# Difficulty and volume

> How Ranqo measures a prompt's difficulty from how many competitors AI names in its answers, and what the volume estimate means and does not mean.

Source: https://ranqo.ai/docs/methodology/difficulty-and-volume

Every prompt in your library carries two five-step meters: **Volume** and **Difficulty**. They answer different questions and come from different places. Difficulty is measured from your answers. Volume is an estimate.

## Difficulty

Difficulty says how crowded the answers to a prompt are: how many competing brands AI names when it answers. An answer that lists eight vendors leaves less room for you than one that names two.

### How it is measured

After every run, Ranqo re-measures the difficulty of each prompt the run asked:

1. For each answer to the prompt, on every engine, count the distinct competitors it names.
2. Average those counts over the answers that named at least one competitor, and round to a whole number.
3. Turn the average into a level:

| Level | Hover label | Average competitors per answer |
| --- | --- | --- |
| 0 | No competition | none named in any answer |
| 1 | Low competition | 1 to 2 |
| 2 | Medium competition | 3 |
| 3 | High competition | 4 to 5 |
| 4 | Very high competition | 6 to 7 |
| 5 | Extreme competition | 8 or more |

The new level replaces the old one. A prompt's difficulty describes the latest run that asked it, not an average of its history. Competitors are counted the way every answer is read, up to 20 per answer. See [How Ranqo measures](https://ranqo.ai/docs/methodology).

A failed query that could not be carried forward from a recent answer is left out, because the empty list it leaves is not a measurement. If none of a prompt's answers in a run could be used, its difficulty stays as it was.

### Before the first run

A prompt that has never been asked has nothing to measure. Prompts that Ranqo writes for you carry the writing model's estimate until their first run: a judgement of how competitive the question is for your brand, where a question that plays to your strengths scores lower. When the model gives no estimate for a prompt, the middle of the 1 to 5 scale is stored in its place, which the meter shows as High competition. Prompts you add yourself carry no estimate, and their meter stays empty until the first run. The **Generate with AI** results show these estimates before you track anything.

A paused prompt is not asked, so it keeps the level from the last run that asked it.

### Reading the meter

The Difficulty meter fills one step per level. It is green up to level 2, amber at level 3, and red at levels 4 and 5. Level 0 and a prompt with no estimate both show an empty meter; hover it to see which ("No competition" or "No estimate yet").

Difficulty is about the answers, not about you. A high level means AI names many brands for this question, which usually makes it harder to be named and to rank near the top. It does not say whether you are named: check the prompt's visibility on the **Tracking** tab or its insights page for that.

## Volume

Volume is the writing model's estimate of how often your target audience asks this kind of question, from 1 (very niche) to 5 (very common). It is set once, when the model writes the prompt, and never changes after that: runs do not update it. When the model gives no estimate for a prompt, the middle of the scale is stored in its place.

The Volume column's tooltip calls it estimated search volume, and the meter's hover reads from "Very low search volume" to "Very high search volume". It is not a count from any search engine or AI platform: no engine publishes how often a question is asked. The model is told to judge the question against your own audience and region, not the whole internet, so the most common question in a niche category can still be a 5. It is also told to score questions that come up repeatedly in the community discussions and trends Ranqo researches for your brand higher, and questions inferred without that backing lower.

Prompts you add yourself have no volume estimate, and their meter stays empty. Its hover reads "No estimate yet", but nothing fills it later.

Use volume to compare prompts within your own library, not as a traffic forecast. See [Prompts](https://ranqo.ai/docs/guides/prompts).

## Where you see them

- **Prompts page:** the Volume and Difficulty columns on the Library tab, and on each theme's page.
- **Prompt insights:** both meters in the header of each prompt's page.
- **Generate with AI:** the estimates for each suggested prompt, before you track it.