Executive Summary
Most providers describe their process in five generic steps that could belong to anyone. Discovery, planning, development, testing, delivery. That tells you nothing, because the process is not where providers differ. The differences sit in the specific decisions inside it.
So this page describes the actual decisions. Who conducts our technical interviews and why. What we do in the first week that most providers leave to the client. How sprint governance runs across a timezone gap. What happens in month nine when someone resigns.
The reason process specificity matters is measurable. PMI reports that organisations with mature project governance complete around 89% of projects on time and on budget, against 36% for low maturity organisations, and that gap holds regardless of where engineers sit. Offshore does not create governance problems. It exposes them.
If you are evaluating an offshore software development company and every provider’s process page reads identically, this one is written to be checkable instead.
Key Findings
- Governance premium: PMI reports mature governance organisations complete around 89% of projects on time and on budget, against 36% for low maturity
- Screening approach: Technical assessment is engineer led, combining live coding, a take home task, and a separate written English communication evaluation
- Retention outcome: Developer churn on client engagements stays below 2%, against the 20 to 30% category norm
- Timezone design: Indonesian delivery teams work Australian business hours, 1 to 3 hours from AEST depending on daylight saving
- Rate structure: AUD 52 to 76 an hour, all inclusive, billed in Australian dollars through an Australian entity
- Scope discipline: Roughly 52% of projects experience scope creep at an average 27% cost overrun per PMI data, which our sprint governance is built to contain
- Local benchmark: A senior Australian developer costs near AUD 210,000 in year one fully loaded, or an effective AUD 123 an hour
Step One: Scoping, and Telling You When Not to Proceed
Every engagement starts with a conversation about the problem rather than the roles. Roles are a solution and it is worth checking the solution fits.
We ask five things. Which roles you cannot fill locally and how long they have been open. What your existing engineering leadership looks like. What the roadmap is over the next twelve months. What your current stack and codebase maturity is. And what you have already tried.
That last question matters more than it sounds. A business that has tried a freelance marketplace and had a bad experience needs a different conversation to one that has never engaged offshore capacity before.
We then produce a costed comparison against your genuine loaded local cost, with our assumptions stated so you can pressure test them. The benchmark we use is not base salary. A senior developer on a AUD 150,000 base costs near AUD 210,000 in year one once you add 12% superannuation, state payroll tax at roughly 5%, workers compensation, leave provisions, equipment, and a recruitment fee.
And sometimes the honest output of this step is that offshore does not suit your situation. Very short engagements, single roles where you have strong local pipeline, or work requiring constant physical presence with non technical stakeholders. We say so at this stage rather than after a contract is signed.
Step Two: Engineer Led Technical Screening
This is the step where providers differ most and disclose least, so here is ours in specific terms.
Our technical assessment is conducted by working engineers, not recruiters. A recruiter running a technical interview from a competency checklist is filtering for interview performance rather than engineering capability, and that difference shows up in your code review queue rather than in anyone’s marketing.
The assessment has three components. A live coding session against a realistic problem, not an algorithm puzzle. A take home task sized to a few hours with a review conversation afterwards, because how someone explains their choices matters as much as the code. And a separate English communication evaluation covering written technical communication specifically, since ambiguous tickets and unclear pull request descriptions cause more rework than accents ever do.
We recruit from the top few percent of Indonesian technical talent, and our engineering team in Indonesia numbers more than 150 specialists, weighted mid to senior rather than junior heavy.
The commercial logic behind that seniority weighting is direct. The hidden cost of a cheaper junior heavy team is your Australian senior engineers’ review time. Ten hours a week at a loaded local cost near AUD 123 an hour is roughly AUD 60,000 a year that appears nowhere as a line item.
Step Three: Named Shortlist and Your Interviews
We present named engineers with CVs, relevant stack experience, and confirmed availability, typically within two to four weeks of a settled brief depending on stack and seniority.
Then you interview them. We treat that as non negotiable rather than as a courtesy, for a straightforward reason: you would not hire locally without meeting the person, and a dedicated engagement is closer to hiring than to buying a project.
We also tell you the allocation percentage for each engineer. If someone is partly committed elsewhere, you need that in the shortlist rather than in sprint three.
Where we differ from a platform matching model is what happens if you reject the shortlist. We go back and screen again rather than widening the criteria to produce a faster yes. That takes longer and produces better outcomes, and it is the main reason our hiring cycle is not the fastest in the category.
Businesses that need to hire software developer capacity urgently should know that trade off up front. If speed is the binding constraint, a platform led provider may genuinely suit you better.
Step Four: The Pilot
New clients start with a defined pilot on real scope, with success criteria agreed in writing before it begins.
Real scope means a self contained feature, a defined integration, or a piece of technical debt with clear acceptance criteria. Not a toy exercise. A pilot with no dependencies and no ambiguity proves nothing about how a team handles your actual environment.
We also suggest introducing a deliberate mid pilot priority change. How a team handles a change request is more informative than how it handles a clean run, and both sides learn more from that single moment than from the rest of the pilot.
Three things worth watching during it. Written communication quality in tickets and pull requests. Code review outcomes, specifically how much rework your senior engineers request. And responsiveness when something breaks, measured in hours.
We expect you to be genuinely willing to walk away at the end. A pilot you have already emotionally committed past is not a pilot, it is an onboarding with extra steps.
Step Five: Onboarding, Including the Part That Is Ours
Most providers hand onboarding entirely to the client and call it collaboration. We split it explicitly, because the split is where engagements quietly fail.
Ours: employment, payroll, equipment, workspace, tooling access provisioning on our side, Australian business hours scheduling from day one, and a named delivery lead assigned before the first sprint.
Yours: repository access, development and staging environments, credentials, a live architecture walkthrough rather than a document handover, and a first ticket sized to be completable in week one.
We will chase you for those five things, which some clients find mildly annoying and all clients benefit from. Engineers waiting two weeks for credentials are billable and idle, and it sets a tone that is difficult to reverse later.
We also assign an internal buddy on our side for the first month and ask you to nominate one on yours. Someone who answers questions without the offshore engineer needing to escalate or wait a day.
“Partnering with Upscalix has allowed us to focus on growing our business while knowing our software development is in capable hands. Their professionalism, communication, and technical expertise have exceeded our expectations.”
Jake, Co-Founder and CEO, Sophiie AI
Step Six: Sprint Governance Across the Timezone Gap
This is where the one to three hour gap earns its keep, because governance routines only work if they can happen as conversations.
Our engineers work Australian business hours from Indonesia. Daily standups happen in your morning, short and live. Sprint planning happens as a meeting with the offshore engineers present and speaking, not as a document exchange with follow up questions the next day. Sprint reviews and retrospectives run fortnightly with the same participation.
Monthly, you get delivery metrics: velocity, defect escape rate, code review turnaround, and hours consumed against forecast. You should not have to email anyone for that data, and if you do, that is a provider problem worth raising immediately.
On scope, we keep a written change process even under a dedicated team model. Roughly 52% of projects experience scope creep at an average 27% cost overrun, and dedicated teams are not immune to it. They just absorb it as slower roadmap progress rather than as a larger invoice, which is easier to miss and equally expensive.
Quarterly, we run an engagement review covering team composition, capacity against roadmap, and whether the model still fits. Roadmaps change and engagements should change with them rather than persisting out of inertia.
What the Process Deliberately Does Not Include
Absence is informative in a process description, so four things we have chosen not to do.
We do not run automated candidate matching. Platform led providers using AI assisted matching present candidates faster than we do, and that speed has genuine value. Our assessment is engineer led, which is slower and produces a different quality distribution. If speed is your binding constraint, that trade off works against us and you should know it before starting.
We do not rotate engineers between clients to maximise utilisation. That practice is efficient for providers and destructive for clients, because accumulated codebase context is the asset an offshore engagement produces and rotation resets it.
We do not fixed price vague briefs. A provider quoting a firm number against a two page description either has not read it properly or intends to recover margin through variation orders. We will tell you when a specification is actually a direction.
We do not bundle non engineering functions. If you need support agents or bookkeepers alongside developers, we cannot help with the first two and will point you toward an enterprise managed staffing provider. Businesses needing broader BPO business process outsourcing capability alongside engineering should plan for two providers rather than expecting one to do both well.
The Upscalix Process at a Glance
| Step | What We Do | What You Do | Typical Duration |
|---|---|---|---|
| 1. Scoping | Costed comparison with stated assumptions, honest fit assessment | Share roles, roadmap, stack, and what you have tried | 1 week |
| 2. Technical screening | Engineer led live coding, take home task, English evaluation | Nothing, this is ours | Ongoing bench plus role specific |
| 3. Named shortlist | Present named engineers, CVs, allocation percentages | Interview the engineers | 2 to 4 weeks |
| 4. Pilot | Deliver real scope against written success criteria | Define scope, agree criteria, be willing to walk away | 30 to 60 days |
| 5. Onboarding | Employment, equipment, AU hours scheduling, named delivery lead | Repo access, environments, architecture walkthrough, first ticket | 1 week |
| 6. Sprint governance | Live standups, sprint ceremonies, monthly delivery metrics | Product ownership, prioritisation, acceptance | Ongoing |
| 7. Scale or adjust | Quarterly engagement review against roadmap | Decide capacity against backlog reality | Quarterly |
How We Handle Scope Changes
Change handling is where process claims meet reality, so this is specific rather than reassuring.
Under a dedicated team engagement, changes go into sprint planning. No variation order, no repricing, no formal approval chain. You reprioritise the backlog and the next sprint reflects it. That flexibility is the main structural advantage of the model and it is the reason most ongoing product work belongs here rather than in a fixed price structure.
What we do insist on is that changes are visible. Reprioritisation is fine. Silent scope expansion, where new work arrives without anything leaving, produces the slower roadmap version of scope creep. Roughly 52% of projects experience scope creep at an average 27% cost overrun, and dedicated teams absorb it as delayed delivery rather than as a larger invoice, which is easier to miss and equally expensive.
So our sprint planning includes an explicit trade off conversation. If something new comes in, something moves out or the roadmap date moves. Stating that plainly at the time is less comfortable than agreeing to everything and less expensive than discovering the gap at quarter end.
Under a project structure, changes go through a written variation process with a stated turnaround commitment and a pricing basis agreed at contract signature. Businesses commissioning a custom mobile app development build under a fixed structure should get that pricing basis documented before work starts, not negotiated when the first change arrives.
What Happens When Someone Resigns
Every provider has a slide about retention. Fewer have a documented answer to this question, so here is ours.
Our developer churn on client engagements stays below 2%, against the 20 to 30% annual attrition typical of offshore delivery centres. That is the number we would ask you to weight most heavily in any provider comparison, because accumulated codebase context is the thing offshore engagements most often fail to deliver.
When it does happen, three commitments apply. We present a replacement shortlist within a defined window rather than an indefinite best efforts arrangement. We run a documented handover covering architecture decisions, known issues, in flight work, and credentials inventory. And the replacement’s ramp up period is not billable at full rate, because you should not pay twice for the same context transfer.
The structural reason our churn stays low is not a culture programme. It is that we recruit mid to senior engineers, pay competitively in the Indonesian market, and place people on engagements long enough to become genuinely valuable rather than rotating them between clients to maximise utilisation.
Reporting and Tooling
Governance only works if the data arrives without being chased, so this part is deliberately specific.
You receive live visibility into sprint velocity, defect escape rate, code review turnaround, and hours consumed against forecast. Not a monthly summary document, which is a generation behind current practice and tells you about last month rather than this week.
We do not ask you to adopt our tooling. Whatever you already use for issue tracking, version control, continuous integration, and code review, our engineers work inside it. Engineers joining an existing team should adapt to that team’s tools, not the reverse, and a provider insisting on its own stack is optimising for its internal convenience rather than your delivery.
The named delivery lead assigned to your engagement is your escalation contact and attends your sprint ceremonies. You get a name and a person rather than a shared inbox.
On documentation, we treat it as a deliverable rather than an afterthought. Architecture decisions, deployment processes, and known issues get recorded as work progresses, which is the only reason a handover is possible if an engagement ever winds down. Any outsourcing provider treating documentation as optional is creating a dependency you will feel later.
How the Process Adapts by Work Type
The seven steps hold, but three work types change the emphasis.
Product roadmap work, meaning ongoing development with shifting priorities, runs on a dedicated team model with sprint level prioritisation. This is the majority of what we do and the process above describes it directly.
Bounded project work with a documented specification can run on a fixed structure with formal acceptance criteria. We will tell you honestly whether your specification is actually specific enough for that, using a simple test: could two different teams read it and produce functionally identical products? If not, you have a direction rather than a specification.
Specialist work changes step two most. Engaging an AI developer or a specialist data role means a different assessment panel and a narrower cost advantage, around 35 to 42% against local fully loaded cost rather than the 45 to 60% available on general engineering, because global demand for those skills has compressed the geographic arbitrage.
Across all three, our rate structure stays the same. AUD 52 to 76 an hour all inclusive for general engineering, covering recruitment, employment, HR administration, tooling, equipment, and delivery management, billed in Australian dollars through an Australian entity headquartered at Level 11, 580 Collins Street, Melbourne.
Whether the engagement is a custom software development build, a dedicated squad, or two engineers alongside your existing team, the quote you receive is the number you pay.
What This Means for Australian Businesses
Process pages are usually the least useful content a provider publishs, because generic steps describe every provider equally well and therefore describe none of them.
The parts of ours worth checking are the specific ones. Who conducts technical interviews. What the screening pass rate is. Whether a replacement’s ramp up is billable. Whether you interview engineers before committing. Whether the provider will tell you when offshore does not suit your situation.
Ask every provider on your shortlist those five questions. The answers vary enormously, and they predict your experience far better than any process diagram does.
If you want to start at step one, tell us the roles you cannot fill and what you have already tried. The scoping conversation and the costed comparison cost you nothing.
Fair enough?
FAQ
How does the Upscalix offshore development process work?
Seven steps: scoping with a costed comparison, engineer led technical screening, a named shortlist you interview, a pilot on real scope, split onboarding, live sprint governance, and quarterly scaling reviews.
Who conducts technical interviews at Upscalix?
Working engineers, not recruiters, using live coding against realistic problems plus a take home task and a separate written English communication evaluation.
Do I get to interview offshore engineers before they join my team?
Yes, and we treat it as non negotiable rather than optional. We present named engineers with CVs, stack experience, and allocation percentages before any commitment.
What happens if an Upscalix engineer leaves mid engagement?
A replacement shortlist within a defined window, a documented handover covering architecture and in flight work, and a ramp up period that is not billed at full rate.
Sources
- https://www.pmi.org/learning/library/scope-creep-rising-11308
- https://ia.acs.org.au/article/2026/australia-s-tech-workforce-shrinks-for-first-time.html
- https://www.ato.gov.au/businesses-and-organisations/super-for-employers/paying-super-contributions/how-much-super-to-pay
- https://www.seek.com.au/career-advice/role/software-engineer/salary
- https://www.deloitte.com/global/en/issues/work/global-outsourcing-survey.html
- https://www.intertec.io/en/resources/blogs/preventing-scope-creep-software-budget-control
- https://expert360.com/articles/true-cost-of-hiring-employee-australia