TL;DR: Fixed price works when you have a detailed specification that won't change. Time & Materials works when the scope will evolve. The Standish Group data shows fixed-scope projects fail at 5x the rate of agile ones (59% vs 11%). Most software projects should use T&M with sprint-level budgets and clear milestones. Fixed price makes sense only for well-defined phases with documented specs.
The pricing model nobody talks about honestly
Every article comparing fixed price vs. time and materials picks a side. Fixed-price advocates say it protects your budget. T&M advocates say it delivers better software. Both are right in specific scenarios and dangerously wrong in others.
The real question isn't which model is "better." It's: how well do you understand what you're building?
Fixed price: when it works (and when it kills projects)
The promise
You agree on scope, timeline, and cost before development starts. The agency absorbs the risk of scope changes. You know exactly what you'll spend.
When it actually works
Fixed price delivers well in these scenarios:
- Repeatable projects. The agency has built this type of system many times. An e-commerce store, a corporate website, a standard mobile app. They know the scope, the pitfalls, and the timeline from experience.
- Documented specifications. You have a detailed PRD, approved wireframes, finalized data models, and defined API contracts. Everything is written down and signed off.
- No discovery needed. You've already done the research phase. You're not exploring what to build; you're executing a plan.
When it goes wrong
Fixed price creates problems when:
- Requirements aren't finalized. You describe features in broad strokes. The agency estimates based on assumptions. Those assumptions are wrong. Conflict follows.
- The project requires experimentation. AI/ML products, user-facing apps that need market validation, or complex integrations with legacy systems. These require iteration, which fixed price penalizes.
- Change orders pile up. Every modification becomes a negotiation. "That's out of scope" becomes the most common phrase in your project meetings.
The data
The Standish Group's CHAOS Report has tracked software project outcomes since 1994. Their findings for waterfall/fixed-scope projects:
- 59% challenged or failed (over budget, over time, or missing features)
- Only 11% of agile projects fall into the same category
Fixed scope doesn't eliminate risk. It redistributes risk in ways that incentivize cutting corners.
How agencies price fixed-price projects
Here's what happens behind the scenes: an agency estimates the real effort at, say, 800 hours. They know that scope changes, miscommunication, and unforeseen complexity add 30-50% to any project. So they quote 1,100-1,200 hours worth of work.
You're paying a 30-50% risk premium. If the project runs smoothly, the agency profits from the padding. If it runs over, the agency either delivers less quality or requests change orders. The "safety" of fixed price costs you money either way.
Time & Materials: the flexibility trade-off
The promise
You pay for actual hours worked. The scope can evolve as you learn from user feedback, market changes, or technical discoveries. You get transparency into how time is spent.
When it actually works
T&M delivers well when:
- The project will evolve. Software products almost always change direction. A competitor launches a feature, user testing reveals unexpected behavior, regulatory requirements shift.
- You have an active product owner. Someone on your team who reviews progress weekly, adjusts priorities, and makes decisions quickly.
- You want transparency. You see exactly what's being built, hours logged, and can redirect effort at any sprint boundary.
When it goes wrong
T&M creates problems when:
- Nobody is watching. Without active oversight, hours can drift. A one-week task takes three weeks because priorities shift mid-sprint or nobody is available for reviews.
- There's no budget framework. "We'll just pay for hours" without sprint budgets or milestone caps leaves your costs unbounded.
- Trust is missing. If you don't trust the agency's time reporting, T&M will be a constant source of friction.
The hybrid model: what most projects actually need
The best projects we've run at Globalbit use a hybrid approach. Neither pure fixed-price nor open-ended T&M.
Sprint-level budgets with flexible scope
Here's how it works:
- Discovery phase (fixed price). 2-4 weeks, fixed deliverables. Produces a detailed project plan, architecture, and accurate estimates. Typically $5,000-15,000.
- Sprint planning with budget targets. Each 2-week sprint has a budget target (e.g., $15,000-20,000). You and the agency agree on sprint goals. If something is more complex than expected, scope adjusts within the budget, not the other way around.
- Monthly budget reviews. You review actual spend vs. planned spend monthly. If the project is tracking over budget, you cut lower-priority features. If under budget, you pull forward next-phase work.
- Milestone-based checkpoints. Major deliverables (MVP, beta, launch) trigger a review. You evaluate progress, adjust the remaining scope, and decide on next phase commitment.
Why this works
- You have budget predictability (monthly targets) without the rigidity of fixed price
- The agency can build the right thing, not just the thing that was specified 4 months ago
- Changes are absorbed naturally into sprint planning instead of becoming adversarial negotiations
- Both sides stay aligned because goals are reviewed every two weeks
A decision framework
| Factor | Fixed Price | T&M | Hybrid |
|---|---|---|---|
| Requirements fully defined? | Yes → FP works | No → use T&M | Partially → hybrid |
| Will scope change? | No → FP works | Yes → use T&M | Probably → hybrid |
| Active product owner available? | Not required | Required | Required |
| Risk tolerance | Low (pay upfront premium) | Medium (need oversight) | Balanced |
| Project duration | Under 3 months | Any length | 3+ months |
| Budget certainty | Exact (with premium) | Range only | Range with monthly caps |
Real-world scenarios
Scenario 1: Corporate website redesign Best model: Fixed price. The scope is well-defined (pages, features, content structure). The agency has built 50 similar sites. Low uncertainty. Estimated cost: $30,000-60,000 fixed.
Scenario 2: Customer-facing SaaS product Best model: T&M with sprint budgets. The product will evolve based on user feedback. Features will be added, cut, and modified during development. Estimated cost: $15,000-25,000/month, with quarterly milestone reviews.
Scenario 3: Legacy system migration Best model: Hybrid. Start with a fixed-price discovery and migration plan. Execute migration sprints on T&M with milestone checkpoints because legacy systems always have surprises. Estimated cost: $10,000-15,000 for discovery, then $20,000-30,000/month for execution.
Scenario 4: AI/ML proof of concept Best model: T&M. The output depends on data quality, model performance, and iterative experimentation. You can't fix-price research. Estimated cost: $10,000-20,000 for the PoC phase, time-capped.
Questions to ask the agency
Before committing to a pricing model, ask:
- "What pricing model do you recommend for this type of project?" If they push fixed price for a vague scope, they're either padding heavily or planning change orders.
- "How do you handle scope changes under fixed price?" If the answer involves complex change order processes, expect friction.
- "What visibility do I get into hours worked under T&M?" You should see weekly timesheets, sprint reports, and burndown charts. If they can't provide this, find another agency.
- "Can we structure this as a paid discovery followed by phased delivery?" The discovery phase often costs 5-10% of the total project budget and saves you 20-40% in avoided rework.
Frequently asked questions
Can I switch pricing models mid-project? Yes, and it's common. Many projects start with a fixed-price discovery phase and transition to T&M for the build. The key is agreeing on transition terms upfront — rates, team composition, and sprint cadence.
How do I prevent T&M from becoming an open checkbook? Two mechanisms: sprint-level budgets and milestone-based reviews. Set a monthly budget target and review progress every two weeks. If the agency can't show meaningful output for hours billed, that's a conversation to have immediately.
Why do agencies prefer T&M? Because it lets them build better software. When an agency isn't penalized for discovering that a feature needs redesign, they'll propose the better solution instead of delivering what was spec'd. That said, not all agencies use this flexibility responsibly. Check references.
What's the best model for a startup MVP? T&M with a maximum budget cap. Startups need speed and flexibility. Budget-capped T&M gives you both: the agency works efficiently, you control total spend, and features can pivot based on early user feedback. Talk to us about structuring your project.