> ## Documentation Index
> Fetch the complete documentation index at: https://www.pierview.ai/docs/llms.txt
> Use this file to discover all available pages before exploring further.

# Troubleshooting

> Common issues and what to check

## No visibility

**Symptom:** Your brand doesn't appear in scan results for prompts where you'd expect to see it.

**What to check:**

* **Brand variants**: Did you add all spellings? "Pierview" vs "Pierview AI" vs "Pier View"
* **Prompt fit**: Is the prompt too broad or unrelated to your category?
* **Model/platform**: Try another model (e.g., Perplexity vs ChatGPT); visibility varies
* **Timing**: AI outputs change. Run another scan to confirm
* **Baseline**: You may have zero visibility today

## Conflicting mentions

**Symptom:** Sentiment or mentions seem inconsistent across scans or prompts.

**What to check:**

* **Sample size**: Few prompts = noisier results. Add more prompts for stability
* **Context**: Same brand can be positive in one answer, neutral in another. We aggregate; check prompt-level detail
* **Competitor confusion**: Similar names? Add clear variants and review attribution
* **Rescan**: Run again; AI responses are non-deterministic

## Missing citations

**Symptom:** You're mentioned but rarely (or never) cited as a source.

**What to check:**

* **Source gaps**: See [Understanding Sources](/core-features/analysis/sources). Which domains does AI cite? Are you on them?
* **Site content**: Do you have citeable pages? "Best X," comparisons, FAQs. See [Citations](/core-features/analysis/citations)
* **Directories**: G2, Capterra, etc., are you listed? Do you have reviews?
* **Playbook**: Run [Create a Best X page](/playbooks/best-x-page-citations)

## Weird sentiment

**Symptom:** Sentiment scores seem off, e.g., you expect positive but see neutral or negative.

**What to check:**

* **Prompt-level view**: Drill into individual prompts; averages can hide nuance
* **Source of sentiment**: Is it from reviews, press, or AI's own wording? Different sources carry different weight
* **Sample size**: Few mentions = less reliable sentiment
* **Rescan**: Sentiment can shift with model updates or query phrasing

## Prompt mismatch

**Symptom:** Results don't match what you see when you manually query the same prompt.

**What to check:**

* **Exact prompt**: We send the prompt as defined. Typos or extra spaces change results
* **Model version**: Models update. Our scans use current versions; manual tests may differ
* **Non-determinism**: AI doesn't return identical answers. Run multiple scans for consistency
* **Regional/session effects**: Some models vary by region or session; we standardize where possible

## Next steps

* See [FAQ](/support/faq) for more Q\&A
* Review [Key Terminology](/introduction/key-terminology) for quick reference