Dashboard anxiety plagues IT pros' nights, weekends, vacations
- Reference: 1757574912
- News link: https://www.theregister.co.uk/2025/09/11/dashboard_survey/
- Source link:
Early mornings, late evenings, weekends. Useless users always demand support [1]READ MORE
More than half of the 616 IT professionals surveyed (52 percent) said they checked dashboards during nights, weekends, or vacations, with 59 percent saying past outages had left them more obsessive about making sure that everything is working. A third of IT pros said they felt compelled to check in at least once an hour.
But that's OK – it should be possible to get alerts when things aren't looking healthy, right?
While a whopping 62 percent of respondents said a dashboard alert had helped prevent a major outage, even more – 76 percent – said the relentless pings disrupted their personal lives, particularly during evenings and weekends. Almost half (43 percent) reported receiving alerts multiple times a day, which often leads to notification overload.
One user said: "False positives or unnecessary alerts constantly interrupt our flow. It's exhausting."
[2]
A worrying 30 percent of respondents said they'd experienced downtime because they hadn't reviewed their dashboard before a problem went from being a warning to a full-blown outage.
[3]Not again! Microsoft blames config tweak for 365 outage in parts of North America
[4]News from a possible future: 'Rampant jellyfish cause AI outage by taking datacenter offline'
[5]IBM Cloud hit by Severity One incident with the same symptoms as other recent SNAFUs
[6]Google agrees to pause AI workloads to protect the grid when power demand spikes
[7]14-hour+ global blackout at Ingram Micro halts customer orders
The study was commissioned by [8]Liquid Web , a hosting outfit that is no stranger to dashboards. CTO Ryan MacDonald called for prioritization in dashboard user interface design as a way to reduce friction. He said: "The next generation of dashboards won't just monitor infrastructure, they'll restore control."
MacDonald also pointed to the use of AI - of course he did - as a way to "boost confidence and reduce noise."
[9]
[10]
Perhaps. Just over half (54 percent) of respondents believed that AI-generated summaries would save time, and slightly fewer (53 percent) said they'd speed up incident response.
Ubuntu users left waiting after Canonical's servers take weekend off [11]READ MORE
However, while 34 percent would trust AI summaries more than a human technician (which arguably says more about the humans involved), 27 percent would trust them less. 39 percent remained unsure.
Although the survey highlights the inability of IT professionals to switch off, the solution – better dashboards and reporting – might not address the core problem. Rather than reducing friction in the user interface, the question needs to be asked: why do IT pros have so little faith in their systems and alerting workflows that they feel the need to keep checking in, even during downtime?
Liquid Web might have suggestions, but a lack of system resilience also needs to be addressed before harassed admins can truly relax and turn away from the baleful glow of the portal. ®
Get our [12]Tech Resources
[1] https://www.theregister.com/2025/02/05/on_call_out_of_contract/
[2] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_offprem/front&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=2&c=2aMKdtprfVMhPMUteye5FtAAAAEI&t=ct%3Dns%26unitnum%3D2%26raptor%3Dcondor%26pos%3Dtop%26test%3D0
[3] https://www.theregister.com/2025/08/21/microsoft_365_outage/
[4] https://www.theregister.com/2025/08/12/france_nuclear_reactor_jellyfish_shutdown/
[5] https://www.theregister.com/2025/08/12/ibm_cloud_severity_one_outage/
[6] https://www.theregister.com/2025/08/04/google_ai_datacenter_grid/
[7] https://www.theregister.com/2025/07/04/ingram_micro_technical_difficulties/
[8] https://www.liquidweb.com/
[9] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_offprem/front&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=4&c=44aMKdtprfVMhPMUteye5FtAAAAEI&t=ct%3Dns%26unitnum%3D4%26raptor%3Dfalcon%26pos%3Dmid%26test%3D0
[10] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_offprem/front&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=3&c=33aMKdtprfVMhPMUteye5FtAAAAEI&t=ct%3Dns%26unitnum%3D3%26raptor%3Deagle%26pos%3Dmid%26test%3D0
[11] https://www.theregister.com/2025/09/08/canonical_server_outage/
[12] https://whitepapers.theregister.com/
Notification overload
This is a huge issue for IT admins across the field (IMO).
We are generating so much data across our infrastructure, and most of these systems independently have their own severities and some default settings when to notify admins. Across the board they spam you until you're numb... Solutions like XDR in the case of cybersecurity can help, but in my experience companies rarely go through the effort of dialing it in to only generate relevant notifications. At least this means you get spammed from one consolidated platform!
"why do IT pros [..] feel the need to keep checking in"
Because when the system goes down, they are first in the line of fire for blame and furious calls to get things working - which only prevents them from actually solving the problem.
But it's comforting for upper management to be able to yell into a phone and, minutes later, see that everything is working again. They get the idea that they actually did something. Apparently, it doesn't happen to them very much.
Re: "why do IT pros [..] feel the need to keep checking in"
And will they be the same managers who regularly refuse to fund training or reliability upgrades or allow downtime for testing? Or maybe they're the ones who refuse to have a proper server room or sweat kit beyond its natural lifetime. Or perhaps the ones who expect 24/7 service from their IT but only pay for 8/5.
If you're working for a manager like this who also shouts at you for doing your job and you're not getting paid for being on-call then my advice would be to ignore the dashboards and find a better job.
I'm seeing something a little more worrying.
The overuse of dashboards. Everything *has* to have a dashboard. Where I work, there are so many damned dashboards it's hard to know what to look at. But hey, management *love* a dashboard.
The srtange thing is, when the dashboard system goes offline, all hell breaks lose. We spend more time diagnosing that then looking at the problems the damned thing is supposed to be alerting us about in the first place.
Management also seem to forget that obtaining data for said dashboard(s) has an impact on the system being queried. You have to wonder at which point the dashboard and it's intrusiveness outweighs the value it's supposed to add.
The other option is email alerts - one system sends me almost 1000 emails a day, trying to get that to feed into a Seim so the software can reduce the noise somewhat….
In a design/manufacturing company I worked for we had the usual monthly reviews with slides and such but we'd just introduced 3 new models on one of the production lines and things weren't going well at all. The directors said that monthly reporting was too laggy so they decided they wanted a huge physical dashboard at the entrance to our section which they would check out daily. This would be good for moral too because we'd all see that they were helping. The head of quality was tasked with writing a "Metrics Plan" for what data would be shown, who would compile it, print it, pin it to the board, how often it would be reviewed, etc. There was a big meeting to discuss the draft Plan and the manager who ran the production line in quesion said that the metrics, collection, printing, posting looked OK, if a bit onerous and he might have to recruit an admin to do it but, he said, there was a whole section missing. After puzzled looks he said that there needed to be a section on exactly what the directors would do when they saw that various metrics were outside their carefully calculated limits; that these actions would need agreed response/clearance times; that there needed to be a section on the board for how well the directors were performing against their targets. Once that was written and approved he'd have no problem signing off the plan.
The big, physical dashboard never happened.
Alerts. What's wrong with us lerts?
More years ago than I like to think, we had a troublesome installation with the customer threatening severe sanctions if the issues were not resolved. We were given a set period (six months) where there must be no faults which prevented machinery operation. Although data-acquisition was not part of the contract, we installed monitoring with remote-access at our own expense.
During preliminary weeks of 'sorting everything out', before the trial, we identified several customer issues including operator deliberate intervention to cause 'emergency overtime'. Also, their station maintenance schedule inadvertently came under our scrutiny..... Clearly they were not keeping up their part of the contract.... Once the operators found out that we could follow what they were doing, the number of faults unexpectedly reduced.
When the actual trial did begin, I found myself logging in at strange hours in order to see what was happening. I received an alert during the night of a major fault and sent a message to our service staff to get to the site immediately. They arrived about 5:00 a.m. and quickly identified the problem (caused by external/customer issues), cleared the problem, reset the particular machine, checked operations and reported to the customer's control room. Who were asleep.....
That morning the customer's project manager phoned me up to angrily demand what we were going to do about the fault. I'm sure I heard him choke on his coffee when he found out it was his problem, and as a favour, we had already sorted it.
I have a certain sympathy for the operators; the new automated system was taking over their jobs, as new technology can and does. But these were skilled people, in demand and very employable.
I know people like tha
One manager logged in 3 days into their holiday to send an email, I only realised when asked to approve a change found it had already been done!
Others have answer calls when on leave (I admit I didn’t know that) and one that I remember was I rang an account manager who answered a call from a beach somewhere (why they didn’t have a voicemail message saying call an alternative contact I don’t know)
Others I have similar behaviour from, me I am on holiday for two weeks and the laptop is put away in its bag and won’t be coming out until possibly a week on Sunday when I will do a quick check to see if I have to change the plans for that week or not (have a few people I need to see but the schedule wasn’t confirmed yet).
No checking the register does not count as work this week.
Unfortunately this sort of behaviour will only lead to more stress and probably time off work. I had the same when on call on “best endeavour” (there were only 2 of us and usually I was the one rung first as more likely to answer) and then formally 1 in 3 weeks on call.
I don’t do call out anymore when it was introduced I was asked if I wanted to and said no thanks…
Re: I know people like tha
I suspect most of those people who said they check dashboards out of hours or during "vacations" are leftpondian, because it's pretty much expected of them and as a result of that there probably won't be anyone else in the company to cover for them.
Missing the point
As with an awful lot of vendor information drivel they're conditioned to completely miss the point.
Businesses that rely on dashboards and reactions to them rather than preventing the problem in the 1st place (a novel idea I know, I doubt it will catch on) or even worse relying on "AI" to detect issues and (horror of horrors) to presumably "fix" them automatically are guilty of venality and crass stupidity of the highest order.
I'm glad I'm retiring in a few years, the corporate world is determined to make a bad situation ever worse.
If the systems are critical 24x7 99.9999% etc etc
Perhaps support and operations need to be 24x7 etc old school three shifts.
Costs a lot more but chump change compared your business critical (ie revenue producing) systems go pear shaped for a couple of weeks.
Who knows the extra hands might routinely test and verify backups.
With the rapidly widening gap between the cost of labour (relatively stagnant) and the exploding costs of running large systems (software and hardware) the cost /benefit of higher staffing might sink in but not before attrition has reduced the pool of skilled candidates to below critical levels.
Dashboards and other manglement fascinators sometimes referred to as a single pane of glass are mostly a singular pain in the arse.
Executive Fascination with
... systems which give them the illusion of control has always helped sell these things.
"Information at your fingertips / at the push of a button."
" (Look at me because) I've got my finger on the pulse of this company."
Does any of this sound familiar?
"while 34 percent would trust AI summaries more than a human technician"
I certainly wouldn't... never trust a bullshiat generator to NOT generate bullshiat !
(Unless it's the bullshiat generator itself that's gone down...)