Your Search Console dashboard can show flat organic traffic while your brand quietly disappears from AI answers. The metrics that caught the decline — rankings, CTR, sessions — were built for a world where users click blue links.
That world is shrinking. Pew Research found users click traditional results 8% of the time when an AI summary appears, versus 15% without one. Ahrefs measured top-1 CTR dropping 58% on keywords that trigger AI Overviews.
You need a new measurement layer. Here is what to track.
The old metrics and what they miss
| Old metric | What it measures | What it misses |
|---|---|---|
| Ranking position | Where you appear in organic results | Whether you appear in the AI answer above results |
| CTR | Clicks per impression | Brand mentions without clicks |
| Organic sessions | Traffic from search | Users who got their answer from AI and never visited |
| Domain authority | Link-based authority | AI-specific citation patterns per platform |
| Conversion rate | Site visitors who convert | Users influenced by AI who convert through other channels |
None of these tell you whether ChatGPT recommends you, Perplexity cites you, or Google's AI Overview names a competitor instead.
The AI visibility KPIs
1. Answer inclusion rate
The percentage of monitored prompts where your brand appears in the AI-generated answer — mentioned by name, cited with a URL, or recommended as an option.
This is the AI equivalent of "ranking." Track it per platform, per prompt category, and over time.
2. Share of voice
Of all brand mentions in AI answers for your prompt set, what percentage are yours versus competitors? If ChatGPT mentions five CRM tools and you are one of them, your share of voice is 20%.
Track the trend, not just the snapshot. A rising competitor share is an early warning.
3. Citation frequency
How often your URLs appear in the cited sources list — distinct from brand mentions. A competitor can be mentioned by name while your URL is the one linked.
Citation frequency matters especially on Perplexity (21.87 citations per answer) and Google AI Overviews (4–8 sources cited).
4. Citation position and framing
Being mentioned is not the same as being recommended. Track:
- Are you listed first or fifth in a comparison?
- Is the framing positive, neutral, or qualified ("good for X but lacks Y")?
- Which specific pages get cited?
5. Source diversity
Which of your URLs, social profiles, and third-party mentions appear as sources? If only your homepage is cited but your product pages and docs are ignored, your content structure needs work.
6. AI referral traffic and conversion
Track utm or referrer patterns from chatgpt.com, perplexity.ai, and other AI platforms in your analytics. Perplexity referral traffic converts at roughly 11x organic search — so even small volumes can matter disproportionately.
7. Prompt coverage gap
The set of prompts where competitors appear and you do not. This is your optimization backlog — each gap is a content or positioning opportunity.
8. AI Visibility Score (composite)
Obsurfable combines mention rate, citation rate, competitive positioning, and trend direction into a single score. A composite metric helps executives track progress without reading 50 individual prompt results.
How to set up measurement
Step 1: Define your prompt set.
50–200 prompts covering: product category queries, comparison queries, problem/solution queries, branded queries, and competitor queries. Use real language your buyers use, not SEO keywords.
Step 2: Establish a baseline.
Run all prompts across your target platforms. Record who appears, who is cited, and how answers are framed.
Step 3: Set cadence.
Weekly for competitive categories. Monthly for stable markets. Re-run after every major model launch (Gemini 3, GPT-5.6, etc.) — citation patterns shift without warning.
Step 4: Connect to action.
Measurement without optimization is reporting. Each gap should map to a content action: new page, refresh, earned media outreach, or structural fix.
Mapping old dashboards to new ones
| Old dashboard | New equivalent |
|---|---|
| Search Console → Queries | Prompt set → Answer inclusion rate |
| Search Console → Average position | AI Visibility Score |
| Search Console → CTR | Citation frequency + AI referral conversion |
| Ahrefs → Organic keywords | Monitored prompts with brand presence |
| Brand monitoring → Media mentions | AI share of voice across platforms |
Keep the old dashboards. Organic search still drives traffic. But add the AI layer — because the trend lines are diverging.
The bottom line
If you are still measuring AI impact through rankings and clicks, you are measuring the channel that is shrinking while ignoring the channel that is growing. Answer inclusion rate, share of voice, and citation frequency are the KPIs that tell you whether AI systems are working for or against your brand.