Know before customers do
Uptime monitors watch your endpoints and heartbeat checks watch scheduled jobs, so a silent failure gets caught.
hearing it from a customer at 11:40Self-serve · Status page
Forty reports,one reply.
One place to say what's wrong. Monitors catch it before customers do, every report joins one incident, subscribers hear when it changes, and anyone who writes in is told straight away.
Tore opens soon. Join the waitlist and we'll email you the day you can start.
app.tore.ai/northwind/incidents
Example data Reports APIGET /v1/reports · 200 · 1 min checks
Web appapp.northwind.app · 200
Sign-inPOST /auth/link · 200
nightly-export jobheartbeat · checked in 02:00
inbox · reports linked10:02
0reports
"Reports won't load"
"Is something down?"
"Is something down?"
Reports are slow to load
INC-88 investigatingresolved- Raised by the Reports API monitor, before the first report
- 40 conversations linked to this incident
- One reply sent to all 40: it's fixed
My dashboard is stuck loading. Anything going on?
Yes, reports are slow to load right now. The team is on it, it's on our status page, and I'll tell you here when it's fixed.
public page · status.northwind.applive
NNorthwind statusSubscribe
All systems normal
Reports are slow to loadInvestigating · update posted 10:04 by Dan
Resolved: reports load normally againFixed at 10:41 · history kept below
Reportsoperational
Web appoperational
Sign-inoperational
Exportsoperational
312 subscribers · nothing to send312 subscribers emailed: investigating312 subscribers emailed: resolved
Stop answering "is it down?"Say it once, in one place.
When something breaks, the hard part is telling everyone. Tore does the telling, so your team can do the fixing.
Forty reports, one reply
Every report about the same problem joins one incident, and one update goes to everyone affected, sent once.
pasting the same apology forty timesSubscribers told on change
Customers who subscribe hear when the status changes, so they stop writing in to ask.
refreshing a status page all morning"We know" straight away
Anyone who writes in about a known issue is told at once, straight from the incident.
"can you tell me more?"From the first failed checkto the last "it's fixed".
Monitors
Caught by a check, not a complaint.
Point uptime monitors at the endpoints that matter. Give scheduled jobs a heartbeat: they check in when they run, and a missed check-in raises the alarm. The jobs that fail quietly at 2am stop failing quietly.
- Uptime monitors on your endpoints
- Heartbeat checks for cron and scheduled jobs
- A failed check alerts your team before the first report
Monitors · Northwind
Example data5 monitors · 2 heartbeatslast check
Web appGET app.northwind.app · every 1 min
upReports APIGET /v1/reports · every 1 min
upWebhooksPOST /hooks/health · every 5 min
upbilling-sync jobheartbeat · every 1 h · checked in 09:00
on timenightly-export jobheartbeat · every 24 h · last check-in 26 h ago
missedIncident clustering
Every report joins one incident.
When customers start writing in about the same problem, Tore links their conversations to one incident instead of forty tickets. When it's fixed, one update goes to all of them, sent once. A person can edit the wording before it goes.
- Matching reports linked to the incident automatically
- One update to everyone affected, sent once
- Edit the update yourself, or let it draft and send
INC-88 · linked conversations
Example data"The reports page is blank for me"10:02"Reports won't load since this morning?"10:03"Is Northwind down?"10:03+ 37 more, all linked to INC-88
40conversations, one update when it's fixed
Dan edited the update · sent 10:42Reports are loading normally again. Sorry for the slow morning, and thanks for telling us.
Known issues
"We know" replies, straight from the incident.
While an incident is open, Tore AI answers anyone who writes in about it with what the team knows, links them to the incident, and tells them when it's fixed. Anyone on your team can take the conversation over at any moment.
- New reports get the known-issue reply at once
- Each one is added to the incident's update list
- Take over any conversation with one click
Conversation · Beacon Data
Example data
Exports keep timing out. Did I do something wrong?
Not at all. Reports and exports are slow right now and the team is fixing it. I've added you to the update list, so you'll hear from us when it's fixed.
Taking this one, Lena is on a tight deadline. I'll send her a manual export now.
The public page
The current state, and the history behind it.
A public page on your own domain shows what's working now and what happened before. It's linked from the chat widget and the customer portal, so people find it before they write in.
- Current status per component, with history
- On your own domain, in your brand
- Linked from the widget and the portal
- Subscribers told when it changes
status.northwind.app
IllustrationNorthwind · all systems normalpast incidents
- Reports were slow to loadtoday · 39 min
- Delayed email notificationsOct 21 · 12 min
- Scheduled maintenance: database upgradeOct 14 · planned
you@company.comSubscribe
Tune it
Known-issue replies straight awayLink matching reports to the incidentIncident updates a person reviews
2kinds of monitor: endpoint uptime and heartbeat check-ins
1 updateto everyone linked to an incident, sent once
Every changeemailed to the page's subscribers
Your domainfor the status page, portal and help center
Tore's own status page runs on Tore.
Keep exploring.
Next outage,tell everyone once.
Add your monitors and your status page today. When something breaks, the telling happens on its own and your team gets on with the fix.