Prompt-Level Analytics
Not all prompts are equal. Some consistently mention your brand across every model. Others never do. Prompt-level analytics shows you the per-prompt, per-model breakdown so you know where you're visible and where you're missing.
Start tracking nowAggregate scores hide the real story
A 50% visibility score could mean you're mentioned in half of all prompts across all models. Or it could mean you dominate ChatGPT but don't exist in Gemini. The aggregate number doesn't tell you where to focus.
Different prompts reveal different things. "Best CRM" might mention your brand consistently. "CRM for enterprise teams" might never mention you. Without prompt-level data, you can't identify which customer questions you're winning and which you're losing.
Marketing teams need granular data to take action. Knowing that you're invisible on a specific prompt gives you a target. Knowing your overall score is 50% gives you nothing actionable.
How it works
Every prompt is tracked individually
Each prompt in your configuration runs independently against all 6 AI platforms. Results are stored per-prompt, per-model — not averaged.
See mention status and position
For every prompt-model combination, you see: mentioned or not, position (1st, 2nd, 3rd), and the full AI response text.
Identify patterns
Filter and sort by mentioned/not mentioned, by model, by position. Find prompts where you're strong and prompts where competitors dominate.
What you see in MRKTlabs
How MRKTlabs compares
| MRKTlabs | Profound | Peec AI | |
|---|---|---|---|
| Per-prompt breakdown | |||
| Position ranking per prompt | |||
| Full response text | |||
| Models tracked | 6 models | 2 models | 3 models |
| Prompt-level competitor data |
Start using Prompt-Level Analytics
Set up your brand, run your first scan, and see prompt-level analytics in action.
Request Analysis