How Rankry Works
Rankry works as a pipeline. It sends your buyer-intent prompts to ChatGPT, Claude, Gemini, Perplexity, and Grok, parses each answer for your brand and your competitors, scores the results, identifies the sources behind them, and turns the findings into an action plan. The same prompts run on a schedule, so you get a trend rather than a one-off snapshot.
What this page covers
Section titled “What this page covers”- The Rankry pipeline, step by step
- What each stage produces
- How often it runs
- Where to see the output
The Rankry pipeline
Section titled “The Rankry pipeline”- Onboarding. You add your brand, confirm your category and location, pick competitors, and review a generated prompt set. See Add your brand.
- Prompts. Rankry builds a tailored prompt set for your category (your Semantic Core) and lets you add your own in Prompt Lab.
- Model responses. Each prompt is sent to the five AI models. Rankry stores the full answer, the brands named, their order, the rationale, and any cited sources.
- Scoring. Rankry computes Visibility, Position, Sentiment, and Diversity per model, and an overall Score. See Metrics reference.
- Sources. Rankry classifies the domains AI engines cite and flags where competitors are cited and you are not.
- Competitors. Rankry scores each competitor the same way and computes head-to-head results.
- AI Readiness. Rankry audits your website across 36 technical checks that affect whether AI engines can read and cite you.
- Action plan and tasks. Rankry converts gaps into a prioritized brief and a task list, and can draft content to close them.
What each stage produces
Section titled “What each stage produces”| Stage | Output |
|---|---|
| Model responses | The raw answer, ranked brands, rationale, cited sources |
| Scoring | Visibility, Position, Sentiment, Diversity, overall Score |
| Sources | Cited domains, source classification, source gaps |
| Competitors | Per-competitor scores, head-to-head win rate |
| AI Readiness | Category scores, grade, specific findings |
| Action plan | Strategic brief, weekly unblockers, roadmap, tasks |
How often it runs
Section titled “How often it runs”Core prompts run when a report is generated, on your plan’s refresh cadence. Lab prompts run on demand, or on a daily or weekly monitor. Because AI answers shift over time, the recurring schedule is what turns variation into a usable trend.
Where to see the output
Section titled “Where to see the output”The dashboard shows your overall Score and the Visibility, Sources, Compete, and Action plan views. Each score links back to the exact AI response behind it, so every number can be traced to evidence.
Frequently asked questions
Section titled “Frequently asked questions”Does Rankry use the real AI models?
Section titled “Does Rankry use the real AI models?”Yes. Rankry queries ChatGPT, Claude, Gemini, Perplexity, and Grok and analyzes their actual answers.
Why do results change between runs?
Section titled “Why do results change between runs?”AI answers are probabilistic and the models and their sources change over time. Rankry runs the same prompts on a schedule so you can see the trend rather than a single noisy sample. See Why did my results change?.
Can I see the actual AI answer behind a score?
Section titled “Can I see the actual AI answer behind a score?”Yes. Every metric links to the response it came from, so you can verify any number.
How many prompts does Rankry run?
Section titled “How many prompts does Rankry run?”Rankry generates a tailored prompt set for your category during onboarding, and you can add more in Prompt Lab. The exact count depends on your plan and your edits.