Understand the business
- What we do
- Immersion in the business model, its actors, processes and constraints. Workshops to discover the domain before talking about technology.
- What you get
- A domain map and the prioritized problem.
The Turpial Method · MTIS
Software that can’t fail starts with understanding your business.
The Turpial Software Engineering Method (MTIS) builds on Scrum and adds the best of test-driven development (TDD), domain-driven design (DDD), experience design and change management. With artificial intelligence throughout the process and quality as a condition, not a final phase.
Why our own method
“We use Scrum” isn't enough
Scrum organizes work; it doesn't guarantee results
Scrum defines how to iterate. It doesn't say what to build or to what quality. That's why we start with the business and close every sprint with a demanding definition of done.
AI accelerates, and it also amplifies
Google's DORA research describes AI as an amplifier: it magnifies a team's strengths and its weaknesses, and links its adoption to lower delivery stability. That's why we use it inside a quality system.
Read the 2025 DORA report ↗Software nobody adopts is useless
A flawless system nobody uses is a failed project. Change management starts in the first phase, not on launch day.
Six phases
From understanding to evolving
The first two define us most: understanding the business and planning the work is what makes development go well.
Structure the plan
- What we do
- Product goal, a backlog prioritized by business value, architecture with documented decisions, risks, definition of done and a quality plan.
- What you get
- A plan, an architecture and a bounded estimate.
Experience design
- What we do
- Prototypes, user testing and a design system. Design doesn't stop when coding starts: it runs alongside construction.
- What you get
- Flows and design validated with real users.
Execute with quality
- What we do
- Sprints that end in a working increment, test-driven development, mandatory code review and continuous integration with quality gates.
- What you get
- A tested increment at the end of every sprint.
Deliver and drive adoption
- What we do
- Deployment, final testing, handover of code and accounts, training and an adoption plan so the system actually gets used.
- What you get
- A system in production, documentation and trained users.
Evolve with data
- What we do
- We measure delivery health with DORA metrics and plan the next iterations on that evidence.
- What you get
- A health dashboard and a roadmap.
Cross-cutting
What runs through the whole method
Quality: MTAC
Our Quality Assurance Method is a gate in every sprint, not a review at the end. It's the same method we use to audit other people's software.
Learn about MTAC →AI, with controls
We use AI in discovery, requirements, design, code, testing, review, documentation and operations. The rule: everything AI produces goes through human review and the same gates as code written by people.
Change management
We manage adoption with the ADKAR model: awareness, desire, knowledge, ability and reinforcement. Awareness and desire are built from the start; knowledge and ability, before each release.
Definition of done
When we say something is done
We publish it because it can be verified. An increment isn't done until it meets all of this:
- Automated tests for new logic
- Code review by at least one other engineer
- Continuous integration green
- Zero critical findings under MTAC
- Documentation updated
- Anything AI-generated passed the same gates
Next step
Two ways to start
Free
Technology Diagnostic
18 questions on strategy, processes, data, software, cybersecurity and people. See your level when you finish and get the full report by email.
Take the diagnostic → ServiceQuality X-Ray (MTAC)
An audit we perform on a specific system: what is broken, what is about to break and what to fix first.
See the X-Ray →
Next step
Have a project that can't go wrong?
Tell us what needs to be built. We start by understanding your business.