Kastr vs SchoolMessenger: an honest read, including where we lose
If your primary requirement is a mass emergency notification system — the thing your safety plan names, the thing that has to make thirty thousand phone calls during a lockdown — you should buy SchoolMessenger and not us. That is the conclusion, on the first screen, because the rest of this page is only useful once it is out of the way.
| Requirement | Belongs to emergency notification | Belongs to everyday communication |
|---|---|---|
| Dedicated emergency console with restricted access | Yes | No |
| Two-person authorisation before a district-wide alert | Yes | No |
| Published uptime commitment and load-test evidence | Yes | Desirable |
| High-volume outbound voice at short notice | Yes | Occasionally |
| Integration with sirens, CAP feeds and public alerting | Yes | No |
| Daily two-way conversation with families | No | Yes |
| Translation with a preview before sending | No | Yes |
| Cost visibility before a send | No | Yes |
| Public API for district automation | Sometimes | Yes |
| Tamper-evident audit of who sent what | Yes | Yes |
| Kastr today: everyday column yes; emergency column no, except the last row | ||
Kastr has no emergency console, no two-person authorisation, no published uptime figure and no load-test evidence. We have an emergency flag on a broadcast that bypasses quiet hours and family channel preferences, gated behind its own capability. That is a smaller claim than an emergency notification system and we are not going to inflate it.
Buy SchoolMessenger instead if any of these is true
Stated plainly, because a district that buys us for the wrong job discovers it at the worst possible moment:
- Your district safety plan names a mass notification system as a control. Ours is not one and we will not let you write our name into that document.
- You require a dedicated emergency console with its own access model separate from everyday sending. We do not have one.
- You require two-person authorisation before a district-wide alert can leave. We do not have it.
- Your RFP asks for a published uptime percentage or load-test evidence. We publish neither, because we are pre-launch and have neither. Any figure we quoted would be invented.
- You need SSO, SAML or MFA for staff. We have a magic link and nothing else.
- You need integration with public alerting infrastructure — sirens, CAP feeds, municipal systems. Not our category.
SchoolMessenger is PowerSchool-owned with a long heritage in exactly this work, particularly high-volume outbound voice. That lineage is worth something real when the requirement is a phone call to every household in twenty minutes.
What we do guarantee at the delivery layer, precisely
We will not claim an emergency capability we do not have. We will describe the delivery engine exactly, because it is checkable and because most vendors describe theirs in adjectives.
- Claimed work that recovers. Workers claim rows with
SELECT … FOR UPDATE SKIP LOCKEDunder a five-minute claim TTL. A worker that dies mid-batch releases its rows automatically rather than stranding them. - Network I/O outside the transaction. A slow carrier does not hold a database transaction open, which is the usual cause of a queue collapsing under its own locks.
- Row-level idempotency. The unique key is broadcast, person and channel. A retry after a partial failure cannot deliver the same message to the same family twice — which sounds academic until an emergency send double-fires at six in the morning.
- SMS to voice failover. On terminal SMS failure the message escalates to a voice call, with the failover linked back to the original attempt so your reporting shows what actually happened.
- Quiet hours, defaulting to 21:00–07:00, bypassed only by an emergency-flagged send, which is itself gated behind a separate capability so not everyone who can post a newsletter can wake a district at 3am.
- Tamper-evident audit. Every state-changing action is chained with SHA-256 over the previous hash and the canonical event, serialised under an advisory transaction lock, with UPDATE and DELETE revoked at the database role and no update policy in row-level security.
That is a strong everyday-communication engine. It is not a certified emergency notification platform, and the distinction is the entire point of this page.
The two-vendor stack, and whether it is worth it
Plenty of districts already run this pattern without naming it: a notification system that fires four times a year, and a family communication platform that runs every day. The question is whether paying for both is defensible.
The arithmetic at our rate: a 6,000-student district pays $19,500 a year for everyday communication, on top of whatever the emergency contract costs. An 18,000-student district pays $54,000. Whether that is worth carrying alongside an emergency vendor depends on what your emergency vendor's everyday product is costing you in staff time and unmeasured reach — systems built for four sends a year tend to be built for authority rather than for a front-office administrator sending eleven messages before lunch.
The counter-argument is real and we will make it for you: one vendor means one contract, one integration, one security review and one support number. If your district is small, or your technology team is two people, that simplicity is worth more than any feature comparison. Ask what the bundled everyday product actually costs as a separate line, and whether it would survive an evaluation on its own merits.
The question that resolves it. Count messages, not incidents. If your district sends fewer than twenty non-emergency messages a month, one vendor is the right answer. If your front office sends twenty before Wednesday, the everyday product is the one you are actually buying, and it should be evaluated as such.
Questions people actually ask
Is Kastr an emergency notification system?
No. We have no emergency console, no two-person authorisation, no published uptime commitment and no load-test evidence. We have an emergency flag on a broadcast that bypasses quiet hours and per-family channel preferences and is gated behind its own capability. If your safety plan requires a mass notification system, buy one.
What does SchoolMessenger do that Kastr does not?
Certified emergency notification with a dedicated console, high-volume outbound voice at short notice, single sign-on, published operational commitments, integration with public alerting infrastructure, a large installed base, and a completed SOC 2 audit through PowerSchool. We have none of those.
Can a district run SchoolMessenger and Kastr together?
Yes, and for larger districts it is often the sensible pattern: the emergency system for emergencies, an everyday platform for the several hundred routine messages a month. Our REST API and signed webhooks are designed for exactly this kind of coexistence.
Does Kastr publish uptime figures?
No. We are pre-launch and have no production operating history, so any percentage we published would be fabricated. When we have twelve months of real data we will publish it with the methodology attached. Treat any pre-launch vendor quoting an uptime number with suspicion, including us if we ever do.
Does Kastr support SSO or MFA for staff logins?
No. No SAML, no OIDC, no MFA. Authentication is a magic link: 32 random bytes, a 15-minute single-use token, atomic consume that defeats prefetch and replay, same-device binding by hashed cookie, and no account enumeration. It is a well-built magic link and it is still not SSO. For many districts that is a disqualifier and we would rather you knew now.
How does Kastr handle quiet hours for urgent messages?
Quiet hours default to 21:00–07:00 and suppress routine sends. A broadcast flagged emergency bypasses both quiet hours and per-family channel preferences, and that flag is gated behind a specific capability rather than available to every sender. Routine traffic pushed through the emergency path trains families to ignore it, which is why the two are separated.
One price. Every feature. Locked for three years.
$3.50 per student per year under 5,000 students. No tiers, no add-on modules, no per-message fees. Published on the site because you should not have to book a call to learn a price.