Verified against Midjourney · 2026-08-08
Manage Midjourney's personalization profile for consistent team brand output
A team-workflow brief for handling Midjourney's always-on personalization — the account-level learned aesthetic bias that quietly shifts default output toward one person's trained taste — so a shared brand asset generated by different team members doesn't silently drift depending on whose account ran the prompt.
The prompt
Ready to copy — highlighted parts are example details you can swap.
WHAT THIS ASSET IS FOR
a set of hero images for a client's product launch, generated by three different team members across the same week — state clearly that this is a shared brand or client asset rather than personal creative work, since that distinction is exactly what determines whether personalization should be on, off, or shared across this task.
THE ACTUAL CONTENT REQUEST
a clean studio product shot of the client's new wireless speaker on a pale grey backdrop
WHY PERSONALIZATION MATTERS FOR THIS JOB
Every Midjourney account accumulates its own personalization profile from that account's past ratings and generation history, and that profile quietly biases default output toward that specific account's learned taste — two team members running the identical text prompt from their own separate accounts can get visibly different results purely because of this account-level bias, with no wording difference responsible for the gap at all.
PERSONALIZATION HANDLING FOR THIS JOB
shared team code — every team member should use the same --p code so results stay consistent regardless of whose account generates them — decide and state one of three approaches: turn personalization off entirely with --p off for a neutral, account-independent baseline; use a shared team personalization code with --p a1B2c3D4 so every team member's generations pull from the same trained aesthetic profile regardless of whose account runs the prompt; or, if brand consistency doesn't require this level of control for this particular asset, note explicitly that personalization is being left on the individual runner's account and any resulting variation between team members' outputs is expected and acceptable.
STYLE ANCHOR
paired with a locked --sref value from an approved prior hero shot, for a second, independent consistency guarantee beyond personalization handling alone — for genuinely brand-critical work, pairing an explicit --sref style reference with either --p off or a shared --p code gives two independent, complementary consistency guarantees rather than relying on personalization handling alone.
WHAT TO AVOID
--no text, watermark, props, a second unit of the product
PARAMETERS
--p a1B2c3D4 --no text, watermark, props, a second unit of the product --ar 4:5 --v 7
OUTPUT
a clean studio product shot of the client's new wireless speaker on a pale grey backdrop --p a1B2c3D4 --no text, watermark, props, a second unit of the product --ar 4:5 --v 7
TEAM DOCUMENTATION NOTE
Whatever approach is chosen for shared team code — every team member should use the same --p code so results stay consistent regardless of whose account generates them, write it down where the rest of the team generating assets for this same brand or client can see it — an undocumented, informal convention ("we all just leave personalization on") is the exact condition that produces silent, hard-to-diagnose output drift between team members months into a project, once nobody remembers whether a shared approach was ever actually agreed on.Customize
Optional — swap in your own details for the highlighted parts above.
Why this works
Midjourney's personalization system trains an account-level aesthetic bias from that account's own rating and generation history, and it is applied by default — this is the direct, mechanical explanation for a specific, confusing team problem: two people on the same brand project, running what looks like an identical text prompt, get results that lean noticeably differently in color, composition preference, or overall polish, purely because their two accounts have each independently trained their own personalization profile from unrelated past use. Without understanding this mechanism, that gap gets misdiagnosed as a wording inconsistency between the two people's prompts, and hours get spent trying to fix wording that was never the actual source of the difference.
Naming three explicit handling options — off, shared code, or accepted individual variation — rather than leaving the decision implicit gives a team a concrete choice to actually make and document, instead of defaulting by omission to "whatever each person's account happens to do," which is the option most likely to produce silent, accumulating drift across a project's asset set specifically because nobody consciously chose it. A shared --p code is the option that actually solves cross-person consistency directly, since it points every team member's generations at the same trained aesthetic profile regardless of whose personal account is technically running the command — closer to a shared brand style guide than to any one person's individual taste.
Recommending a companion --sref alongside whichever personalization approach is chosen, rather than treating personalization handling as sufficient on its own, reflects that the two mechanisms solve genuinely different, complementary problems: personalization handling controls which account-level aesthetic bias is in play, while --sref controls a specific, chosen visual style transferred onto new content — using both together for brand-critical work closes two independent gaps rather than assuming one setting compensates for the other.
The closing instruction to document whichever approach is chosen, rather than treating the decision as a one-time fix, targets the actual long-term failure mode this workflow exists to prevent: an informal, undocumented team convention is exactly the kind of decision that quietly stops being followed months into a project once new team members join or the original context is forgotten, and by the time the resulting drift becomes visible across a large asset set, it is far more expensive to diagnose and correct than writing the agreed approach down once at the start would have been.
Verified against
Midjourney v7 · 2026-08-08
Changelog
- 2026-08-08 — Initial publish, verified against Midjourney v7 --p personalization handling for team brand-asset consistency.
Need this built into your business?
If a prompt isn't enough — what Scult builds, built and maintained for you — that's Scult's day job.
EXPLORE WHAT SCULT BUILDS
