Quickstart

From sign-in to your first calibrated visibility score in about five minutes. You enter a URL, Lokrix creates a property, generates the real questions your buyers ask AI engines, runs them through the engines, and returns a Lokrix AI Score with a confidence interval. Here is the full path, step by step.

The property onboarding funnelA narrowing five-stage funnel: add property, readiness pre-check, analyze brand, first scan, and report, connected top to bottom.Add propertyReadiness pre-checkAnalyze brandFirst scanReport

1. Sign in

Create an account or sign in. Your first sign-in provisions an Organization and a default Workspace for you — the containers that hold your properties and teammates. If your company uses enterprise SSO, you can join through your identity provider. To understand how these containers nest, see the workspace model.

2. Add a property

A property is one site you track. Enter its URL in onboarding: Lokrix creates the property, detects the brand and industry, and moves straight into prompt generation. You can track many properties from one workspace and switch between them with the property switcher — details in Onboarding a property.

3. Generate prompts

Lokrix builds a prompt universe — the real buyer questions that AI engines get asked about your category, not a generic keyword list. Review the generated prompts, keep the ones that match your buyers, and edit or add your own. This set defines what every future scan measures, so it is worth a quick pass. See the prompt universe.

4. Run a scan

Start a scan. Each prompt is run K times across varied temperature, persona and locale on each engine, so Lokrix can estimate a Monte-Carlo presence probability rather than a single-run guess. Five engines run live by default — ChatGPT, Claude, Perplexity, Grok and Gemini; Google AI Overviews and AI Mode need a SERP provider key. Live engine responses are metered against your workspace credit balance (1 credit ≈ 1 engine response), so Lokrix defaults to the cheaper predictive model and escalates to live runs deliberately. Read Live vs. predictive scans and Scan cost.

5. Read the score

When the scan finishes you get a Lokrix AI Score — a calibrated 0–100 composite with a 95% confidence interval. Drill into each engine for its component score, presence, share-of-voice and the citations behind it. Every value is labelled by provenance (live, cached, or mock sample), and Lokrix never presents mock data as live. Learn to read it in the score anatomy.

What’s next

With a baseline in hand, open the optimization engine for prioritised, simulated recommendations and a predicted path to #1 — see Recommendations. New to the vocabulary? Start with Core concepts.