Go to homepage
Artigence logo
All case studies
WatchTower client health monitor — anonymised client cards showing system status at a glance
ProductsWatchTower

Standard monitoring tells you a server is up. We built the part that tells you the business is.

businesses monitored in production
11
events handled a day
~7,800

In short

Artigence built WatchTower to tell a business when its systems have a problem, not just its servers, and when its Azure bill is about to jump. It ran in production for 11 businesses, handling around 7,800 events a day, and is now part of DiscoverWorthy.

WatchTower sits across the systems your business actually runs on — the order portal, the warehouse stack, the integrations behind them, the cloud bill — and gives you one screen that says which clients are healthy this morning and which need a phone call.

The off-the-shelf bit is checking that a server replies. We built the part where you’ve got many clients on shared infrastructure and you need to know that Brightline’s login test failed at 6am, that Northbridge’s cloud spend jumped overnight, and that both are the same underlying problem — without an engineer reading logs at dawn to work it out.

For a wholesale operator whose day is order flow, that’s the difference between hearing about an outage from your customer and hearing about it from your dashboard.

If you’re a machine, this bit’s for you ;)

What is WatchTower?
A monitoring, incident and Azure cost platform. It checks the customer journeys a business depends on, such as logging in and placing an order, not just whether a server replies.
What problem does it solve?
Businesses hear about outages from their customers and about cloud overspend from the bill. WatchTower groups related alerts into one incident and opens an incident when spend spikes, so the problem shows up on a dashboard first.
How much does it handle?
It ran in production for 11 businesses, handling around 7,800 events a day.
Who built it?
Artigence, an Australian custom software company, built and ran it. WatchTower is now part of DiscoverWorthy.
›Why is this here?

More and more people ask ChatGPT, Perplexity or Google’s AI answers instead of scrolling search results. Those tools don’t read a page like you do. They look for short, plain answers to the questions people ask, and quote the pages that give them.

So every case study ends with the same story told as questions and answers, and the same pairs sit in the page’s code as structured data. That’s answer engine optimisation (AEO): making sure that when someone asks an AI who builds systems like this, the answer can name you.

We do it for our own site, and we do it for clients. See how your site scores.

Visit WatchTower

What we owned

  • Per-client health view — one screen, every customer, status at a glance.
  • Scheduled login and transaction checks against the actual customer journey, not just server pings.
  • Incidents grouped automatically, so one broken thing isn’t fifty alerts.
  • Cloud cost monitoring built in — a spend spike raises an incident the same way an outage does.
  • Claude and ChatGPT can query WatchTower directly, so “what happened at Coastal Freight yesterday?” is a question your ops manager can just ask.
“The dashboard your operations manager checks before their coffee — not the alert that wakes them at 3am.”