You will receive a documented estimate: an indicative range, named assumptions, costed work items and a delivery timeline. Not a figure quoted over the telephone and quietly revised three weeks later.
We quote in ranges rather than single figures because an early estimate rests on incomplete information. A precise number implies a certainty nobody has before the data model has been opened and the interfaces exercised, and once spoken it tends to harden into a budget that gets defended rather than a hypothesis that gets tested. A range with its assumptions written beside it behaves differently: when an assumption changes, so does the number, and both parties can see why.
A fixed price becomes available once several things are true at once. Scope has to be described in enough detail to be estimated rather than imagined; data due for migration has been inspected rather than assumed clean; third-party systems involved have known behaviour and known costs; acceptance criteria are agreed; and one person on your side is able to make decisions. Where any of those is missing, a fixed price is time-and-materials with the estimating risk handed to whichever party guessed worst. We would rather have that argument in week two than at the end of the build.
Unknowns are handled in two visible ways. Contingency is priced as its own line, so you can see how much of the estimate is allowance rather than known work, instead of it being buried inside an hourly rate. Beyond that, any change is assessed, costed and approved in writing by both sides before it enters the schedule — slower than a verbal agreement and considerably cheaper than an unrecorded one.
Our figures exclude GST, which we apply as required by law. They also exclude third-party licences and subscriptions held in your name, cloud consumption billed to you directly by the provider, data cleansing carried out by your own teams, travel beyond visits agreed in advance, and specialist services such as independent penetration testing that we coordinate rather than perform. Exclusions are itemised rather than implied, because a surprise about what is excluded damages a working relationship more than the amount itself.
Discovery and architecture work is chargeable, and that arrangement protects the client at least as much as it protects us. Paying for discovery buys time from engineers who would otherwise be estimating from a two-page brief, inflating the build price to cover what they cannot see, or declining the work outright. It also produces artefacts — current-state process map, target architecture, integration inventory, risk register — that belong to you whether or not the build proceeds with us. A client who takes that document elsewhere is not a problem; a client who discovers in month four that the work was mispriced is.
Each step has a stated turnaround, and each asks something specific of your side.
An engineer reads the request and replies confirming receipt, either accepting it or explaining plainly why we are not the right practice. Where an answer is ambiguous we ask two or three precise questions rather than sending a questionnaire. All we need from you here is this form completed honestly, including when you have marked the budget as undetermined.
One video session with the engineers who would do the work and whoever in your organisation understands the process best. We cover how work happens today, which system holds each piece of data, what currently fails, and what would need to be true for this to be worth funding. A short written summary follows so any misunderstanding surfaces immediately.
The estimate carries an indicative range, a breakdown of the major work items, every assumption the range depends on, identified risks with their likely cost, hosting costs, and a milestone timeline. Where we could not establish something — an API's rate limits, the state of a legacy database — we name it as an open item rather than averaging it away.
You review the document internally and either proceed, request amendments, or decline with no obligation. Construction typically starts three to six weeks after acceptance, which covers contracting, environment provisioning and your own availability. Larger builds occasionally begin with a paid discovery phase, credited against the build if you continue with us.
These bands exist so you can tell early whether a conversation is worth having. They are drawn from delivered mid-market work in Western Australia, are indicative only, and frequently move once scope is written down. Every figure is exclusive of GST and of third-party costs, and none is confirmed until it appears in a written scope accepted by both parties.
Two cautions on reading the bands above. Systems at the lower end usually have few integrations, clean source data and a single user role; those at the upper end tend to involve regulated data, offline field use, or several systems that must stay in step. Your position can only be established from a written scope, and we would rather revise a range upward in an estimate than midway through a build.
Two projects described identically can differ by a factor of four. These are usually the reasons.
Each connection has its own authentication, rate limits, failure modes and documentation quality. A modern REST endpoint may take a day; an on-premise system reached only through a nightly file drop takes considerably longer to make reliable.
Migration cost depends less on volume than on condition. Duplicated records, inconsistent codes, free-text fields holding structured data and twenty years of historical rows all require reconciliation and a documented rollback if any of it does not add up.
Privacy, work health and safety, financial reporting or industry regulator requirements add evidence trails, audit logging, retention rules and role-based access. Those are design constraints rather than features bolted on, and they affect how every screen is built.
Working without reliable connectivity means local storage, conflict resolution, queued synchronisation and deliberate handling of partial uploads. Sites with poor coverage also demand interface work suited to gloves, glare and intermittent taps.
Every role carries its own permissions, screens, notifications and approval paths, all of which need specification, building and testing. Four roles is not four times one role, but it is meaningfully more than one.
Reporting is routinely underestimated. Each report needs its own definition of truth, edge-case handling and performance work, and anything feeding a regulator requires reconciliation against source systems before it can be signed off.
Allow about ten minutes. Anything you cannot answer yet can be left open and settled on the call.
Fields marked with an asterisk are needed for us to reply with anything useful.
Yes. Discovery occupies engineers who would otherwise be building, and charging for it means the estimate reflects what was found rather than what was hoped. You receive documents that remain yours whether or not the build proceeds with us.
We can give an indicative band from a conversation, which is useful for deciding whether to proceed. A figure intended as a commitment follows inspection of the data model, interfaces and volumes, because that is where budgets are usually wrong.
Yes, delivered remotely on the same terms, with any interstate travel quoted separately and agreed before it happens. Where a project genuinely needs people together, such as discovery on a complex site or a go-live, we plan those visits into the schedule rather than leaving them to chance.
Often. We begin with a short paid review establishing what is salvageable, what is fragile and where the documentation stops. Occasionally the honest recommendation is to rebuild one module rather than continue extending it, and we will quote both paths.
Change is normal and priced openly. Contingency covers unknowns identified early; anything beyond it becomes a written variation with its own cost and schedule effect, approved by both parties before it enters the plan.
You do, on full payment: bespoke deliverables, infrastructure definitions, test suites and documentation transfer to you. We retain ownership of pre-existing frameworks and grant a perpetual, non-exclusive licence to use them as embedded in your deliverables.
Send the problem in plain language and we will tell you whether software is the answer, and which discipline applies.
