Toil And Trouble, Catchpoint Clears The Mist On The Internet Performance Cauldron

Shakespeare wasn’t online. But if he was, he might have been thinking about Internet and network system interconnectivity when he wrote the opening lines for the witches at the start of Macbeth.

When shall we three meet again… in thunder lightning or in rain? When the hurlyburly's done, when the battle's lost and won.

The hurlyburly in this case is the ever-present need for enterprise organizations to keep their web presence connected and online with site reliability engineers being physically and technically capable of addressing downtime issues so that the battle is more often won than lost.

Technical Toil & Trouble

In keeping with the witch’s cauldron theme, Internet performance monitoring company Catchpoint says that toil (and trouble) levels are on the rise despite the widespread proliferation of AI-based intelligent automations being applied inside modern enterprise IT stacks. The company’s 2025 SRE (site reliability engineering) report suggests that after five years of steady decline, the median reported percentage of “work spent on toil” has increased to 30% from 25% in 2024. This trend arguably raises questions about the practical impact of AI on daily workloads.

Always keen to tell us that “slow is the new down” when it comes to Internet performance and the subsequent lag that this effect has on enterprise applications, Catchpoint’s market analysis of several hundred executives across system architecture and engineering roles as well as management, agrees that poor performance is as harmful as downtime and elevating user experience is now a key reliability metric.

“Success starts with individuals owning their role in the bigger picture and that starts with embracing SRE as more than a technical enhancement process,” said Mehdi Daoudi, CEO and co-founder of Catchpoint. “When [IT operations and developer] teams understand how their work drives outcomes, it becomes easier to align around the opportunities that matter and the steps to seize them, and what’s a major concern this year is that organizations are feeling pressured to prioritize release schedules over reliability.”

Now in its seventh year, Daoudi calls his firm’s SRE market analysis the “authentic and independent voice of the reliability community” as it aims to underscore the role of SRE as an indispensable practice in maintaining high-performing, resilient digital services and applications. This year's report highlights challenges and opportunities facing these teams in an era marked by rapid technological advancement and escalating performance expectations.

So what are the operational problems (okay let’s say challenges) in network-connected IT teams today?

Real IT Team Challenges

One of the key issues is pressure.

While IT teams are usually happy enough to be left running to schedule and be able to release enterprise application updates on what we might imagine to be a fairly well-defined cadence and schedule, hiccups (for want of a more serious ailment term) often occur. As a result of pressures typically emanating from an organization’s business function which are then filetered down through the IT function, over two-thirds of professionals surveyed by Catchpoint acknowledge frequently feeling pressured to prioritize application and data service release schedules over reliability. This says the company reflects the ongoing struggle between agility and stability.

A further issue is thrown up due to the number of monitoring tools a modern enterprise usually runs. Because most reasonable sized enterprises might use somewhere between two and ten monitoring or observability tools, emphasizing a “value over cost” mindset for effective monitoring across technology stacks, we actually end up with too many cooks (or indeed witches) spoiling the broth in this case.

Burdensome Toil Exacerbation

“What was most eye-opening from our report findings this year was that, for most teams, it seems the burden of operational tasks has grown for the first time in five years,” said Leo Vasiliou, Director, product marketing at Catchpoint and author of the SRE Report. “The expectation was that AI would reduce toil, not exacerbate it.”

Looking at the situation in hand then, Vasiliou bemoans whet the team calls a “continued misalignment on reliability priorities” today, with the root cause of those misalignments very often being a disconnect between the IT department and business management. While overall responses paint a positive picture of reliability practices, significant differences emerge when analyzed by managerial responsibility, highlighting a gap in alignment on priorities and approaches notes Vasiliou and team.

When it comes to actual Internet-related incident reporting and response, some 40% of respondents said they were handling between one to five 5 incidents in the last 30 days. Notably, incident response is a shared responsibility across all business operations levels, with higher-level managers just as involved as individual employees.

The Internet Is A Stack

What we should perhaps take away from these reports comes down to two factors.

First, Catchpoint is an Internet performance monitoring company and there are a whole host of Internet performance monitoring hiccups (again, a more serious term would be preferable) happening at any one time. So the company is naturally disposed to highlight areas where firms are failing to monitor the health and wealth of the web pipes that they draw their IT lifeblood from in a way that is commensurate with the amount of transactions and commerce that they put through these channels.

Second, Catchpoint is an Internet performance monitoring company because - across the whole "stack" of the Internet itself - service connectivity, bandwidth, scope, resilience and flexibility are real issues and tools such as real user monitoring (pleasingly shortened to RUM), tracing, “Internet synthetics” (methods used to prototype web services to see how they perform under poor, moderate, good or very good coverage) and tracing are all in the company’s wheelhouse to help provide what it calls high-fidelity data and flexible visualizations with advanced analytics.

Switching Shakespeare plays from Macbeth to Hamlet if the artistic licence allows it. The prince of Denmark’s last line clearly showed how much he valued being online and connected when times are good… otherwise, the rest is silence.