Fixed price agrees a price for defined software work; time and materials pays for agreed work at agreed rates. Neither label alone establishes what happens when scope changes, a dependency fails or a budget limit is reached. Compare those decisions as carefully as the initial quote.

This guide keeps the choice practical: what is known, what remains uncertain, who can authorise change and what evidence shows that the work is acceptable. It is a planning guide, not legal advice or a market price list.

Fixed price: what it covers

A fixed-price proposal needs an identifiable scope, assumptions, exclusions and acceptance conditions. It may follow a discovery phase that investigates the work before a build price is agreed. Record how milestones are reviewed and paid, including any dependencies the customer must provide.

Fixed pricing can suit bounded work with understood interfaces and stable requirements. It does not make missing information disappear. Check how the agreement treats changes, delays, third-party charges and work outside the original scope before treating the headline amount as the whole budget.

Questions to resolve before committing

  • Can both parties describe the same finished behaviour and demonstrate acceptance?
  • Are testing, deployment, documentation and handover included?
  • Which assumptions have evidence, and who investigates those still uncertain?
  • How are defects against agreed behaviour distinguished from new requests?

A supplier may allow for uncertainty in its estimate; ask what risks and exclusions the quote assumes rather than applying a generic premium. Vague requirements can leave both parties with incompatible expectations. The software development problems guide discusses that wider delivery risk.

Time and materials: what you review

Under time and materials, agree the rates, chargeable activities, third-party costs and reporting arrangements. The work can be reprioritised within the agreed process, but a changed priority still consumes capacity and may displace something else.

This can suit investigation, evolving products or ongoing improvement. It needs someone authorised to order priorities and review progress. Ask for understandable records of work, demonstrations and a forecast of remaining effort. Timesheets explain expenditure; they do not by themselves establish that the required outcome works.

Compare total scope and evidence, not day rates alone. A cheaper rate does not establish a cheaper accepted outcome. Equally, a higher rate or a familiar supplier is not proof of effectiveness. Make review and stop decisions explicit under either model.

Compare the decisions, not just the labels

Questions for the two commercial approaches
DecisionFixed priceTime and materials
What is being bought?The defined scope under the agreed conditions.Agreed work charged by time and applicable costs.
What changes?Review the effect on scope, price, acceptance and timing.Review priorities, effort, spend and displaced work.
What does the customer provide?Requirements, dependencies and timely acceptance decisions.Priority decisions, feedback and expenditure review.
What needs protection?Scope clarity and acceptance quality.Visibility of spend, progress and remaining uncertainty.
Who carries a particular risk?Check the agreed allocation; the model's name is not a complete risk assessment.

Distinguish an estimate, a threshold, a cap and a commitment

An estimate forecasts effort or cost using stated assumptions. An approval threshold identifies when someone must authorise further work. A spending cap limits authorised expenditure under the agreed terms. A delivery commitment describes what must be delivered and accepted.

Do not assume that a capped T&M arrangement automatically requires the supplier to finish any remaining scope without further payment. Agree what happens as the cap approaches: work may pause, scope may be reduced or a revised commitment may need approval. Also define whether third-party costs, tax and other charges are inside the stated figure.

Phased fixed pricing is another option. Investigate a bounded question, review the findings and decide whether to commission the next phase. Do not treat approval of discovery as automatic approval of the build. A hybrid needs the same clarity about evidence, expenditure and stop points as either basic model.

What happens when an assumption changes?

Use a change record before continuing work affected by a failed assumption. Keep the original requirement visible; silently rewriting it can conceal a missed acceptance condition.

  1. Original assumption: what was believed, and where was it recorded?
  2. New evidence: what has been observed, in which environment, and what is still unknown?
  3. Affected outcome: which requirement, acceptance check or dependency is affected?
  4. Current position: what work is already accepted, what remains useful and what further work is proposed?
  5. Options: investigate within a bounded allowance, reduce scope, change the approach, revise the agreement or stop.
  6. Decision: record expected cost and timing effects, the approver, date, conditions and next review. Do not substitute silence for approval.

Hypothetical example: an unavailable export field

A fictional booking project assumes that an external system can export a reference needed to match customers. A test account produces an export without that field. The team does not yet know whether this is a permission issue, an account limitation or a missing capability.

Under a fixed-scope arrangement, compare the finding with the recorded assumption and agreed investigation responsibilities. Under T&M, identify what investigation is authorised and whether it fits within the remaining allowance. Neither approach makes the technical uncertainty disappear.

The decision-maker could approve a bounded investigation, consider a different matching process or pause the affected feature. Record the acceptance consequences of each option: a manual workaround is not equivalent to an automated match unless that revised outcome is explicitly accepted. Update the forecast and delivery plan after the decision, not merely the invoice.

This example illustrates a decision process, not an actual project, quotation or claim about who would legally owe additional work.

How to choose: five questions

  1. How much is understood? Identify evidence for the scope and interfaces, not just confidence in the feature list.
  2. What budget constraint is real? Clarify authorised spend and which scope could change if the budget cannot.
  3. Can the work be separated into useful phases? Each phase needs a decision and an outcome, not simply a billing date.
  4. Who can review and decide? Both approaches need customer participation; agree availability rather than assuming a universal number of hours.
  5. How will you evaluate the supplier? Use transparent evidence and a bounded first engagement rather than assuming a pricing model creates trust.

If the build decision itself is unresolved, start with the custom versus off-the-shelf guide. For assumptions and cost drivers, read the project estimation guide; its figures are not a quotation for your project.

Protect the outcome under either model

Write down the working process, acceptance checks and route for disagreement. Keep review evidence tied to the actual version delivered. The software QA acceptance guide explains a release evidence pack.

Ask appropriate advisers to review ownership, licences, access, notice and handover provisions for the proposed agreement. Preserve a workable exit path: who can access repositories, retrieve documentation and continue operation if the relationship ends? This guide does not determine intellectual-property ownership or contractual remedies.

A small discovery phase or milestone can test the working relationship before a larger commitment. Use the supplier vetting checklist alongside the evidence from that work.

Discuss requirements through Visuosofts application development, or contact the team with the intended outcome, known dependencies and spending constraints. The useful starting point is a shared understanding of the work, not a promise that one pricing model always wins.

Software pricing questions

Which is better: fixed price or time and materials?

Choose using scope evidence, uncertainty, budget constraints and decision capacity. Neither model guarantees a successful outcome; both need clear acceptance and change arrangements.

Does a spending cap guarantee delivery of every feature?

No such conclusion follows from the label alone. Agree the delivery commitment and what happens when the cap approaches, including pause, scope and approval rules.

How should we compare developer rates?

Compare the included work, relevant evidence, supervision needs and total expected cost. Obtain current quotes for the defined scope rather than treating a generic rate range as a budget.

What should happen when an assumption proves wrong?

Record the evidence, affected acceptance condition, remaining uncertainty and options. Obtain the authorised decision before continuing work that depends on a changed scope or expenditure approval.

Who owns the code and supporting assets?

Have the proposed ownership, licence and access provisions reviewed for the actual agreement. Check third-party components and the handover requirements rather than inferring ownership from the pricing model.