How we work
Fixed scope,no surprises midway
How we turn a loosely defined problem into a fixed price and timeline, and what we need from you for that date to hold.
How we plan a delivery
Six stages, from the first conversation to the metric in production. You always know which one the project is in, and what you get at the end of each.
- S01
Study: Understand
We listen and map the context before talking technology. You get the problem stated on one page.
- P02
Plan
Scope, phases, acceptance criteria and price. You decide whether to go ahead.
- A03
Assemble: Build
Short cycles with visible progress. Repository open from day one.
- D04
Demonstrate
You exercise the real flow, in your environment, before go-live.
- E05
Endorse: Validate
Quality and security review, criterion by criterion.
- S06
Ship & Sense
Deploy, operating documentation and a reading of the success metrics.
Indicative schedule
Reference figures for a mid-sized project. The real schedule, with dates, comes out of the scoping sprint in the delivery plan.
- 15 days
01 Discovery and technical alignment
Detailed scope, prototypes and architecture decisions.
- 6 weeks
02 Development and integration
Frontend, backend and APIs, with deliveries you validate each cycle.
- 2 weeks
03 Testing and technical validation
Automated, integration and end-to-end tests.
- 1 week
04 Documentation and observability
Solution documentation, logs, metrics and alerts.
- Final milestone
05 Final delivery and go-live
Code, access and the production environment, with metrics being read.
Phases overlap: testing starts alongside development. Dates are fixed in the contract and only move through a change of scope you request.
What you get before deciding
The scoping sprint is short, fixed-price and ends with four documents. If you decide not to proceed, they are yours, and you can take them to another vendor.
Fixed scope
A numbered list of what will be built, and an explicit list of what will not. It is the second list that prevents the argument at the end.
Acceptance criteria
For each item, the sentence that defines done in a testable way. No adjectives, no interpretation.
Delivery plan
Phases, milestones and dates, with what you receive at each one. No gap between payments goes past six weeks without a verifiable delivery.
Fixed price
Total value, payment terms and the change-of-scope rule. The price does not move because the work turned out hard: that risk is ours.
What we need from you
The most common cause of delay is not technical difficulty, it is waiting. We list these items before fixing the date and treat each one as a dependency with an owner and a deadline.
Decisions and people
Someone with authority to approve scope and sign-off, a technical point of contact, and agreed availability. Without a decision-maker, every question becomes a committee meeting.
Access and credentials
Cloud account, repository and DNS in your name. Infrastructure is born in your account and the code is yours from the first commit.
Systems and integrations
For each system: documentation, a test environment and a technical contact. Without a sandbox, testing happens in production, and we do not do that.
Data
Representative test data and an anonymisation rule. A system validated against three made-up records breaks on its first real day.
Environments and compliance
Who pays for infrastructure, where you validate before go-live, and the data-protection footing of the project, which changes architecture and logging.
After delivery
Who operates the system, where alerts go, and when handover happens. An alert that reaches nobody is decoration.
How missing items shape the plan
No classification ritual: we sequence the delivery around what exists. Work that depends on a missing item moves later; work that is ready moves up.
The schedule always reflects reality, not the original wish.
When something only you can provide is missing and cannot be worked around, the affected phase waits and the date moves with it, agreed in contract, no penalty.
The questions every founder asks
What if I change my mind mid-project?
Fixed scope is not frozen scope. You request the change in writing, we return the impact on timeline and price within two business days, and you choose: accept, swap it for an equivalent item we have not started, or drop it. Swapping costs nothing and is the most common path. Nothing gets built before your answer.
What if you get the estimate wrong?
It will happen, and the price does not move: estimation risk is ours and was priced in. If the timeline is affected, you hear about it the day we do, not the day before the milestone. And we propose cutting a low-value item before we propose a delay, the choice is yours.
Is the code really mine?
Yes, from the first commit, in your repository and your cloud account. No proprietary component of ours sits inside your system. You do not need us to keep going, and that is deliberate: a vendor who locks a client in through technical dependency does not have to deliver quality.
What if a prerequisite arrives late?
We flag it the day it becomes a risk, not the day it becomes a delay. We propose a workaround when one exists, synthetic data instead of the real set, for example, and we record the rework that workaround will cause. If it is blocking and there is no workaround, the phase freezes and the date shifts by the same amount.
Shall we fix the scope?
Thirty minutes, no charge. You leave knowing whether the problem is worth solving, whether it is worth solving with software, and whether we are the right people.