Please tell me about Sysprep after a feature update.

Yokoyama Yosuke (横山 洋介) 0 Reputation points
2025-10-07T00:38:54.9833333+00:00

I perform a clean installation using a VL ISO, then customize the system by configuring Windows settings and installing applications. After that, I plan to perform a feature update to upgrade from Windows 11 23H2 to 24H2, and then run Sysprep to create an image. Would this process be considered a recommended procedure?

Windows for home | Windows 11 | Install and upgrade
0 comments No comments
{count} votes

1 answer

Sort by: Most helpful
  1. Ian-Ng 6,125 Reputation points Microsoft External Staff Moderator
    2025-12-05T16:39:58.3333333+00:00

    Hi @Yokoyama Yosuke (横山 洋介)

    Welcome to Microsoft Q&A forum. 

    Thank you for reaching out regarding your Windows 11 deployment strategy. 

    Regarding your query about performing a feature update (23H2 > 24H2) prior to running Sysprep:  To maximize stability and image integrity, we recommend building your golden image from a clean Windows 11 24H2 installation rather than upgrading a 23H2 reference device and then Sysprepping. In‑place upgrades alter the OS state, especially with Microsoft Store (Appx) provisioning and system configuration, making Sysprep /generalize more likely to hit validation issues or post‑generalize boot issues on 24H2.

    Recommended procedure: 

    1. Clean install Windows 11 24H2 (Volume Licensing media as needed). Windows 11 image creation for OEMs 
    2. Customize in Audit Mode (Ctrl+Shift+F3) and avoid Store app updates or per‑user changes. Sysprep (Generalize) a Windows installation
    3. Run Sysprep (/generalize /oobe /shutdown) and capture the image (DISM/WinPE/MDT).

    This approach yields a generalized, clean image without artifacts from prior versions and aligns with standard Windows deployment methodologies. 

    I hope it clears your concerns. Should you have any further questions or concerns, please feel free to reach out.

    0 comments No comments

Your answer

Answers can be marked as 'Accepted' by the question author and 'Recommended' by moderators, which helps users know the answer solved the author's problem.