Blog
The 7 IT reflexes that will no longer be acceptable in 2026 (but that you still have)
IT is perhaps the field with the most stubborn habits.
We do things “the way we've always done them.” Because it worked once.
Because no one has time to take another critical look at it.
At the same time, we see the opposite: as soon as a new hype comes along, we all rush headlong after it. AI, SASE, zero trust, XDR, you name it. Often hopeful of finding something really good. And sometimes... afraid of missing the boat. Afraid of looking “old-fashioned.”
At Sciante, we do things differently these days.
Every year, we evaluate our suppliers. Not because they are ‘bad’, but to check whether what they deliver no longer brings us value. Those who no longer contribute to results are likely to be dropped. This is how we keep our company simple, effective, and affordable.
We do exactly the same with our own IT. Anything that no longer serves a purpose is removed. This not only saves costs, but also reduces your attack surface. Less stuff, less risk, less hassle. And we are more IT-heavy than many people think; our own server park runs like clockwork. Since the rollout of the current dual DC setup in 2017, we have had zero seconds of downtime.
And then there are the habits and reflexes. The ways of working that are never on an invoice, but do cost money every month. That's often where the biggest leaks are.
In this blog, I will take you through seven IT reflexes that you really need to stop using in 2026. Not because it is trendy, but because they cost your organization money, add risks, and/or frustrate your people unnecessarily.
Do you recognize them? Then you know right away where you need to start in 2026.
Reflex 1 — Thinking you are free because your supplier “takes care of everything”
Vendor lock-in is one of the biggest business risks today. Not because your supplier is “bad,” but because you have no way out when they change the rules of the game.
Consider Broadcom's acquisition of VMware more than two years ago. The financial consequences are now painfully clear: for some customers, license costs have increased by a factor of 15. That's right: not 15%, but a 1400% price increase. And Broadcom is getting away with it because many organizations have nowhere else to go without enormous migration pain.
Microsoft is moving in the same direction. From July 1, the 365 packages will increase significantly in price: between 5% and 33%. The official reason: 1,100 new features have been added. The reality: with “renewed endpoint management” and more active “restriction of shadow IT,” you are being pulled even more firmly into the Microsoft ecosystem. Using alternatives is becoming increasingly difficult.
Add to that the creeping integration of Copilot, and you arrive at an uncomfortable conclusion: more and more of the data your company relies on is coming under their control instead of yours. That's not an IT detail; it's a strategic risk.
Good resolution for 2026: request the vendor lock-in checklist from Sciante.
Reflex 2 — Thinking that one cloud provides an “overview” (while you build one kill switch)
Convenient, right? Everything in the cloud. No more hassle with disks, no more servers to worry about.
Until someone else decides you're no longer allowed in.
From Google flagging a medical photo as child pornography to Microsoft blocking an account without explanation, you don't even have to do anything wrong to lose access. And if you use Windows with BitLocker, even your own laptop can become unusable if your Microsoft account is locked. Your key is there. No account = no key = no data.
For companies, this is an existential threat. If all your core data is stored in a single cloud environment, you have effectively placed an external kill switch on your business operations. If you lose access, you might as well close up shop.
Therefore:
-
Always ensure you have an up-to-date backup of essential data outside that single cloud environment.
-
Keep encryption keys under your own independent control.
-
And if local storage is really not possible, make sure you have a second cloud location, preferably in another country and with a different provider.
The cloud is useful. But complete dependence is not a strategy, it's a gamble.
So if you have one cloud, put setting up that second location on your to-do list for 2026.
Reflex 3 - Hype first, thinking later
Everyone wants “something with AI.” Suppliers promise miracles, management wants “something like what was in that presentation,” and before you know it, another pilot project is up and running.
Unfortunately, 95% of these AI and hype projects fail. Not because the technology is worthless, but because there is no real problem underlying it. A problem is being sought for a solution.
Don't have a concrete problem? Then you don't need a solution. And if you start inventing a problem because you want the new shiny tool so badly, you're not executing a strategy, you're burning money.
At home, in my new house, I play with home automation. Another hype. But there, it's a hobby: fun, room to experiment, in my own time and with my own money.
In a commercial company, the same rules apply: hypes are only acceptable if they demonstrably contribute to turnover, margin, risk, or productivity. Anything less than that is a toy, often with a price tag that is far too high.
Reflex 4 — Thinking you can outsource governance (and still maintain control)
You don't leave the steering wheel to the accountant, even though they prepare the annual figures.
Yes, you hire expertise, but you remain at the wheel.
You know how the company is doing, you make the decisions.
The same applies to your core business. You can easily outsource production, logistics, or customer service. But no one hands over their brand promise with the message: “You tell me, surprise me.”
But then there's IT.
That's exactly what happens there.
As soon as things get complex, many organizations pass everything on to one large supplier. Cloud, infrastructure, security, monitoring, sometimes even architecture: “Just set it up and make sure it works.”
But IT is your business. Without IT, there is no turnover, no process, no customers.
If you don't set the boundaries—which risks are acceptable, what performance is required, what is and isn't possible—then your supplier will determine the playing field.
The result: rising costs, declining agility, and IT that mainly benefits your supplier.
You can't buy governace.
You have to take control.
On the agenda for 2026: regaining governance over all IT.
Reflex 5 — Immediately blaming sluggishness on “insufficient resources”
“Then we need an extra server.”
“Install faster disks.”
That's still the standard response when an application is slow. But if you're stuck in traffic, you can't solve that problem with a bigger engine or wider tires. The problem isn't your car, it's the road.
Of course, if you go from 100 to 200 users, you need more capacity. But I've never seen an application scale neatly from 100 to 1,000 users just by adding a bigger server. At some point, you reach the limit.
Scale matters. On a different scale, you have to work differently. This also applies – and perhaps especially so – in IT.
An application that “suddenly” no longer performs rarely has a real resource problem. There is almost always something else going on: inefficient queries, chatty services, clumsy integrations, telemetry that crashes everything.
Extra resources can provide temporary relief. But it's an expensive band-aid that never closes the real wound. The underlying problem will return. Always.
Only when you expose and solve that problem will you achieve real performance – without having to upgrade your infrastructure every time.
On the agenda for 2026: figure it out... not write it down immediately, but something along the lines of making an appointment with you.
Reflex 6 — Being surprised by disruptions that were clearly foreseeable
“If it ain't broken, don't fix it” is sound advice.
But in IT, the real question is: when do you call it “broken”?
Your car gets an annual inspection.
Not because it's broken down on the side of the road, but to prevent that from happening.
Does the garage see too much wear and tear? Then you replace that part before you hit the road. Safety first and costs under control, please.
With applications, that almost never happens. I know very few companies where software is actually checked periodically. Let alone anything resembling an MOT.
“But Hugo, we have monitoring.”
Yes. Except that by the time that monitoring kicks in, you're often already incurring damage. Productivity is down, customers are angry, IT is stressed. Honestly, you can usually see it coming – the line of frustrated users at the IT department. That line is the advance warning of your monitoring alert.
Fact: software wears out. Not because the code spontaneously decays, but because the world around it is constantly changing: new Windows versions, database upgrades, growing datasets, additional integrations, more users, more logging, more everything.
Those changes pile up. Until “suddenly” something goes wrong.
But “suddenly” is usually the result of years of no maintenance.
In 95% of cases, you can see IT problems coming. And prevent them.
Read that again: you can stay ahead of 19 out of 20 issues.
You can stop putting out fires and start scheduled maintenance on your IT.
New years resolution intention: proactively monitor and act before an incident occurs.
Reflex 7 — Automatically dismissing open source because it would be “too much hassle”
In many boardrooms, there is still a kind of reflex: open source is scary, unreliable, “for hobbyists.”
But take a look in your own pocket. Huh?
Do you have an iPhone? iOS is built on FreeBSD.
Do you have an Android phone? Android runs on Linux.
Both are open source. Both have proven to be stable, secure, and extremely reliable.
You trust your phone.
But in your IT landscape, you often ignore open source. That's a real shame.
At Sciante, we run our production environments on Linux servers. Not because it's a “fun” hobby project, but because the same application runs 2-3 times faster on the same hardware than on an average closed-source alternative. That's free money: less hardware, fewer cloud resources, lower licensing costs. It works exactly the same way in the cloud.
There's something else to consider: open source rarely comes with mandatory spyware telemetry. When you turn it off, it's really off—not “only the mandatory basic set of data continues to flow to the supplier for a while.”
In dozens of projects, I have never seen a situation where we couldn't find a good open source alternative. Not always the exact same product, but the same or better functionality, more control, and often higher performance.
Does this mean you should dump all your closed-source software tomorrow? Of course not.
But if you only look at “the usual suspects,” you're buying convenience – and giving up flexibility, margin, and control.
Avoiding open source is not risk management.
It is missing an opportunity.
Break with costly IT reflexes – start here.
Turn your IT into an advantage again – not a cost drain
Many organizations unwittingly end up making knee-jerk decisions, creating dependencies, and investing in expensive hype.
Not because IT teams are doing anything wrong, but because the playing field has become so complex that old reflexes are becoming increasingly ineffective.
Want to get rid of knee-jerk decisions, vendor lock-ins, and money-burning hype?
And do you want your IT to do what it's supposed to do again: deliver more results, less waste, and more peace of mind in your organization?
Then make a no-obligation appointment with me.
In 30 minutes, we'll look at your IT landscape together and:
-
we'll identify 1–2 stubborn reflexes that are currently costing you money
-
you'll see where you can immediately do more with less
-
you will know which button to press first for the greatest effect
No sales pitches, no thick reports.
Just an honest conversation, sharp questions, and concrete options to make your IT work to your advantage again.
👉 Do you want your IT to become a strategic advantage again?