> ## Documentation Index
> Fetch the complete documentation index at: https://docs.getsurfaced.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Quickstart

> From sign-up to your first results, and what to do with them.

Setup is four screens and a wait. The decisions that matter come after, when you choose which questions are worth tracking every night.

**1. Name your company and website**

Sign-up pre-fills both from your email domain where it can. The website is not cosmetic: the audit reads it, and a workspace without a valid one shows onboarding instead of Insights on every screen.

You can skip the website, but you will have no tracking until you add one.

**2. Check what Surfaced read about you**

Your site is analysed into an industry, a product type, an audience, likely competitors and keywords. Every field is editable, and two minutes here pays off: everything suggested later is built on it.

If the analysis fails or is wide of the mark, enter the details by hand.

**3. Choose your prompts**

Surfaced generates a set of questions and tells you how many you can run inside your plan's monthly response allowance.

Pick the ones a real buyer would type before choosing a supplier. Leave the rest. You can add more later, and [Prompts](https://docs.getsurfaced.ai/set-up/prompts.md) explains what makes a good one.

**4. Run the first tracking run**

This usually takes a few minutes. The first run deliberately asks every supported platform, so you get a complete picture before settling into your workspace defaults.

Results appear as they land, so you do not have to wait on the page.

## Then look at four things, in this order

**[Analytics](https://docs.getsurfaced.ai/insights/analytics.md).** Your five headline figures. Read [brand visibility](https://docs.getsurfaced.ai/metrics/brand-visibility.md) first, and the note on [periods and filters](https://docs.getsurfaced.ai/metrics/periods-and-filters.md) before you draw any conclusion from it, because the default view is narrower than most people assume.

**[Prompts](https://docs.getsurfaced.ai/insights/prompts.md), grouped by brand visibility.** This turns one number into a list of questions you win and questions you lose. Open two or three of the losses and read the answers themselves.

**[Competitors](https://docs.getsurfaced.ai/insights/competitors.md).** Click the rival that appears most. You get every prompt where they are recommended and you are not, which is the most directly useful list in the product.

**[Audit report](https://docs.getsurfaced.ai/insights/audit-report.md).** The one screen you can act on today without waiting for another run. It tells you whether assistants can read and quote your site at all.

## What to expect in week one

**One run is a reading, not a measurement.** The same question asked twice returns different wording and sometimes different recommendations. Give it several days before treating a movement as a trend.

**Platforms differ more than days do.** Before concluding something changed, check the per-platform charts. The spread between two assistants on the same day is usually wider.

**Nothing you publish will show up yet.** The engines have to re-read your site before a new page can be cited, and then a run has to ask the questions again. Judge content work over several runs.

## Next

* [How Surfaced collects data](https://docs.getsurfaced.ai/how-surfaced-collects-data.md). What a run does, in five minutes.
* [Metrics overview](https://docs.getsurfaced.ai/metrics/overview.md). The five numbers and what each cannot tell you.
* [Map of the app](https://docs.getsurfaced.ai/app-map.md). Every screen and the page that explains it.


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.