Prompts & Tags

Prompts are the search queries BuzzView uses to analyze AI responses. Writing the right prompts and organizing them with tags unlocks the full power of BuzzView.

What are Prompts

A prompt is a question or search query that real users type into AI-powered search engines. BuzzView runs your prompts through the AI platforms you've selected and analyzes the responses for brand mentions, citations, and sentiment.

The quality of your prompts directly determines the quality of your insights. Generic prompts produce generic results. Prompts that mirror what your customers actually ask produce data you can act on.

Writing Effective Prompts

Follow these principles to build a high-quality prompt set:

  1. 1
    Think like your customer
    What questions do they ask when researching your product category? Start from their perspective, not yours.
  2. 2
    Be specific
    "Best CRM for small businesses" is far more useful than just "CRM software". Specific queries reflect real user intent and produce more meaningful results.
  3. 3
    Include comparison queries
    "X vs Y" and "best alternatives to X" prompts reveal how your brand is positioned competitively in AI responses.
  4. 4
    Cover the buying journey
    Mix awareness-stage questions ("what is X?"), consideration-stage ("best tools for X"), and decision-stage ("X pricing and reviews") queries.
  5. 5
    Test different phrasings
    AI responses can differ significantly based on how a question is phrased. Including variations of the same intent gives a more complete picture.

Prompt Volume

There's no fixed minimum or maximum, but 30–200 prompts typically provides a statistically meaningful dataset. Below 20 prompts, results may not be representative enough to draw reliable conclusions. Above 500, consider splitting into focused sub-projects organized by topic or market segment.

Quality outweighs quantity. A focused set of 50 well-crafted prompts will generate more actionable insight than 500 vague or repetitive ones.

Using Tags

Organizing with Tags

Tags let you group and filter prompts within a project. You can assign tags by topic (e.g., "pricing", "features", "competitors"), by buying stage, by product line, or any other dimension that matters to your analysis.

Once tagged, you can filter the entire project dashboard to analyze any subset of prompts independently — so a tag like "competitors" instantly shows you your brand's visibility only on comparison queries.

Tag Strategy

Think about tags before you import prompts — it's much easier to tag while importing than to go back and tag later. Common tagging schemes include: by topic area, by product category, by competitor mentioned, and by funnel stage.

Importing Prompts

Upload a CSV or .txt file to add prompts in bulk. CSV format: column A for the prompt text, column B (optional) for a comma-separated list of tags to apply to that prompt. The .txt format accepts one prompt per line with no tags.

The importer validates your file format and shows a full preview before adding prompts to your project, so you can catch any issues before committing.

Avoid prompts that are too generic (just a category name) or too brand-dependent (just your company name). These don't reflect how real users discover brands through AI search and will produce less useful results.