Pre-Project Preparation: The Foundation for Better Electronics and Software

Most overruns in electronics and software projects trace back to decisions that were never actually made: who the product is for, what it must do on day one, and what it is allowed to cost. A discovery phase forces those decisions early, while changing your mind costs a conversation rather than a board respin. This article sets out what that phase should produce and how to tell when it has done its job.

At a glance

Scope
A written list of what the first release must do, and what it deliberately leaves out.
Feasibility
The riskiest technical assumptions tested with small experiments before the design depends on them.
Estimate
Effort and cost given as a range, with the assumptions behind each end stated.
Risk register
Known unknowns ranked by impact, each with an owner and a next step.

The problem

Early conversations about a product are full of agreement that turns out to be shallow. Two people can say “low power” and mean a week on a coin cell or a day on a phone battery. Until requirements are written down and challenged, every estimate is a guess about a different product.

The work at this stage is to replace assumptions with decisions: define the user and the use case, separate requirements from wishes, check that the hard parts are physically and commercially possible, and say plainly what is still unknown.

An estimate is only as good as the requirements behind it. Discovery is where you find out which requirements you actually have.

What good looks like

A product brief: target users, the problem being solved, the competing options and the constraints that cannot move, such as unit cost, size, battery life or a launch date.

A technical specification covering functional and non-functional requirements, candidate architectures and key components, and any certification the target market requires.

A feasibility summary, an estimate expressed as a range, and a risk list, so the decision to proceed is made with the uncertainty visible.

What it changes

Planning Impact

Everyone works from one description of the product, so later disagreements are about trade-offs rather than about what was meant.

Engineering Impact

Architecture and component choices are made against stated requirements and can be defended when they are questioned.

Commercial Impact

Budget and schedule are discussed as ranges with reasons, which makes them easier to plan around and to revisit.

Risk Impact

Problems that would have appeared during prototyping are found when the only thing to change is a document.

Where to go next

A good discovery package converts directly into a plan: an architecture, milestones with acceptance criteria, and a first prototype aimed at the largest remaining risk.

The same approach works for a product that already exists. Reviewing a prototype or a fielded design against its real requirements shows what to keep, what to redesign and in what order.

Problems this prevents

  • Requirements that mean different things to different people
  • Technical assumptions nobody has tested
  • Estimates given without stated assumptions
  • Risks discovered only after the design is committed

Discuss your project

Email support@aavico.com or use Get in Touch when the website finishes loading.