How do you choose new business software?

Start by writing down the problem in one paragraph, before looking at any product. Then shortlist against that paragraph, trial with your own data rather than the vendor's, talk to two businesses your size already using it, and settle two questions in writing: who owns your data, and how would you get it out.

Most bad software purchases go wrong at the beginning rather than the end. A demonstration is impressive, features get compared, and nobody has clearly stated what the business actually needs fixed, so the decision drifts toward whichever product performed best rather than whichever solves the problem. A written problem statement, agreed before the first demo, is the cheapest protection available: it turns a comparison of features into a comparison of outcomes.

Three checks are worth more than the rest of the evaluation combined. Trial with your own data and your own awkward cases, because every product looks good on clean demonstration records and your business is not clean. Talk to two comparable businesses already running it, ideally not the reference customers the vendor supplied, since a small market makes this genuinely easy in Adelaide. And check integration against the systems you already have, because software that cannot talk to your accounting or job-management platform means double entry, and double entry means two records that eventually disagree.

Then the two questions vendors find least comfortable, and the reaction to them is itself informative. Who owns the data you put in, and can you export it in a usable format? What happens if you leave, how much notice, what does it cost, and how long do you have to retrieve everything? A confident vendor answers both plainly. A vague answer is not a reason to walk away on its own, but it tells you what leaving will feel like, and that is worth knowing while you still have a choice. If you want a second opinion before signing, call 1800 456 567.