How to choose job management software for an NZ trade business

Choosing job management software is a decision most trade businesses make once, badly, and then live with for years, because moving is painful enough that the second decision keeps getting deferred. This guide is the test sheet we would hand a mate: seven checks that separate software built for a working crew from software built for a head office, and the red flags that predict an expensive mistake. It applies whether you end up on ONYX1 or on something else.

Test one: does it run your actual loop?

Every trade business runs the same loop: capture the enquiry, quote it, schedule it, do the work, invoice it, get paid. The test is whether the software runs that loop end to end without you re-keying anything between stages. Ask the vendor to show a phone call becoming a job, the job becoming a quote, the quote being approved by a customer, the crew clocking onto the work, and the finished job becoming a GST-correct invoice. If any stage lives in a different module you pay extra for, or worse in a different product, you will be copying data between systems within a month.

Test two: what happens to the bill when you hire?

Per-seat pricing looks harmless at sign-up and taxes you forever after. At around NZ$52 per user a month, a five person crew pays roughly NZ$260 and an eight person crew is over NZ$400, and every apprentice, sub-contractor and part-time administrator moves the bill the day they start. Worse than the invoice is what it pushes owners into: shared logins and people left off the system, which is how records get holes in them. Ask the plain question: if I hire two people next year, what is my new bill? ONYX1's answer is a flat price per plan, NZ$79, NZ$149 or NZ$299 plus GST, and hiring within your plan changes nothing. Whatever tool you choose, get that answer in writing.

Test three: is it actually built for New Zealand?

Localisation is more than a currency symbol. GST at 15% computed by the software rather than typed by a person. NZBN on your documents. Xero-shaped accounting data, because your accountant lives there. Payroll export shaped for the systems NZ businesses run. The Privacy Act 2020 rather than a US privacy policy with the state names swapped out. If the demo shows sales tax fields, imperial units or a chart of accounts that has never met IRD, the localisation is a skin, and you will find the gaps at the worst time, usually at GST return time.

Test four: does it survive the field?

The people using this software most are not at desks. They are on ladders, under floors, in basements and up back-country driveways, holding a phone in a glove at arm's length in the sun. That means big targets, legible type, high contrast, and above all it means offline: writes that queue on the phone and sync when coverage returns. Ask the vendor what happens when a tech logs parts in a basement with no signal. If the answer involves 'they can do it later', the truth is the data will never be entered.

Test five: does safety gate the work, or trail it?

Paper pre-starts get filled in at smoko, after the work started, which defeats the point. The structural fix is software where the signed safety form gates the clock-in: no signed form, no timer. Look also at the incident register. If an incident can be edited or deleted after the fact, the register is not evidence, it is a diary. ONYX1's register has no delete path for any role, which is the standard worth demanding anywhere.

Test six: can you leave?

The honest test of a vendor is how easily they let you go. Your customers, jobs, timesheets and financial records should export in usable formats at any time, not as a favour at termination. Ask what exports exist, what format they take, and what happens to your data after you cancel. If the answer is vague, the lock-in is the product.

Test seven: if it has AI, where do the permissions live?

AI is arriving in every product in the category, and the question that separates a toy from a tool is structural: when the AI acts, whose permissions does it act under, and where are they enforced? If the connector runs as an administrator, then everyone using it has admin rights, whatever the interface pretends. The standard to demand: the AI inherits the role of the person who connected it, enforcement lives in the database, money moves need explicit confirmation, and every action lands in an audit trail. That is how ONYX1 wires its MCP connection, and it is the right wiring whoever you buy from.

The red flags

• A demo call required before you can see a price. Opaque pricing is a negotiation strategy, and you are the party without the data.
• A setup fee plus an implementation project measured in months, for a business of eight people.
• Per-seat pricing wearing a costume: 'from $X per month' where X buys one user.
• Modules everywhere: payroll module, safety module, asset module, each priced separately. The loop should be one product.
• No offline story, no export story, or an AI story that cannot answer the permissions question above.

The short version

Buy the loop, not the feature list. Price the whole crew, not the first seat. Demand NZ-native money handling, field-grade usability, safety that gates the work, data you can take with you, and AI that lives inside the permission system rather than above it. ONYX1 is our answer to those tests, and every claim it makes is on public pages you can check: onyx1.co.nz/pricing, onyx1.co.nz/security and onyx1.co.nz/ai. Hold every vendor, including us, to the same sheet.