Switching accounting software is a migration, not a subscription swap

The conversation nearly always starts with a renewal email and ends with a comparison table. The price has gone up again, somebody has found a rival product that costs less, and the table shows the two side by side with ticks in most of the same boxes.

The table is not wrong. It is answering a question about subscriptions, when what is actually being proposed is a data migration.

Write the price difference down as an annual figure first

Do that before anything else, because it is the number the whole decision gets measured against and hardly anyone says it out loud. A few pounds a month is a few tens of pounds a year. That is the prize.

Now price the move. Somebody has to map the chart of accounts, agree a conversion date, get the balances in, reconcile them against the old system, reconnect everything that was plugged into the old ledger, and answer questions for a month afterwards. Whether that is your bookkeeper's evenings or an invoice from a firm like ours, it is not tens of pounds.

For a good many businesses those two numbers are a decade apart and the decision is over before it starts. The interesting cases are the ones where money was never really the point, and we will come to those.

What has to move, roughly in the order it bites

A ledger is not a document. It is a set of balances, a history behind them, and a web of connections to other systems, and those three travel very differently.

The balances. You choose a conversion date and the trial balance as at that date has to arrive intact. This is the part every supplier's import tool is built for, and the part worth worrying about least.

The history. Most migrations carry balances and summaries rather than years of transaction detail. That is fine until somebody wants to know what a supplier charged two years ago, or until the comparative column in the accounts looks thinner than it used to. Decide deliberately how far back the detail is coming, rather than finding out in March.

The attachments. Receipts and supplier invoices attached to individual transactions are the thing we see left behind most often, because they sit alongside the ledger rather than inside it. If your record of a purchase is the entry plus the scan, then half of every record is at risk.

The bank. Feeds do not transfer. Each one is set up again at the new end, and there is usually a gap in the middle where statement lines have to be brought in by file. Items still unreconciled in the old system on the conversion date need an owner and a plan, because those are what make the first month refuse to balance.

Payroll. Moving mid-year means carrying year-to-date figures for every employee and the submission history behind them. Anyone who has done it once will tell you to move at the start of a tax year instead. They are right.

Filed returns. Whatever you submitted to HMRC through the old software, the new software did not submit. Know which periods are closed, and where the authorisation to file sits, before switch-over rather than after.

Everything that was plugged in. This is the one that catches people. The card payment feed, the till, the job or project system, the stock app, the expense app everyone photographs receipts into, the collection mandate. Each is a separate reconnection with its own history that does not follow, and at least one will be something nobody remembers arranging.

Then the part that is not data at all

People get slower for a while. A bookkeeper who could code a purchase invoice without looking now has to look. Reports live somewhere else and are called something else. The first month goes on checking figures that would previously have been trusted.

That cost is real and temporary, which makes it easy to wave away. It should be budgeted instead, because it is the reason a switch made in a busy month is remembered badly however cleanly the data landed.

When the sum still says go

None of this is an argument for staying put. It is an argument for switching for a reason that survives contact with the invoice, and price rarely is one.

The reasons that hold up are about fit. The ledger cannot do something the business now does: stock, projects, job costing, a second currency, a second company. Or three tools are covering overlapping ground and one of them could absorb the others. Or your accountant cannot work in it properly, which quietly costs more in their time than the subscription ever saves. Or the integration you need does not exist and is not going to.

Those are structural, and a structural problem does not improve by being carried another year. The test for that is the one we apply to any system: does it still describe how the business actually works?

A price rise on its own is not structural. It is an irritation with a number attached, and the number is small.

70%: Software spend cut for one client by consolidating tools rather than switching them
Uninterrupted: Filing, through a full practice migration from Digita to IRIS

The order matters more than the tool

We moved an entire accountancy practice from Digita to IRIS, which is this job at a scale most businesses will never have to attempt. What made it work was not a piece of software. It was the sequence.

01Map every account code and field to its destination before touching the new system
02Pick a conversion date at a period end, and a quiet one
03Load, then reconcile against the old system until the two agree
04Keep the old system readable, not merely backed up, for a year
05Retire each old integration only once its replacement has run a full cycle

The fourth is the cheapest insurance on the list and the one skipped most often. Read access to the old ledger for another twelve months usually costs very little, and it is the difference between answering an awkward question in ten minutes and reconstructing the answer from paper.

If you are sitting with a renewal email and a comparison table, the useful hour is not the one spent on the table. It is spent writing down the annual price difference, then listing everything that logs into your current ledger. Put those two beside each other honestly and the answer tends to present itself, in one direction or the other.

If it does turn out to be go, the migration is the project and the sign-up is a formality. If you want a second opinion first, the first conversation costs nothing, and "stay where you are and spend the money on something more useful" is an answer we give more often than not.

We work with businesses across Knutsford, Alderley Edge, Wilmslow, Altrincham, Stockport and Warrington, and remotely for clients anywhere in the UK.