Technical content is measured by a content freshness KPI that is reviewed monthly with engineering and content publishing leadership. It's important to note this KPI is an engineering KPI, not just a content team measurement. Content freshness is defined as shown in the table below.
|Most fresh (0-60 days)|
|Fresh (61-90 days)|
|Stale (91-120 days)|
|Abandoned (121 days +)|
You can find freshness data for your service in the CSI team's BI portal: https://aka.ms/skyeye.
How to mark an article as "fresh"
Updates for content freshness are driven by content teams working together with engineering teams. The schedule and cadence of content freshness work is determined by those teams.
You indicate content is fresh by setting the date for the article. In articles published on docs.microsoft.com, you set freshness using the ms.date metadata attribute. Freshness should never be based on an autogenerated date.
What is included in a freshness review?
When you review an article for freshness and update the date, it means:
- You reviewed the entire article and the content is accurate, relevant, and complete for the scope of the article.
- You reviewed all the screenshots and they reflect the current customer-facing product UI.
- All crosslinks go to current, accurate, and relevant content.
- You reviewed customer feedback for the article (verbatims and Livefyre) and have incorporated changes to address the customer feedback.
You should not update the ms.date value for the article when you make typo fixes or minor, quick updates. Content freshness review should be a comprehensive review of the full article - it's your promise and contract with the customer that the content is up-to-date and accurate.