Contact

How to evaluate a software development proposal: criteria for decision makers

A still life on a marble table: two heavy books, a quill, an inkwell, coins and an envelope. (William Michael Harnett, The Banker's Table, 1877)
William Michael Harnett: The Banker's Table, 1877 · The Metropolitan Museum of Art · Public domain
Category
Project
Published
Reading time
~3 min read
Author

Summary

The bottom-line figure of a proposal says little on its own. Compare what the developer commits to, what is excluded, and what happens when things change mid-project.

Why looking only at the bottom line is not enough

Two proposals can differ dramatically even when responding to the exact same brief. The reason is rarely that one vendor is expensive and the other cheap. Far more often, they interpret the scope differently: one includes testing, data migration and user training, while the other does not.

A proposal primarily reveals how a developer thinks about your problem. Read it for that insight: the price tag is merely the consequence.

What should a strong proposal include?

  • Understanding of the problem: the developer summarises your business goal in their own words.
  • Scope: what will be built, and explicitly what is excluded from the quote.
  • Assumptions: the premises behind the estimate and what could cause it to change.
  • Timeline and milestones: when you will see working software you can actually test.
  • The team: who will do the work and who your day-to-day contact will be.
  • Handover and IP ownership: who owns the source code and what deliverables you receive.
  • Post-launch terms: warranty, maintenance and ongoing operations.

Fixed price or time and materials?

Comparing pricing models
ModelAdvantageRiskWhen it fits
Fixed pricePredictable upfront costEvery change requires renegotiation; vendors price in contingency for uncertaintyWell-defined, tightly scoped projects
Time and materialsFlexible; priorities can shift as you learnFinal total is not fixed upfront; requires active budget trackingEvolving requirements or exploratory products
Phased (discovery first, then fixed phases)Keeps uncertainty short while making delivery predictableMore decision gates and coordinationLarger, complex initiatives

No pricing model is inherently superior; the right choice depends on how well defined the task is. Demanding a fixed price without a detailed specification is risky for both sides: either the vendor pads the quote heavily with contingency, or every minor clarification turns into a dispute. With time-and-materials billing, ask for regular itemised reports and an agreed budget cap.

Warning signs to watch for

  • A quote for a complex system arrives almost immediately, without a single clarifying question.
  • The proposal never states what is out of scope.
  • Testing, data migration or user training are not mentioned at all.
  • There is no mention of source code ownership or handover deliverables.
  • Only a single lump-sum figure is given, with no breakdown or milestones.
  • The price is drastically lower than competing bids, with no clear explanation of what was left out.

None of these automatically disqualifies a vendor, but each one warrants a direct question. The quality of the answer will tell you at least as much as the proposal itself.

How to compare proposals side by side

  1. Create a comparison grid based on the main points of your brief and mark what each proposal covers.
  2. Ask explicitly about missing items: they may be included, just not spelled out.
  3. Compare assumptions: wherever assumptions diverge, the vendors are solving different problems.
  4. Calculate total cost of ownership: build, rollout and several years of operation.
  5. Meet the actual team who will build your software; working chemistry shows up there first.

Key takeaways

  • A proposal shows how well the developer understands your problem: the price is a consequence of that understanding.
  • A good proposal explicitly states what is out of scope and what assumptions the estimate rests on.
  • Fixed-price contracts fit well-defined scope; time-and-materials fits evolving requirements.
  • Large price differences almost always reflect different scope assumptions: ask vendors to walk through them.
  • True project cost includes the years of operation and maintenance after launch.

Frequently Asked Questions

How many proposals should we request?

Only as many as you can review and discuss thoroughly. Approaching three or four carefully selected developers almost always yields better results than skimming ten superficial quotes.

Why do quotes for the exact same brief vary so widely?

Because vendors make different assumptions about scope, engineering quality and risk contingency. Comparing itemised breakdowns and stated assumptions reveals where the gap comes from.

Should we commission a paid discovery phase before asking for a development quote?

Yes, for complex projects it is strongly recommended. Discovery produces a detailed specification that enables accurate pricing, and you keep that deliverable even if you ultimately choose another development partner.

Facing a similar decision? We are happy to talk it through.

Let's talk