Can Wavesteam estimate a project before the requirements are complete?
Yes. When requirements are incomplete, Wavesteam can provide a dated, assumption-based range for early investment screening. It is not a final commitment. A fixed-price contract needs a first-release boundary, integrations, data, quality and production responsibilities, and acceptance evidence. Every early range should show its baseline, lower and upper scenarios, exclusions, and the variables most likely to move it.
An early estimate can reveal whether the available budget is clearly insufficient, broadly plausible, or worth further discovery. The less evidence available, the wider the range must be.
| Estimate stage | Available evidence | Responsible output | Decision it supports | It cannot support |
|---|---|---|---|---|
| Order-of-magnitude | Business objective, main users, channel, a few constraints | Wide range, assumptions, exclusions, major risks | Continue discovery, buy, or build | Fixed contract or release date |
| Budget estimate | Main flow, prototype, clients, interface list, data samples | Work-package ranges, external fees, risk scenarios | Internal approval and funding | Treating an untested interface as fixed effort |
| Contract quotation | Scope baseline, acceptance, responsibility, real interfaces, quality targets | Rates or total, milestones, change and payment terms | Formal procurement and delivery | An unlimited future roadmap |
Even the broadest estimate needs a short basis: the problem, user task, first-release clients and roles, administration, payment/map/AI/hardware/legacy links, migration, expected region and volume, release responsibility, and exclusions. Mark each unknown as an assumption and state how a contrary finding changes the range.
When defining budget, scope, and cost assumptions, also compare How much does a custom app cost? and Can a project launch its core in phases when the budget tightens halfway through?; the linked guidance adds context that should be considered in the same decision.
Resolve the variables that can change the decision
Simple submission and lookup is different from concurrent inventory, refund, split settlement, approval, and audit. A documented API with a sandbox differs from a legacy system that must be reverse-engineered across several owners. Clean new data differs from years of conflicting keys and zero-downtime migration. One web client differs from native mobile, several mini-programs, offline work, or hardware. Normal business-hours service differs from strict resilience, cross-region recovery, or continuous on-call.
Wavesteam turns these into a conditional estimate, for example: one WeChat mini-program and administration client, no historical migration, two sandboxed APIs, and business-hours support form the baseline; native mobile, undocumented interfaces, or historical-order migration add named work packages and trigger re-estimation. The numbers come from the team's current role rates and effort—not another client's case.
When the interval is too wide, spend discovery on the unknown most likely to reverse the decision: whether the legacy interface works, the device protocol is stable, users complete the critical task, data can migrate, or the platform permits the category. A runnable proof or trial migration often narrows cost more than a long feature inventory.
Wavesteam supplies the initial range with its date and assumptions and tells the client which business materials would strengthen it. When prototype work, interface testing, data profiling, and a vendor-comparable breakdown are needed, we define a separate discovery deliverable that remains useful even if construction stops. We do not promise unlimited detailed design for free or hide consulting effort in a later build price.