Skip to main content

​
Overview

Position Score measures the average ranking position of your brand (or products) when they appear in AI-generated answers. Instead of asking “are we mentioned?”, Position answers “where do we appear relative to others?” — for example, whether you are usually mentioned first, second, third, or later in the response. Lower values are better:
  • 1.0 → typically mentioned first.
  • 2.0–3.0 → often mentioned but behind other brands/products.
  • Higher values → show up later in the answer, with lower prominence.

​
How Position is Calculated

​
Brand Position (Brand Prompts)

For brand-level analytics, position is derived from the brand_mentions table in the analytics schema:
  • Each row in brand_mentions has a mention_order integer (1, 2, 3, …) indicating the order in which a brand was mentioned within a chat.
  • The Brand Position Score is: ```text Position Score = AVG(mention_order) across all brand mentions ```

​
Calculation details

  • Input: brand_mentions.mention_order values for your brand.
  • Aggregation: Average (mean) of all mention orders for the brand in the selected period.
  • Range: 1+ (1 = first mentioned, 2 = second mentioned, etc.).
  • Lower is better: A lower score means the brand is typically mentioned earlier in the answer.
  • Filters Applied: Project, date range, AI models, topics/prompts, and optionally brand domain.
In SQL terms (as used in the analytical views documented in DATABASE_SCHEMA.md): 
avg_position = AVG(bm.mention_order)

​
Product Position (Product Prompts)

For product-level analytics, position is derived from the product_mentions table:
  • Each row in product_mentions has a mention_order for that product mention.
  • The Product Position Score for a given product prompt is: 
    Product Prompt Position = AVG(mention_order) across all product mentions for the prompt

​
Calculation details

  • Input: product_mentions.mention_order values filtered by prompt_id.
  • Aggregation: Average of mention orders for all products mentioned when that prompt was executed.
  • Range & interpretation: Same as brand position (1 = first product mentioned).
  • Filters Applied: Project, date range, AI models, product prompt IDs, and optionally specific product URLs.
In SQL terms (as used in getProductPromptsAnalytics): 
AVG(pm.mention_order) as avg_position

​
Display formats

  • Brand dashboard:
  • Position appears in the brand metrics table alongside Visibility and Sentiment.
  • Product prompts dashboard:
  • Position is shown per product prompt, so you can see which prompts produce earlier recommendations for specific products.

​
Why Position Matters

​
Brand Authority in AI Answers

Position reflects how authoritative and top-of-mind AI models consider your brand for a given topic:
  • A position close to 1 implies:
  • Your brand is often the first option AI mentions.
  • You are treated as a primary reference in your category.
  • You are highly relevant to the user’s query.
  • A higher position (e.g. 3–5) implies:
  • Your brand appears but is overshadowed by others.
  • Users may see competitors first and consider them before you.

​
Competitive Advantage

Because position is calculated for all brands and products in the conversation:
  • You can see whether AI tends to lead with you or with competitors.
  • You can identify:
  • Competitors that consistently rank ahead of you.
  • Topics where you already “own” the top position.
Higher authority (lower position) combined with good sentiment and visibility is a strong indicator of category leadership in AI search.

​
Topic Relevance

Position can vary significantly by:
  • Topic / prompt (e.g. you may be first for “budget-friendly tools” but lower for “enterprise solutions”).
  • Model (different models can have different preferences).
This helps you identify:
  • Topics where you’re strongest and should double down.
  • Gaps where you’re visible but not ranked early — an opportunity to strengthen content and sources.

​
Understanding Your Position Data

​
In the Overview Dashboard

On the Overview (brand) dashboard:
  • The Brand table shows:
  • Visibility: how often you’re mentioned.
  • Sentiment: how positively you’re described.
  • Position: how early you appear when mentioned.
  • Use filters (models, date range, topics/prompts, brands) to:
  • See if position is better on certain models.
  • Compare your average position vs individual competitors.

​
In the Product Prompts Dashboard

On the Product Prompts dashboard:
  • Position is calculated per prompt, across all product mentions for that prompt.
  • This lets you see which prompts:
  • Make your products show up first or early.
  • Consistently rank competitor products ahead of yours.

​
Interpreting Ranges

Use these ranges as a practical guide (exact thresholds depend on how many brands/products appear per chat):
  • 1.0–1.5 → You are usually the first or among the top mentions.
  • 1.5–3.0 → You are regularly in the main comparison set, but not always first.
  • >3.0 → You tend to appear later; competitors are generally prioritized ahead of you.
Always interpret position together with visibility:
  • Low position (good) + high visibility = strong authority and presence.
  • Low position + low visibility = strong performance when mentioned, but not seen often.
  • High position (poor) + high visibility = often mentioned, but not top-ranked — a signal to improve authority.
Position works best in combination with other core metrics: Together, these metrics answer:
  • Are we present? (Visibility)
  • How are we framed? (Sentiment)
  • Where do we rank? (Position)

​
Improving Your Position

To move your brand or products closer to position 1:
  1. Improve source quality and authority
  • In your Sources / Prompt Sources views, identify domains and URLs that strongly influence prompts where you rank well.
  • Aim to secure similar or better coverage on those domains for topics where you rank lower.
  1. Strengthen category fit in content
  • Ensure your content clearly and explicitly addresses the category and use cases covered by your prompts.
  • Use comparison pages, “best X for Y” guides, and detailed product pages that match real user queries.
  1. Optimize prompts and topics
  • Refine prompts so they reflect how your ideal customers ask questions where you should naturally be top-of-list.
  • Add prompts for topics where you already have strong content and third-party coverage.
  1. Monitor changes over time
  • Track position trends after major launches, new reviews, or content updates.
  • Watch for competitors that start to appear ahead of you and investigate which sources or narratives are driving that change.
Position is a direct signal of who AI treats as the primary answer. By combining position insights with visibility, sentiment, and sources, Orvi AI helps you systematically move your brand toward the top of AI-generated recommendations.
Last modified on January 25, 2026