Know exactly what your competitors are doing.
Lightspace Labs automatically discovers your competitors, analyzes them across five intelligence dimensions, and delivers continuously updated reports to your portal — without any manual research on your part.
From sign-up to intelligence in hours.
The platform runs automatically from the moment you sign up. Here's what happens behind the scenes.
Brand Research
Claude researches your brand — positioning, audience, products, and market context — to understand who your competitors should be.
Query Generation
AI generates search queries representing how your customers actually look for solutions like yours.
Visibility Scan
Queries run against Perplexity, ChatGPT, and Gemini. Every cited domain is logged with its citation frequency.
Competitor Extraction
Claude scores each cited domain for relevance — filtering out news sites, directories, and social platforms — and ranks the remainder.
Your Portal
You log in to a ranked list of competitor candidates. Star the ones you want to track. Full CI analysis begins immediately.
A three-layer AI pipeline on every analysis cycle.
Each starred competitor runs through three Claude model layers — with five parallel scout agents working simultaneously at Layer 1.
Five Haiku agents run simultaneously per competitor, each with a dedicated mission: pricing, content strategy, marketing messaging, proof signals, and tech stack. Each agent autonomously decides which pages to fetch and when it has enough data.
A Sonnet agent receives all five scout outputs and synthesizes them into a structured CompetitorProfileCard — a six-dimension business analysis covering positioning, product, pricing, proof, visibility, and growth signals.
Opus receives all competitor profiles for your market and produces a StrategicIntelligenceBrief: where to win, where to defend, active threats, prioritized opportunities, and content briefs. Every plan tier receives Opus synthesis — quality is not gated by price.
Five dimensions, one portal.
Each starred competitor is analyzed across five modules on a rolling schedule. All intelligence is available in your portal with side-by-side comparison views.
Citation Visibility
Track how often your competitors appear in Perplexity, ChatGPT, and Gemini — daily.
Learn moreShow less
How often does each competitor appear in AI-generated search results compared to your own site? The citation visibility module runs your standard query set against Perplexity, ChatGPT, and Gemini daily — counting how often each starred competitor is cited vs. how often you are, broken down by engine and by query.
This is the fastest-moving signal in the CI suite. A competitor launching a content push or restructuring their site for GEO will show up here within days.
What you see
- Your citation rate vs. each competitor
- Breakdown by AI engine (Perplexity, ChatGPT, Gemini)
- Per-query breakdown — which queries favor competitors
- Top competitor domains by citation frequency
- Trend over time
Marketing Strategy
A weekly snapshot of how each competitor positions, who they target, and how their tone is shifting.
Learn moreShow less
Every week, Claude fetches each competitor's homepage, about page, and blog content and extracts a structured marketing profile. This tells you how they're positioning their product, what audience they're targeting, which channels they're active on, and how their tone and messaging have shifted since last week.
Useful for spotting competitive pivots early — if a competitor shifts from SMB messaging to enterprise, you'll see it in their messaging pillars before it shows up in the press.
What you see
- Market positioning statement
- Tagline (current and historical)
- Messaging pillars — their key value claims
- Content themes and topic focus
- Tone (professional, casual, technical, etc.)
- Active marketing channels
Tech Stack Detection
Identify every technology powering each competitor's site — decoded from public signals alone.
Learn moreShow less
Claude fetches the competitor's public HTML, response headers, and script sources — then interprets the signals to identify the technologies powering their site. No special access required; everything is decoded from what their site publicly exposes.
Useful for inferring their infrastructure costs, their team's technical choices, and which ad platforms, analytics tools, and chat systems they're investing in.
What you see
- Frontend framework (React, Vue, etc.)
- CMS (WordPress, Webflow, Contentful, etc.)
- Analytics platforms (GA4, Mixpanel, etc.)
- Ad platforms (Google Ads, Meta Pixel, etc.)
- Chat and support tools
- CDN and infrastructure signals
Pricing Intelligence
Catch competitor price drops, new tiers, and plan changes within a week — not when a customer mentions them on a call.
Learn moreShow less
Claude checks each competitor's pricing and plans pages weekly, extracting a structured breakdown of every plan they offer. Price changes, new plan tiers, and feature adds or removals are all captured in the snapshot timeline.
If a competitor quietly drops their entry price or adds a free tier, you'll know within a week — not when a customer mentions it in a sales call.
What you see
- Whether they publish pricing at all
- Plan names, prices, and billing cadences
- Which plan is marked as 'most popular'
- Feature inclusions per plan
- Historical pricing snapshots for comparison
Product Deep-Dive
A full hierarchical map of each competitor's product — modules, features, and screenshots — refreshed monthly with vision-powered analysis.
Learn moreShow less
Once a month, Claude runs an in-depth product analysis using vision — fetching product and features pages, encoding the top screenshots, and passing them to Claude Opus alongside the text content.
The result is a full hierarchical product outline: products, modules, and individual features — with screenshots and demo videos linked inline. This is the deepest competitive intelligence available without direct product access, useful for spotting gaps in your own product or preparing for competitive sales situations.
What you see
- Full product list with descriptions
- Module / subproject hierarchy per product
- Individual features with names and descriptions
- Product screenshots linked inline
- Video demos linked (YouTube, Vimeo, Wistia)
- Monthly snapshot timeline for change tracking
Everything in one place.
All CI intelligence surfaces in a clean client portal. Overview, competitor management, intelligence comparison views, and run history are organized into tabs so you can focus on what matters.
The Competitors tab shows pending candidates (auto-discovered) alongside your actively tracked competitors. Star or reject candidates, add competitors manually, and see the analysis status for each tracked competitor.
The Intelligence tab gives side-by-side comparison views for each module — a citation visibility bar chart per engine, a marketing strategy table, a tech stack grid, a pricing matrix, and a product outline accordion with screenshots.
Star to track. Unstar to pause.
Competitor candidates arrive with a Pending status — discovered by our AI but not yet costing you a tracked-competitor slot. You review the list and star the ones you want to follow.
Each plan has a cap on the number of starred competitors. Starring activates the full CI schedule for that competitor and claims one slot. If you hit your limit, you'll see an upgrade prompt.
Un-starring a competitor moves them to Paused — analysis stops but all historical data is retained. Their slot is freed immediately. Dismissed competitors move to Rejected and are hidden from the main view.
Competitor statuses
- PendingAuto-discovered, awaiting your review. Does not count against your plan limit.
- ActiveStarred and being tracked. Full CI analysis schedule runs on this competitor.
- PausedWas active but un-starred. Analysis paused; historical data retained. Slot freed.
- RejectedDismissed from the pending list. Hidden from main view; re-add manually if needed.
Common questions about the platform
What is automated competitive intelligence?
Automated competitive intelligence is the continuous, AI-driven process of monitoring competitors' public digital presence — their pricing, marketing strategy, product offerings, technology stack, and search visibility — without manual research. Lightspace Labs runs a three-layer AI pipeline per competitor: five parallel Claude Haiku scout agents gather raw data simultaneously, a Claude Sonnet specialist analyst synthesizes it into a structured profile, and a Claude Opus strategic synthesizer produces a cross-competitor intelligence brief with prioritized opportunities. The pipeline runs on a rolling schedule — daily for citation visibility, weekly for marketing and pricing, monthly for product analysis — so intelligence stays current without any client effort.
How quickly will I have competitor data after signing up?
Most clients have a ready-to-review competitor candidate list within a few hours of signing up. Lightspace Labs immediately researches your brand, generates search queries for your market, and scans Perplexity AI, ChatGPT, and Gemini to identify which domains co-appear with your business in AI-generated results. After you star the competitors you want to track, the first full five-module intelligence pass completes the same day.
What does citation visibility track and why does it matter?
Citation visibility measures how often each competitor appears as a cited source in AI-generated search responses — broken down by engine (Perplexity AI, ChatGPT, Gemini) and by query. It updates daily. As AI search becomes a primary discovery channel, appearing in AI-generated answers is a direct measure of market authority. Citation visibility shows exactly which competitors are gaining or losing AI search presence, and how your own site compares.
How is AI-powered competitive intelligence different from manual research?
Manual competitive intelligence — checking competitor sites, reviewing pricing pages, reading their blog — takes hours per competitor and produces a snapshot that is immediately out of date. AI-powered CI runs continuously: citation visibility updates daily, marketing strategy and pricing update weekly, and product hierarchy updates monthly, with no time investment from you. AI also processes signals a manual researcher cannot: response headers for tech stack detection, visual product hierarchy extraction using Claude's vision capability across screenshots, and citation frequency tracking across three AI search engines simultaneously.
Is competitive intelligence legal?
Yes. Lightspace Labs uses only publicly available information — the same data any person could access by visiting a competitor's website. The platform analyzes public web pages, pricing pages, product descriptions, and publicly visible response headers and script sources. It does not access data behind login walls, scrape private systems, or circumvent any technical protection. Gathering and analyzing publicly available competitor information is a standard, legal business practice.
Which AI models power the competitive intelligence pipeline?
Lightspace Labs uses three Claude models in its CI pipeline. Layer 1 uses Claude Haiku — five parallel agents per competitor, optimized for speed on high-volume extraction tasks. Layer 2 uses Claude Sonnet — a specialist analyst that synthesizes all five scout outputs into a structured competitor profile. Layer 3 uses Claude Opus — reserved for strategic synthesis across all competitor profiles to produce the StrategicIntelligenceBrief with prioritized opportunities and content briefs. All plan tiers receive Opus synthesis; quality is not gated by price.
Can I track competitors that weren't automatically discovered?
Yes. Beyond the auto-discovered candidates, you can add any competitor by URL directly from your portal. Manually added competitors are treated identically to discovered ones — star them to activate the full five-module CI schedule. This is useful for tracking specific regional competitors, known rivals in adjacent markets, or businesses that do not yet appear in AI search alongside your brand.
Start tracking your competitors today.
Sign up in minutes. Your competitor list is populated automatically — no setup calls, no waiting.