Skip to content

Limits and Known Issues

AI visibility tracking has real limits, and Rankry is honest about them. AI answers are probabilistic and change over time, so a single report is a sample, not a verdict. This page lists the main limits so you can read your data correctly.

  • The limits of AI answers
  • The limits of citation data
  • Plan and data limits
  • Model variability. The same prompt can produce different answers on different runs. Read the trend across reports, not a single result.
  • Prompt volatility. Small wording changes can change an answer. Keep prompts fixed so results are comparable.
  • Model differences. The five models disagree, which is expected. Track them separately.
  • Indexing delay. A new or changed page takes time to affect AI answers. Allow several reporting cycles after a change.
  • Citations are shown only where a model surfaces them. Answers from training memory carry no citations.
  • Citation coverage differs by model, with Perplexity the most consistent.

Each plan sets limits on brands, prompts, audit runs, content drafts, history, and export formats. For exact and current numbers, see the pricing page.

Why did my results change without any action on my part?

Section titled “Why did my results change without any action on my part?”

AI answers shift as models and their sources update. This is normal. The recurring schedule is what separates signal from noise.

Can Rankry guarantee a position in AI answers?

Section titled “Can Rankry guarantee a position in AI answers?”

No. AI answers are probabilistic, and no tool can guarantee placement.

Usually several reporting cycles, because AI answers update gradually.