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

# Methodology

> How Alura estimates the numbers Etsy doesn't publish, and how the methods are versioned

Etsy publishes some numbers: a shop's total sales, a listing's price, its reviews and favorites. It doesn't publish how many sales each listing makes, or when. Alura estimates those, and every estimated value in the API has a name starting with `est_`. Everything else is Etsy's data or Alura's own records.

| Page | Version | Covers |
| - | - | - |
| [Sales estimates](/methodology/sales-estimates) | 2026-10 | Sales and revenue of products and shops: lifetime, monthly average, last month and monthly history |
| [Sales trend](/methodology/sales-trend) | 2026-10 | `est_trend_3m`, the last three months against the three before |
| [Keyword market](/methodology/keyword-market) | 2026-10 | `est_avg_monthly_sales` on keywords, `null` for now |

## How versions work

* Each page has a version, written `YYYY-MM`, for the month the method took effect.
* When a method changes in a way that changes numbers, its version changes and the page's version history says what changed and from when.
* Rewording a page or fixing its description doesn't change the version. The page notes the date instead.
* Sales history responses carry `methodology_version` and `methodology_url`, because late reviews can change past months. Other responses don't, so check the page's version if you store estimates over time.
* In the [API reference](/api-reference/research/products/get), every `est_` field, filter and sort value links to its page and version.

## What estimates are good for

Estimates are for comparing products, shops and keywords and for spotting direction. They are not accounting: a seller's own Etsy stats are the only exact source for their own shop. When Alura doesn't know a value, it returns `null`, and a missing value never reads as 0.


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