Blog / consulting
The case for keeping the technology you have
Why I often tell clients to keep the equipment they already own, and the questions I ask before recommending any new technology spend.
A surprising amount of my consulting work ends with me recommending that a client buy nothing.
That sounds like a strange business model, and maybe it is, but it follows directly from being independent. I resell no hardware, no licences, no managed services. When someone asks whether they should replace their network, their server, or their phone system, I have no reason to say yes beyond whether it is true. And quite often it is not.
Replacement is the default answer, not the right one
Every vendor you ask about aging equipment will recommend replacing it. This is not a conspiracy; it is just incentives. Vendors sell new things. An MSP quoting your network refresh earns nothing by telling you the switches are fine.
The result is that most businesses replace technology on the vendor's schedule rather than their own. A five-year-old switch is described as "end of life" as though it were a safety hazard, when what the term usually means is that the manufacturer would like to sell you another one.
I wrote a couple of weeks ago about what a second opinion on a quote catches. The single most common thing it catches is spending that did not need to happen at all.
When old gear is genuinely fine
Some equipment ages badly. Most of it ages slowly, and some of it barely ages at all.
A gigabit switch from eight years ago moves packets exactly as well as it did on day one. If your office needed gigabit then and needs gigabit now, the switch is not the problem, whatever the quote says. Good structured cabling lasts decades. Racks, UPS chassis, mounts, and cable plant almost never need to move when the electronics do. Servers running steady internal workloads can run far past their depreciation schedule if the workload has not grown.
Winter is when I see this most, because January budgets bring January quotes. A business gets a renewal-season proposal to "modernize," and half the line items are replacements for things that work.
The honest test for old equipment is not its age. It is three questions: Does it still do the job? Is it still getting security updates, or can it be isolated so that stops mattering? And can you still get parts or a replacement if it dies? Gear that passes all three can stay. Gear that fails the security question usually cannot, and I will say so just as plainly.
The questions I ask before recommending spend
When a client asks me whether to replace something, I work through a short list.
What is failing, specifically? "It's old" is not a failure. Slow file transfers, dropped calls, and outages are failures, and each one traces to a specific component, which is often not the one the vendor proposes replacing.
What does the current equipment cost to keep? Sometimes the answer genuinely favours replacement: rising maintenance, unsupported software, power-hungry hardware, one irreplaceable unit with no spare. That is a real case for spending, and I make it when it exists.
What breaks if we do nothing for two more years? Usually the honest answer is nothing, and "do nothing, revisit in two years" becomes the recommendation.
And finally: if we do spend, is this the right size of spend? Replacing like with like is often cheaper and better than the upsell. The same instinct that pads quotes, which I covered in my piece on cabling quote red flags, also inflates replacements.
What this is not
This is not an argument for running everything into the ground. Unsupported firewalls, operating systems past their update window, and single points of failure with no spare are real risks, and pretending otherwise is its own kind of malpractice. Backups, in particular, are one place I never tell anyone to coast.
The argument is narrower: replacement should be a decision, made from your requirements, not a reflex triggered by a vendor's calendar.
If you have a refresh proposal on your desk and you suspect half of it is optional, that suspicion is worth testing before you sign. An independent assessment of what you have and what actually needs to change is a fixed fee, and the most common outcome is a shorter shopping list.