict, not a vibe check.

Most tools let the model publish whatever it writes. Radar runs every draft through a three-stage funnel and gives it a number — so weak pages get fixed or killed before they ever touch your site.

post ≥ 70review 55–69regenerate < 55
Stage 1 · free, instantHeuristics

Length, readability and CTA checks. Costs nothing, runs in milliseconds, and prunes the obvious misses before a single token is spent.

12 candidates in
Stage 2 · embeddingsRelevance & originality

Vector comparison against the topic and against everything you have already published. Near-duplicates are killed here, not caught by a reader later.

7 pass
Stage 3 · LLM judgeEditorial judgment

The survivors are ranked head-to-head by an LLM judge, which picks the strongest and says why the others lost.

3 ranked
ResultThe page that ships

One winner, with a score and a list of fixes attached. Below your threshold, it goes back for another pass instead of going live.

78 · post
Built for teams & agencies

Every client walled off. Every workspace on autopilot.

Radar is multi-tenant by design, not by configuration. Each client's keywords, content, rankings and CMS connections live in an isolated workspace — and one daily scheduled run fans out across every one of them.

01

Isolated by design

Data scope comes from the signed-in session — never from a request a client could tamper with. One client cannot see another's anything.

02

Roles & invites

Owner, manager and read-only access per workspace, so a client can watch progress without touching the controls.

03

Fans out on schedule

One daily run generates, scores, rank-checks and citation-checks across every workspace you manage — hands off.

FAQ

Questions, answered.

What is Spinlap Radar?

An AISO and SEO intelligence platform. It shows whether AI engines cite your brand, explains why competitors get picked instead, and runs a full engine to research, write, score and rank the content that earns those answers.

Is it an SEO tool or an AI tool?

Both. It combines traditional SEO — keyword research, content generation, rank tracking and site audits — with AI search visibility tracking and generative engine optimization in one platform.

What is AI Search Optimization (AISO)?

AISO — also called Generative Engine Optimization, or GEO — is the practice of improving how AI-powered search engines and large language models discover, understand and cite your brand in their answers. Spinlap Radar measures your AI visibility and tells you what to publish to improve it.

How is content quality controlled?

Every draft passes a three-stage funnel before it can ship: free instant heuristics, embedding-based relevance and originality checks, then an LLM judge that ranks the candidates head-to-head. The result is a number — post at 70 and above, review from 55 to 69, regenerate below 55 — so weak pages get fixed or killed before they touch your site.

Start ranking

Point it at a keyword. Watch it climb.

Spin up a workspace, drop in your first topic, and let the engine take the first page from research to published while you review.