Fixed Price vs Time and Materials: Choosing the Right Software Contract
How fixed price and time and materials contracts compare on risk, flexibility, and true cost, and how to choose the right model for your next project.
How many people does it take to build your software? An MVP needs three to five, a mid-size product needs six to nine, and an enterprise system needs ten or more split into smaller squads. Those numbers matter because team structure is where proposals hide both padding and gaps: a vendor quoting eight people for an MVP is charging you for coordination you do not need, and one quoting two people for a multi-role platform is deferring costs you will pay later. This guide explains what each role actually does, what a sensible team looks like at each project size, and how AI-assisted development is shifting the ratios.
Seven roles cover almost every project, though one person often holds more than one of them on smaller teams.
A team matched to scope, with shared and fractional roles at the small end and dedicated specialists as complexity grows. Here is the pattern we staff against:
| Project size | Team size | Typical composition |
|---|---|---|
| MVP ($25k–$75k) | 3–5 | 2 full-stack developers, part-time designer, project manager who also coordinates QA, fractional tech lead |
| Mid-size product ($75k–$250k) | 6–9 | 3–4 developers, dedicated QA engineer, designer, project manager, tech lead, fractional DevOps |
| Enterprise system ($250k+) | 10+ | Multiple squads of 5–8, each with developers and QA; shared architect, design, DevOps, and delivery management |
The enterprise row follows a principle worth knowing: the Scrum Guide recommends teams of ten or fewer because smaller teams communicate better, so large systems are built by several small teams rather than one big one. Headcount above ten on a single backlog is a coordination tax, not extra speed.
The ratios shift, and review capacity becomes the constraint instead of typing speed. Google's 2025 DORA report found 90% of technology professionals now use AI at work, with over 80% reporting productivity gains, and the same research found roughly 30% place little or no trust in AI-generated code. Both findings show up in staffing. A two-developer MVP team now ships what took three or four developers in 2023, which is part of why MVP budgets have held steady while scope has grown. At the same time, experienced engineers who can judge generated code matter more, because the risk has moved from writing code slowly to accepting bad code quickly. QA shifts earlier and reviews harder. We cover the mechanics in how AI is changing software development; the structural summary is smaller teams, higher seniority, and no reduction in testing.
Three patterns account for most of the troubled projects we inherit.
Outsource the first build, augment when you have an engineering core with specific gaps, and hire when the product is proven and ongoing. Building this structure in-house means recruiting six to nine people before writing a line of code, and our cost-to-hire breakdown shows that often exceeds the cost of the software itself once salaries, benefits, and recruitment are counted. An outsourced team arrives already structured, with the fractional roles genuinely fractional on your invoice. The middle path puts rented specialists inside your own process and management, and we compare when each model wins in staff augmentation vs. project outsourcing. Whichever route you take, the contract shape matters as much as the roster; our guide to fixed price vs. time and materials explains how pricing models change team behavior.
Ask three questions and insist on specific answers. Who on this roster is dedicated to my project and who is shared, and at what fraction? Who is the single technical decision-maker, and will I meet them before signing? When does QA start, with the right answer being the first sprint rather than the last month. A vendor with a sound structure answers all three in a minute; a vendor selling a bench answers with role titles and rate cards. At Kodenique we staff projects from our Southeast Asia delivery team using exactly the compositions in the table above, and we will show you the named roster, each person's allocation, and who holds the technical decisions before anything is signed. Tell us what you are building and we will propose the smallest team that can deliver it, or get in touch if you want a second opinion on a roster someone else has quoted.
How fixed price and time and materials contracts compare on risk, flexibility, and true cost, and how to choose the right model for your next project.
Realistic app development timelines by project type, from 8-week MVPs to year-long platforms, and the factors that decide whether you ship on time.
What EV charging management software costs in 2026, comparing per-port SaaS fees, white-label platforms, and custom builds, from a team that built one.