Services · Support · Hosting & maintenance
Uptime monitoring. We find out before your customers do.
Checks on the journeys that earn money, alerting that reaches a person, and someone whose job it is to answer. A red dashboard nobody is watching is not monitoring.
Nobody notices monitoring until the day they need it
Most monitoring answers one question, is the homepage returning a 200, which is the least interesting failure mode a site has. Checkout can be broken, the contact form can be silently discarding submissions, and the homepage will still be green.
We monitor the journeys that actually matter: can someone complete a purchase, submit an enquiry, log in, load a key template. Synthetic checks that walk the flow rather than pinging a URL.
And we monitor the response, not just the detection. An alert that fires into an unwatched channel at 2am is theatre. Escalation has to reach a human with the access to fix it, which is what makes this part of hosting and maintenance rather than a tool you buy.
How we actually do it
Detect, escalate, respond.
- 01 · Monitor journeys, not URLs
- Synthetic checks that complete the transaction or submit the form, from more than one region. This catches the failures that matter and that a status ping never sees.
- 02 · Watch the things that fail quietly
- Form submissions arriving, payments settling, feeds updating, certificates and domains approaching expiry. Silent failures cost more than outages because they run for weeks.
- 03 · Alert with an escalation path
- Alerts routed to a person with a fallback if nobody acknowledges. Tuned to avoid noise, because a channel that cries wolf gets muted and then ignored.
- 04 · Respond and write it up
- Someone whose job is to act on the alert, then a short written note on what happened and what stops it recurring. Recurring incidents are a build problem, not a monitoring one.
What you get
Deliverables, not adjectives.
- Synthetic checks on your revenue-critical journeys, multi-region
- Silent-failure monitoring: forms, payments, feeds, certificates, domains
- Alert routing with escalation and acknowledgement tracking
- A named response process with defined hours and expectations
- Post-incident notes, and fixes for the causes that repeat
Questions, answered straight
No fluff, as promised.
Is this not just an uptime tool?
Tools detect. The value here is the response, a person who acts, and a process that removes the cause. Buying the tool is the easy ten per cent.
What response times do you offer?
It depends on the plan and what the site does. We would rather agree honest hours we can genuinely hold to than advertise a number we would miss at 3am on a Sunday.
Do you monitor sites you did not build?
Yes, after a short review so we understand the stack and the risks. That review sometimes finds the reason things keep breaking.
Related services
This never works in isolation.
Proof from real work
Clients where this was the job.
Who finds out
first, right now?
Tell us what you're working on. You'll get an honest answer about whether we can help, and exactly how.
Prefer email? hello@saigon.digital