What to keep and what to replace: a practical integration approach

The short answer: fixing a tangled tool stack isn't all-or-nothing. The practical approach is to keep the tools that genuinely work (accounting, email), integrate around them, and replace only the ones causing more trouble than they're worth. You end up with a connected stack without a risky, expensive rip-and-replace.

What's the question I should ask about each tool?

For each tool in your stack, ask two things: does it do its job well, and does it fit how you work? Those two questions sort every tool into a clear category:

• •Good, and fits how you work (accounting software, usually) → keep and integrate.
• •Good at a standard job it doesn't need to change (email, payroll) → keep.
• •Fights your specific operation and forces workarounds → candidate for replacement.
• •Redundant with another tool you already have → consolidate.

Run every tool through that and you get an honest keep / integrate / replace / consolidate list — which is far more useful than a vague sense that "we've got too many apps."

Which tools should I keep and integrate?

Your accounting package, email, payroll — standard functions done well by products you'd gain nothing from rebuilding. These are settled problems. Xero and MYOB handle Australian accounting; your payroll tool handles awards and superannuation; your email just works. Keep them, and connect your operational system to them so data flows without re-keying.

You get the benefit of proven tools plus a connected stack. That's the whole point of integration — you don't have to choose between "familiar but disconnected" and "connected but foreign." Connecting to your accounting software and your CRM is exactly this move.

Which tools are worth replacing?

The tools worth replacing are the ones causing the workarounds — the ones your operation is bent around, the ones held together with spreadsheets and manual steps. Usually these are the operational tools that don't fit your specific process: the booking system that doesn't handle how you actually book, the generic tool everyone's found a clever hack for.

That's where a system built to fit earns its place. Not because custom is better in the abstract, but because the operational core is the part of your business that's genuinely unique, and it's the part generic tools serve worst. The same logic decides whether to replace or integrate an ERP — replace only where the fit is genuinely wrong.

A worked example: sorting a real stack

Take a growing services business in Canberra with eight tools. Run them through the two questions — does it do its job well, and does it fit how you work — and the list sorts itself fast. Xero: does the job well, standard function, no reason to touch it — keep and integrate. Email and payroll: same, keep. The CRM the sales team knows: works fine, keep and integrate. A separate quoting tool and a scheduling tool that overlap and half-duplicate each other: consolidate. And the booking system everyone's built spreadsheet workarounds around because it can't handle how the business actually books: that's the replacement candidate.

Notice what that exercise produced. Out of eight tools, most stay, two collapse into one, and exactly one gets rebuilt to fit. That's a targeted, affordable change — not the terrifying "replace everything" project the owner had been dreading. Naming each tool honestly is what turns a vague sense of "our systems are a mess" into a short, specific action list.

How does keeping Xero and MYOB fit the Australian picture?

For Australian businesses, the accounting package is almost always in the "keep and integrate" column, and it's worth saying why plainly. Xero and MYOB aren't just familiar — they're built around Australian GST, BAS lodgement, Single Touch Payroll and superannuation rules. Rebuilding any of that from scratch would mean recreating a mountain of compliance logic that these products already handle, and that your accountant already relies on.

So the sensible move is never to replace the accounting tool. It's to keep it as the financial system of record and connect your operational system to it, so the operation feeds the books automatically and correctly coded. The same goes for payroll: keep the tool that already knows awards, super and STP, and integrate around it. Replacement effort belongs on the operational tools that fight your specific way of working — not on the settled, compliance-heavy functions the standard products already do well.

Why does this beat rip-and-replace?

Replacing everything at once is expensive, disruptive and risky. You're migrating all your data, retraining everyone, and betting the whole operation on a single big switch working first time. That's a lot of ways for things to go wrong, all at once.

Keeping what works and replacing only what doesn't is cheaper, safer, and faster to value. It respects the investment in the tools that are perfectly fine, and it focuses effort and money on the parts that actually hurt. You improve the stack in targeted moves rather than one all-or-nothing gamble — and you see benefits along the way rather than only at the end.

There's also a discipline to it: it forces you to be honest about which tools are genuinely a problem versus which ones you're just used to grumbling about. Often a tool that feels frustrating is actually fine, and the real issue is that it's disconnected — an integration fix, not a replacement.

The honest first step

Knowing what to keep and what to replace comes from seeing your whole stack honestly — which tools work, which fight you, which overlap. Mapping it tool by tool gives you a clear keep / integrate / replace list, and usually reveals that the fix is smaller and more targeted than the vague dread of "we need to sort out our systems" suggests. If the stack has grown past anyone's full understanding, start with the signs your tool stack has become a tangle.

Related reading

• •Do you need to replace your ERP, or integrate with it?
• •Integrating custom software with your CRM
• •Signs your tool stack has become a tangle

Tell us what you run, and we'll show you what's possible.