Case study · Digital publishing

The checks someone ran every 30 minutes now run themselves.

A global news publisher's critical reader journeys of sign in, sign out, homepage, and paywall were checked by hand every half hour, around the clock, and logged in a spreadsheet.

~240 hrs
of manual checking eliminated per month
Every 30 min → continuous
journey coverage, day and night
4
critical reader journeys watched: login, logout, homepage, paywall
Failure-only
alerting: silence means all is well
Client
Global news publisher
What we built
Synthetic journey monitoring
Watches
Login, logout, homepage, paywall
Alerts via
Failure-only alerts to the ops team
Status
In production, managed
The problem

Around-the-clock diligence, spent on clicking the same four pages.

Every 30 minutes, on every shift, someone signed in, signed out, loaded the homepage, and hit the paywall, then logged the result in a spreadsheet. Roughly 240 person-hours a month of skilled operations time, spent confirming that things were still fine.

And the coverage still had holes. A paywall failure one minute after a check could sit undetected for another 29, on flows that directly gate subscription revenue.

What we built

The same journeys, walked by software instead.

Synthetic monitors that behave like a reader, and only speak up when something breaks.

Real journeys, not just uptime

The monitors don't just ping a server. They sign in, sign out, load the homepage, and trigger the paywall exactly as a reader would.

Continuous, not every 30 minutes

Checks run around the clock at a cadence no human shift could sustain, shrinking the window a failure can hide in.

Alert only on failure

No green-tick noise, no log to maintain. The team hears about it when a journey breaks, and only then.

The spreadsheet, retired

The manual log is gone. The monitoring history is captured automatically, as a by-product of the checks themselves.

The outcome

A full workweek of checking, given back, with better coverage.

Before

  • Manual checks every 30 minutes, on every shift, forever
  • Roughly 240 person-hours a month on repetitive verification
  • Failures able to hide for up to half an hour between checks
  • Results hand-logged in a spreadsheet

After

  • Journeys verified continuously by synthetic monitors
  • That time redeployed to investigation and improvement work
  • Failures on revenue-critical flows surfaced in minutes
  • History captured automatically, alerting only on failure
Running in production

Going live was day one. It's been looked after every day since.

Login pages get redesigned, paywalls get re-tuned, and synthetic checks that nobody maintains start failing for the wrong reasons, or worse, passing for the wrong reasons. We keep the journeys current as the site evolves. That's the difference between a script and a service.

StatusIn production
MonitoringWatched around the clock
Maintained byThe team that built it

Tell us about the process that's eating your week.

A 30 minute scoping call. No pitch deck, no obligation. Just an honest answer on whether we can fix it.

Book a Scoping Call