Separate, private conversations with the right people
Families hold a separate thread with each teacher and staff member, and can reply privately to any post or message. The class administrator decides who can message whom, so a parent reaches the people who can actually help, and no one else.
Translated both ways, and kept for the record
A parent writes in their home language and the teacher reads English; the reply lands back in the parent's language. Every word is saved, and either side can print or export the full history when they need a record.
One conversation, two languages, the parent writes in Tagalog, the teacher reads in English. Tap “Show original text” to reveal what each side actually typed.
Family Messaging: parents reach their own child directly, on the district's terms
Older students already have their own Bloomz account to hear from teachers, so the one edge missing was letting a parent message their own child directly, a quick pickup change, a check-in after practice, without relaying it through a teacher or the front office. ParentSquare built a second, separate app for this (StudentSquare), middle and high school only. Bloomz just unlocks the edge on the account students already have, and it's off until an administrator says otherwise.
Off by default. Turn it on district-, school-, or class-wide — and existing threads close the instant you turn it back off.
Profanity screening with an admin escalation queue
Posts, comments, and messages are screened against a transparent wordlist before they deliver — deterministic, no AI judging anyone's tone. Flagged language warns the sender first; choosing "send anyway" doesn't deliver it, it holds it in a review queue where the group's moderators or a school administrator approve, reject, or — admin-only — suspend the author. Severity decides the outcome: threats are blocked outright with no override, while safety-sensitive language is delivered and quietly alerts your team, because blocking a cry for help is the wrong failure. Off by default; the school decides if and where it turns on.
A deterministic wordlist screens posts, comments, and messages before they deliver — no AI judging anyone. Overriding the warning holds the message for human review instead of sending it. Off by default; your admin switches it on.
How this reinforces the Bloomz pillars
Connected across Bloomz
20 minutes with Bloomz
answers more than 20 pages ever could.
Bring your hardest scenario, a multilingual community, a PBIS rollout, a vendor consolidation, and we'll walk through the actual product.