Getting Started with StretchSearch

StretchSearch is the StretchSuite platform for governed site search, crawling, indexing, and ranking. This article introduces the core concepts and walks you through your first project so the rest of the documentation makes sense.

What StretchSearch does

StretchSearch manages the full lifecycle of turning web pages into trustworthy search: you register the sites you're allowed to crawl, configure governed crawl jobs, review the SEO intelligence and ranking previews that come out of indexed pages, and expose that indexed evidence to search and to AI. Unlike a hosted search box, it is governance-first — every crawl is bounded by authorization, robots policy, risk level, rate limits, and an approval workflow.

Core concepts

• Project — the container for a body of work. Every target, crawl job, finding, and ranking belongs to a project. Project types include owned_site, client_site, competitor_research, brand_audit, seo_intelligence, and research.
• Target — a domain you have declared StretchSearch is allowed to touch, with its authorization status, robots-policy review, allowed domains/prefixes, and risk level.
• Crawl job — a governed plan to fetch pages: seed URLs, scope, depth, page cap, and rate limit, plus governance and readiness state.
• Indexed page — a stored page record (title, meta description, H1, status code, canonical URL, extracted text) that feeds findings, candidates, and ranking.
• SEO finding — a deterministic, severity-ranked issue detected on indexed pages.
• Ranking preview — a scored, ordered candidate produced by the StretchRank engine.
• Evidence — the citable records that ground search results and AI answers.

The most important thing to understand first

Creating things in StretchSearch does not start a crawl. Registering a target or creating a crawl job produces a governed record, not live fetching. Crawler execution is disabled by default and is enabled centrally by your platform operators only after authorization and readiness are satisfied. Throughout the product you'll see this reflected as crawler_enabled: false. This is by design and is what makes StretchSearch defensible for enterprise and agency use.

Set up your first project

1. Open StretchSearch from your StretchSuite workspace. The dashboard Overview shows your projects, recent crawl jobs, open SEO findings, and ranking-preview counts.
2. Go to Projects and choose New project.
3. Give it a clear name (for example, Acme Marketing Site Audit).
4. Pick a project type that matches the work — use owned_site for a property you own, or client_site for one a client has authorized.
5. Optionally set an owner brand and a description so teammates know the scope.
6. Save. Your project is created with status active and its own project key.

Add your first target

1. Inside the project, open Targets and choose Register target.
2. Enter the root URL (a full https:// URL) and the domain.
3. Set the target type (owned_property, client_authorized, competitor_research, or public_reference) and the authorization status.
4. Review the robots policy and set risk level. Save the target.

Plan your first crawl job

1. Open Crawl jobs and choose New crawl job.
2. Select a crawl mode (for example, owned_site_deep_crawl).
3. Add at least one seed URL and set the scope, max depth, max pages, and rate limit.
4. Save. The job is created as a draft — no crawl runs yet. From here it moves through the readiness checklist and approval before execution can ever be enabled.

A realistic example

Maya runs SEO for a small agency. She creates a client_site project called Northwind Coffee — Q3 SEO, registers northwindcoffee.com as a client_authorized target after confirming written permission, and drafts an seo_content_crawl job seeded with the blog index and the shop landing page, capped at 300 pages with a 1,000 ms rate limit. She completes the readiness checklist, and her platform operator approves execution. Once indexing runs, Maya reviews SEO findings and ranking previews without ever leaving her workspace.

Tips

• Name projects for the engagement plus the time frame — it keeps federated search results readable.
• Always confirm authorization before registering client or competitor targets. The metadata fields exist to record that permission.
• Start crawl jobs small (low page cap, seed_urls_only scope) and widen scope only after a clean first pass.
• Use the dashboard Overview as your daily home base — it surfaces the highest-severity findings and the latest crawl activity.

FAQ

Do I need admin rights to use StretchSearch? No. Everyday work — projects, targets, crawl jobs, reviewing findings and rankings, and federated search — happens in your workspace. Governed actions like enabling crawler execution or generating findings are handled by platform operators.

Will registering a competitor's domain crawl it immediately? No. It creates a governed record with crawler_enabled: false. Nothing is fetched until authorization and readiness are satisfied and execution is enabled centrally.