Start With the Problem, Not the Product

Most software decisions in a small business begin in the wrong place. Someone sees a slick demo, a competitor mentions a tool, or a member of staff gets frustrated with a spreadsheet, and suddenly you are comparing feature lists for products you did not know you needed ten minutes earlier. The result is usually a subscription you half-use and a team that quietly keeps doing things the old way.

Try reversing the order. Before you look at anything, write down the actual task that is causing pain. Be specific. "Invoicing is a mess" is too vague to buy for. "It takes three hours every Friday to raise and send about forty invoices, and two of them went out late last month" gives you something to measure against.

Then ask a few blunt questions:

• How often does this task happen — daily, weekly, or once a quarter?
• How long does it take now, and who does it?
• What goes wrong when it is rushed or forgotten?
• Would a better process, rather than a new tool, fix most of it?

If the answer to the last question is yes, you have saved yourself a subscription. If it is no, you now have a benchmark. A tool that cuts that Friday session to forty minutes has paid for itself. A tool that saves ten minutes a month probably has not.

Trial It With Real Work, Not a Demo

Free trials are generous, but they are also short. A fortnight disappears quickly if nobody is actually using the thing. Pick two or three people who do the task every day, give them real jobs to run through the system, and set a fixed review date in the diary before the trial ends. Trials that drift past their end date usually convert to paid subscriptions by default.

During the trial, keep a simple note of every workaround. Any moment where someone says "I'll just do this bit in the old spreadsheet" is a signal worth recording. One or two workarounds are normal. Five or six on a routine task means the software does not fit how you work, and no amount of training will fix a poor fit.

Also test the boring edges. What happens when a customer's details change halfway through a job? Can you correct a mistake without deleting and starting again? Can a new starter use it after twenty minutes with no hand-holding? These small tests reveal more than any sales presentation.

Check How It Talks to Everything Else

Integration is where SMEs lose the most time, and it is the question least often asked properly. A tool that does not connect to the systems you already rely on creates a second version of the truth, and someone has to reconcile the two by hand.

Ask directly, and get answers in writing:

• Which other systems does it connect to natively, and which need a paid add-on or a developer?
• Can you export your data in a common format such as CSV, and how often?
• Does it handle the tax and payroll reporting formats you are required to submit?
• Where is the data stored, and who can access it?

If a supplier is vague about exports, treat that as a warning. Being able to take your customer list, your order history or your accounts elsewhere when you choose is not a nice-to-have. It is the difference between a supplier and a hostage situation.

Price the Training, Not Just the Subscription

The monthly fee is the easy number to compare and the least important one. The real cost is the hours your team spends learning the system, plus the dip in output while they do. For a five-person business, twenty hours of collective learning time is a genuine expense, and it is worth writing down before you decide.

Ask what support is included. Is there a phone line, or only email? Are there sessions for new starters, or do you pay each time? If your main user leaves, how quickly can someone else pick it up? Businesses that lose a key member of staff and find the software locked behind one person's knowledge learn this lesson the hard way.

Where a tool genuinely saves time, training is an investment. Where it is merely newer, training is a cost you absorb for very little return.

Plan the Rollout and the Exit Before You Sign

Roll out in stages rather than switching everything on a Monday morning. Run the new system alongside the old one for a month on a limited part of the work, and only retire the old method once you are confident. Keep your data tidy during this period — duplicate records will haunt you later.

Before signing, read the notice period and check what happens to your data if you leave. Some contracts auto-renew annually with a narrow cancellation window, which is fine until you want to change your mind. Diarise the renewal date the day you sign, not three weeks before it.

Finally, name one person who owns the system. Not a committee, not "whoever is free" — one individual who checks it is being used, gathers feedback and raises issues with the supplier. Software does not improve a business on its own. Someone has to make it earn its keep.