Know before your customers do

Why we built it

SSAMT started with an outage that should have been found in minutes and was found by a customer. An internet link went down. The business on the other end of it could not trade, its customers could not reach it, and the hosting company only found out when someone phoned to tell them. The people paying for the service were the ones reporting that it was broken.

There was monitoring. It just sat in the wrong place. It ran inside the same network as the link it was meant to watch, so when the link dropped, the monitor dropped with it and had no way to tell anyone. From the inside, a dead line looks exactly like silence.

That taught us the lesson SSAMT is built on: monitoring has to watch from somewhere else. Something outside your network has to notice when your network disappears, and something inside it has to see what the outside cannot. So we built a small service that did both, and pointed it at that link.

It worked, so we kept going. The firewalls came next, then the branches and their LTE backup lines, then the nightly jobs and backups that fail without making a sound, then the status pages and reports our clients' own customers wanted to see. Clients asked whether they could use it themselves. What began as a fix for one bad morning has grown into a full monitoring platform, and it is the tool we now use to look after every system we are responsible for.

If the customer is the first to know, you found out too late. SSAMT exists so that you hear it first, from something that was watching.

What it watches

From the internet, the way your customers see you

Website and API checks look at status codes, keywords and returned values, so you know the site actually works rather than just that the server answers. Ping, port and DNS checks cover your servers and services, with real latency, jitter and packet loss. TLS certificates are checked on any port, and domains are watched for expiry, because both take a working service offline on a date you could have read off a calendar.

Inside your network, where the outside cannot see

A small on-site node runs on any always-on Windows PC or server, Linux machine or Docker host. It watches private servers, intranet sites and LAN devices, and it only connects out over HTTPS, so there is no port to open and nothing new exposed to the internet. Network discovery scans the private subnets you choose and lists every device with its address, MAC and maker, name, open ports and what it appears to be, whether a printer, camera, PC or network device. After the first scan it emails you when something new connects, and any device can be monitored with a single click.

Branches, firewalls and internet links

Add each branch with its fibre and LTE links, and SSAMT tells you the moment a site fails over to its backup line, loses its redundancy or goes down altogether, rather than three weeks later when the data bundle runs out.

Speed and line quality: up is not the same as usable

The on-site node runs scheduled speed tests from each branch, anywhere from hourly to daily, against Cloudflare's nearest server: download, upload, latency and a bufferbloat grade that shows why calls break up the moment someone starts a download. It alerts you when a line is slower than it should be. It also measures latency, jitter and loss to the gateway and to the ISP, scores them as call quality (MOS), and times how long Microsoft 365 and your own applications take to respond from the office. That is the difference between "the line is up" and "the line is usable". Speed tests use the branch's own data, and the dashboard shows how much.

Backups and scheduled jobs

The most expensive failures are the ones that make no noise: the backup that did not run, the integration that never started. Heartbeat monitors expect each job to check in when it finishes, and raise the alarm when one stays silent.

Security exposure

An exposed-ports audit catches RDP or admin pages left open to the internet. Blacklist checks tell you when your mail server's IP has been listed, usually before your emails start disappearing into spam.

Why it went down, not just that it did

The on-site node reads your firewalls, switches and UPSes over SNMP (v2c or v3) and receives their syslog, so reboots, power failures, link drops and configuration changes become evidence. When an outage ends, SSAMT drafts a root-cause report with the cause, the impact, the timeline and how to prevent it. You review it and send it to the client in one click, or print it to PDF. Firewall traffic logs never leave your network.

Alerts that make sense

Alerts go out by email and Telegram according to severity, with reminders until someone acts. For the outages that cannot wait until someone checks their inbox, SSAMT sends an SMS, a WhatsApp message or places a voice call. On-call rotas decide who that is: alerts go to whoever is on call this week and escalate if they do not respond, with swaps and leave handled in the rota rather than in a group chat. Dependencies mean one outage raises one alert: when a branch's link goes down, you hear about the link, not about forty devices behind it. Maintenance windows, one-off or recurring, keep planned work from waking anyone up.

Insights: the record over time

Availability, incidents, downtime, time to recover and root causes over 7, 30 or 90 days, for everything you watch or for a single customer, measured against their SLA. It turns "the internet is always down" into a number, and a number into a decision about the line, the provider or the hardware.

Showing your customers the record

A public status page with uptime history, incident updates, maintenance notices and email subscriptions. Monthly SLA reports per customer, emailed automatically with a CSV and printable to PDF. Groups per customer, read-only customer logins limited to their own services, a weekly digest for your team, a wall display for the operations room, two-factor sign-in, and a REST API for your other tools.

Running in five minutes

1. Sign up. Create a company account and start the 14-day free trial. No card and no contract.
2. Add what matters. Paste a URL, IP address or host name, or add a branch with its links. For your internal network, run the on-site node. SSAMT builds the checks for you.
3. Hear it first. Get the alert before the phone rings, show your customers a live status page, and send them a clear report after every outage.

Pricing

Two plans, each with a 14-day free trial of everything in it. Prices are in rand and exclude VAT. Paying annually gives you two months free.

Both plans include website, API, port, ping, DNS, heartbeat and certificate checks; email alerts with severity levels and reminders; your own status page with incident updates; maintenance windows; dependencies; insights on availability, incidents, downtime, time to recover and root causes over 7, 30 or 90 days; sites, firewalls and internet links with failover alerts; on-site nodes; speed tests with a bufferbloat grade, line quality (MOS) and web experience from inside the network; network discovery with new-device alerts; and two-factor sign-in.

Professional adds the exposed-ports audit, IP blacklist and domain expiry checks; Telegram alerts; SMS, WhatsApp and voice-call alerts for serious outages (100 a month); escalation to on-call contacts; on-call rotas with swaps and leave; email subscriptions for your status page; monthly SLA reports for your customers; a weekly digest; REST API keys; read-only customer logins; SNMP and syslog evidence; and drafted root-cause reports for every outage.

Run it yourself, or let us run it

SSAMT is built to be self-service: an IT team or an MSP can sign up and be watching its first services within the hour. But a monitoring tool on its own only does half the job. The other half is deciding what matters, who is told, and what they do next, and then doing it at two in the morning.

If you would rather hand that over, we run SSAMT for you. We map what your business depends on, from the internet line to the nightly jobs, install the on-site nodes, set the checks and tune the alerts until every one of them means something. Then we receive the alerts, act on them, and send you a monthly report of what failed, what we fixed before anyone noticed, and what is trending towards trouble. It sits naturally alongside our application support and infrastructure and security work, and is quoted on the size of what is being watched.

The thinking behind all of this is set out in when the customer calls first, it's too late.