Timing does more of the work than argument does.
Hardware conversations go furthest when infrastructure is already on someone else’s mind. Five moments reliably create that opening: after an incident, during a business change, when an external party asks a question, when costs shift, and when vendor support ends. There are also moments worth letting pass, because raising it at the wrong time can set the conversation back.
Infrastructure is easy to discuss in the week after something breaks and surprisingly hard to raise at any other time. That isn’t a communication problem. It’s that infrastructure stays invisible to everyone outside IT until it stops working.
Which means the same information can land very differently depending on when it arrives. A conversation that goes nowhere in March can land in June, unchanged, because something shifted in between.
These are five moments worth watching for, and four worth waiting out.
1. Immediately after something goes wrong
A failure, an outage, or a near miss is the only time the cost of ageing hardware is visible to people outside IT. For a short window, the abstraction disappears and the business has just experienced the thing you’ve been describing.
Most teams miss this window, and understandably so. You’re occupied fixing the problem, and raising budget in the middle of an incident feels opportunistic.
The move isn’t to raise it during. It’s to raise it in the week after, when the disruption is still fresh but the pressure has lifted. Not as an argument, but as a short factual note: what failed, what it cost in downtime and staff time, how many other devices are in a comparable position.
That last part is what turns a resolved incident into a pattern. One failure is bad luck. Eleven more machines of the same age and condition is a forecast.
Worth linking this to your broader risk picture too, since ageing infrastructure and ransomware exposure tend to travel together.
2. When the business is changing shape
Growth, restructuring, an office move, a new ERP or line-of-business system, a shift in how people work. Any of these puts infrastructure on the agenda without IT having to put it there.
The reason this works is that it changes the category of the spend. Infrastructure investment attached to a business initiative is assessed as part of that initiative’s cost. Infrastructure investment on its own is assessed as overhead. Same money, very different conversation.
Practically, this means keeping a current view of what the fleet could support, so that when someone asks whether the business can take on a new system or add thirty people, you can answer immediately rather than going away to check. The answer “yes, and here’s what it needs” carries more weight than a separate request three months later.
3. When someone outside IT raises it
An external party asking about your infrastructure carries a weight that an internal request doesn’t. It’s no longer IT’s opinion.
The most common versions are client security questionnaires, supplier onboarding and procurement reviews, particularly if you work with enterprise or government clients. Audit and compliance reviews do the same job. So do cyber insurance renewals, where applications have become considerably more detailed than they once were.
Patching operating systems is one of the eight controls in the ASD Essential Eight, which means hardware too old to run a supported operating system becomes a compliance question rather than a preference. We’ve covered what Essential Eight compliance involves in more detail, and separately what the transition to the ASD Essential Series means for organisations working toward it.
On insurance, one specific action beats an assumption. If your renewal is coming up, ask your broker directly how your policy wording treats unsupported operating systems and whether extended support arrangements are treated differently from fully supported ones. The answer varies between insurers, and it’s better to know before a claim than after.
When one of these lands, the case writes itself, because someone else has already established that it matters.
4. When the numbers move
A settled decision can be reopened when the inputs change. That isn’t reopening a debate, it’s reporting new information, and it’s a materially different conversation to have with a finance team.
Right now there’s a live version of this. Gartner expects average PC prices to be 17% higher than 2025 levels by the end of 2026, driven by memory manufacturers redirecting capacity toward AI infrastructure. If your allocation was approved in May or June, it very likely buys fewer devices than intended, and that’s not a planning failure by anyone.
The same logic applies to any input change: a vendor price rise, a lead time extension, a support contract ending, a licensing change. The trigger is always the same. Something outside your control moved, and the previous decision was made without it.
5. When vendor support ends
An end of support date is the most useful trigger available, because it’s external, published and not negotiable. It isn’t IT’s judgement about when equipment is too old. It’s the vendor stating when they stop fixing it.
Windows 10 reached end of support on 14 October 2025. Extended security updates are available for business devices as a paid program for a limited period, but they cover security patches only, with no new features and no technical support, and eligibility depends on the version and edition you’re running.
The reason this opens a conversation well is that it reframes the question. It stops being “should we replace these devices” and becomes “what’s our plan for the devices the vendor no longer supports.” The second question is much harder for a business to leave unanswered, and it invites a plan rather than a purchase.
One thing worth clarifying early: recent coverage about extended Windows 10 updates has largely concerned the consumer program, which has different terms from the commercial one. If someone in your leadership team has seen a headline suggesting the deadline moved, that’s worth correcting before it shapes a decision.
And four moments to let pass
Not every opening is a good one, and a conversation raised badly can make the next one harder.
- During the incident itself. Everyone is focused on restoring service, and budget raised mid-outage reads as opportunism even when it isn’t. Wait until the following week.
- When age is the only argument you have. “These are five years old” invites the obvious reply that they still work. Without failure rates, support hours, or a specific capability the fleet can’t deliver, the conversation tends to end in a deferral that’s harder to reopen later.
- When you can’t answer the follow-up. The first question after any refresh proposal is what it costs and whether it can be staged. Raising the topic without those answers usually produces a request to come back with them, which turns a live conversation into a delayed one.
- When leadership attention is somewhere else. An acquisition, a restructure, a major client issue, an audit. Infrastructure won’t win attention against any of these, and a proposal deferred for lack of bandwidth still counts as deferred.
The common thread is that a premature conversation spends the opportunity without gaining anything. The moments above keep coming around. A rejection tends to stay on the record.
What the five have in common
None of them are about presentation. Every one attaches the conversation to something the business is already paying attention to, whether that’s an incident it just experienced, a change it’s already funding, a question it’s been asked, a number that moved, or a date it didn’t set.
The practical implication is that the work happens before the moment arrives. When a server fails on a Tuesday, the useful thing isn’t building a case that week. It’s already knowing the age profile of the fleet, what’s in warranty, what’s out of support, and roughly what replacement costs. Then the moment becomes an opportunity rather than a scramble.
That preparation is also what makes the four waiting moments avoidable, since most of them come down to being asked something you don’t yet have the answer to.
It’s the part most in-house IT teams don’t have room for, not because it’s difficult, but because it competes with keeping everything running. Our infrastructure services and earlier piece on infrastructure lead times both cover the planning side in more detail.
How to prepare ahead of time
The five moments arrive without notice. What makes them usable is having the numbers ready before one does.
That’s what our free hardware refresh business case produces: a scoping session with a written assessment of your environment, the total cost of ownership, the risks in your current setup, and a phased plan. Something you can put in front of your executive team the week an opportunity opens, rather than starting from scratch. No obligation attached.
If you’d rather run your own numbers first, the cost of delay calculator gives you an indicative figure in a couple of minutes.
FAQs
When's the best time to raise a hardware refresh?
When infrastructure is already on the business's mind: the week after an incident, during planning for growth or a system rollout, or when a client, auditor or insurer has asked a question about your environment.
How do you start a conversation about replacing IT hardware?
Attach it to something already happening rather than raising it on its own. A recent outage, a planned business change, an external requirement, a shift in pricing, or a vendor end of support date all give the conversation a starting point that didn't come from IT.
Â
When should you avoid raising it?
During an active incident, when equipment age is the only argument available, when you can't yet answer questions about cost and phasing, or when leadership attention is consumed by something larger.
What should you have ready before the conversation?
The age profile of the fleet, what's in and out of warranty, what's approaching or past end of support, and a rough replacement cost. The moments arrive without warning, so the preparation matters more than the pitch.
Does Windows 10 end of support justify replacing devices?
It justifies having a documented plan. Options include upgrading eligible devices to Windows 11, enrolling in paid extended security updates for a defined period, or replacing hardware that can't be upgraded.
