How to assess unreliable IT before you replace anything

When an owner tells me the systems keep breaking, the next thing I usually hear is a plan to replace something. It might be a new IT provider, a new phone system, new laptops for everyone, or a new ERP. Sometimes a replacement is the right call. More often the new system inherits the old problems, because nobody found out why the old one kept failing.

This article walks through what to check before you spend money on a replacement and how to decide what to fix first. You don't need special tools for any of it.

Who this is for

• Owners and GMs dealing with repeat outages, slow fixes, or vendors blaming each other
• The person inside the company who became the IT contact because somebody had to
• Finance leaders trying to understand why technology costs keep climbing
• Leaders about to approve a replacement project who want to know it will solve the real problem

Symptoms and causes

Most of what owners describe are symptoms. An outage comes back every few weeks, tickets stay open for days, software works for some people and not others, or two vendors each say the problem is on the other side.

The causes underneath are usually fewer and harder to see. The common ones are:

• No single person owns the system, so decisions and fixes wait on whoever has time.
• The setup was never documented, so every fix starts with someone rediscovering how it works.
• A key system is past its supported life and no longer gets security updates.
• The business has changed and the core software no longer fits how the work gets done.
• The support agreement covers less than everyone assumes.

A replacement can help with an unsupported system or one the business has outgrown. It does very little for the other causes, and those tend to follow the business onto whatever it buys next.

Start with an inventory

Before you replace anything, write down what you have in a simple spreadsheet. For each important system, capture:

• What it does and which part of the business depends on it
• Who inside the company owns it and makes decisions about it
• Who supports it, and what the support agreement actually covers
• When the contract ends or renews
• Who holds admin access

Pay close attention to admin access for email, your domain name, the finance system, and the core business system. It isn't unusual to find those credentials sitting with a former employee or with a vendor the company stopped using years ago. Finding that out during an inventory is a lot better than finding it out during an outage.

Rank by what stops the business

Some unreliable systems deserve much more attention than others. Rank them by what happens to the business when they go down. Put the systems that stop revenue or cash at the top, such as order entry, invoicing, payroll, and anything customers use directly. A slow shared drive is annoying, but an invoicing system that fails at month end costs money.

Then check the backups for those top systems by actually restoring something. A report that says the backup succeeded tells you a job ran. It doesn't tell you whether you can get the data back or how long it would take. Pick a file or a database from last week, restore it somewhere safe, and time it. If nobody can do that, you have found your first priority.

The ownership question

Most IT checklists skip this part, and it is often why the same problems keep coming back. For each critical system, someone inside the business should be able to say who decides how the system is used and changed, and who is accountable when it fails. If the answer is "the vendor" or "I guess me," that gap is worth fixing before any new technology.

Then look at your outside support. Plenty of IT providers do good work inside the scope they were hired for. Trouble starts when that scope and the business's expectations drift apart. A provider hired to keep the laptops and network running slowly gets expected to manage the ERP vendor and plan the next upgrade. Nobody formally agreed to that, so it doesn't get done well.

Read the agreement and compare it with what you need today. The goal is a written line between what the provider owns and what the business owns, somewhere both sides can see it. Sometimes the answer is a scope change with the same provider. Sometimes the business needs a different kind of help.

Decide what to fix now and what to plan

Once you have the inventory, the ranking, and the ownership gaps, sort every issue.

Fix now covers anything that puts the business at risk today. Admin access held outside the company, backups that don't restore, systems that no longer get security updates, and active accounts for people who have left all belong here.

Plan covers issues that need a decision or a budget, such as replacing an unsupported system or changing a support agreement.

Everything else goes on a watch list with a date to look at it again.

Give every item one owner and a way to confirm it is done. An item like "look into backups" tends to sit open for months. "Restore last week's invoicing database and record how long it took" can be checked off.

For most small businesses, a workable pace is to clear the fix-now list in the first 30 days. Make the planning decisions over the following 60, and hold off on large replacement projects until those decisions are made.

The same thinking applies to automation. If a process or system isn't reliable yet, adding AI to it mostly makes the failures show up faster, which I wrote about in AI makes bad process faster.

When an outside review makes sense

You can do most of this yourself. An outside review helps when the people closest to the systems are also the ones being evaluated, or when a large replacement decision is coming and you want an independent view before you commit.

A scoped review should hand back something you can act on. At minimum, that means:

• Written findings on the current state
• Realistic options and the tradeoffs between them
• A prioritized roadmap with an owner for each step
• The open questions that still need a decision

Be cautious if an assessment ends with a proposal for ongoing services and little else.

Get a clear picture first

BaronTek works with owners and leaders of small and mid-sized businesses on IT planning. That work is a defined review of current systems, contracts, and dependencies, and it ends with the decisions, open questions, and next steps documented for the people who will carry them out.

If the same problems keep coming back and you are weighing a replacement, a short note on what keeps breaking and what you are considering is enough to start a conversation.