---
title: "What is feature freshness? | Snowplow Signals"
description: "Feature freshness is how recently a served feature reflects what a customer did. A stale feature describes a past version of the customer, not the current one."
source: https://signals.snowplow.io/learning/feature-stores/what-is-feature-freshness
---

[Learning](https://signals.snowplow.io/learning) / [Feature stores](https://signals.snowplow.io/learning/feature-stores) 

# What is feature freshness?

Feature freshness is the lag between something happening and the served feature reflecting it. The fresher a feature, the more recent the behavior a model can act on. Stale features make a model answer for a customer who has already moved on.

## What freshness means

Freshness is the time between an event happening and the feature that depends on it being updated. A feature updated once a night is up to a day behind. A feature updated as events arrive is current to the last few seconds. Both can be correct; they describe the customer at different points in time.

## Why it decides what a feature can do

A model can only act on what its features tell it. If the features are a day old, the model answers for yesterday’s customer, even when the person is on the page now. Freshness sets the ceiling on how reactive anything built on the feature can be.

## The update latency nobody quotes

Read latency, how fast a served feature comes back, is quoted often. Update latency, how fast a new event changes the served value, is quoted far less, and it is the half that decides freshness. A feature store is only as current as the pipeline feeding it.

## Put it to work

* [Real-time attributes on Signals →](https://signals.snowplow.io/products/real-time-attributes)

## See it on your own traffic 

Add the SDK, define one attribute, and see an attribute updating against your own traffic in the first session. The free tier runs on the real engine: no card, no sales call.

[Start free](https://signals.snowplow.io/start) [Talk to an engineer](https://signals.snowplow.io/start?intent=engineer) 

More explainers in the [Learning Center →](https://signals.snowplow.io/learning)