Common Mistakes in Radio Tenders

Radio procurement goes wrong in a small number of ways, repeatedly, across organisations that have nothing else in common. The errors are recognisable enough to list, and each has a cheap fix at specification stage and an expensive one afterwards.

"120 handheld radios, 15 vehicle radios, 2 repeaters" describes a delivery, not a system. It says nothing about who needs to hear whom, where coverage is required, or what happens in an incident.

Bidders answer what is asked. A device list gets device prices, and the parts that make devices useful arrive as chargeable extras or not at all.

Fix: describe the operation first — teams, sites, areas, simultaneous users, incident behaviour — and let the device counts follow from it.

The commonest and costliest. A system is bought, installed, and then found not to cover the basement, the far warehouse or the lift lobby. The supplier says the delivery met the specification. They are right, because the specification never said where coverage was required or how it would be measured.

Fix: require a coverage survey as part of the works, list coverage areas by name, and write a measurable acceptance criterion with a method and a pass threshold.

Published range figures assume free space, clear line of sight, an ideal antenna and no interference. None of those describe a factory, a hospital or a city.

The same radio delivers fifteen kilometres across open ground and twenty metres from a basement. Range is a property of the path, not of the device.

Fix: use datasheets to compare devices with each other; use a link budget and a measurement to predict coverage.

Specifications that mix technologies — TETRA-grade availability requirements, DMR pricing assumptions, a passing mention of PoC — receive offers that are impossible to compare because each bidder resolved the ambiguity differently.

Fix: choose the technology deliberately, state it, and keep the rest of the document consistent with it. The reasoning is in analogue, DMR, TETRA and PoC compared.

Awarding on price is defensible only when compliance is defined tightly enough that a non-working system cannot be compliant.

When coverage, programming, training, licensing and service are absent from the mandatory requirements, the cheapest offer is precisely the one that omitted them. It wins fairly, and it fails operationally — and the second procurement costs more than the difference ever saved.

Fix: make the operationally essential items mandatory rather than scored, so price competes between offers that would each work.

A radio system is a decade-long relationship. Parts availability, repair turnaround, a local service capability, and someone who can reprogram a fleet when the organisation reorganises — these determine what the system is like to live with.

Fix: require named service capability, stated response times, parts availability for a defined period, and evidence of authorised service status where the manufacturer operates one.

Batteries are consumables with a service life of two to three years under daily cycling. Accessories decide whether a radio is usable at all: a loudspeaker in a plant is not heard, and a radio that is not heard is not coverage.

Tenders that omit them produce a year-two budget request that nobody planned for, and a fleet that users work around.

Fix: itemise batteries — capacity and quantity per radio — and accessories matched to the environment, and state replacement expectations.

"Supply and installation" without definition produces disputes over who mounts antennas, who runs cable, who provides power, who programmes, and who trains.

Fix: name each work item and assign it. Include the channel plan as a deliverable, agreed before delivery rather than after.

Frequency assignment takes real time and is often the critical path. Unassigned, it produces installed equipment that cannot lawfully transmit, with each party reasonably believing it was the other's job.

Fix: name the responsible party and a deadline, and sequence the project so tuning and programming follow the assignment rather than preceding it.

A system sized precisely for today's headcount, on a single frequency, with no spare capacity, hits its ceiling the first time the organisation adds a department or a site — and expansion becomes a new project rather than an increment.

Fix: state the expected growth, require the architecture to accommodate it, and ask each bidder to describe how their proposal scales. Discussed further in scaling a radio system.

Nine of the ten share one root: the tender described a purchase when the requirement was a capability. Radios are a component of a system whose other components are coverage design, licensing, configuration, training and support — and components that are not specified are not delivered.

A specification written around the outcome takes longer, attracts fewer but better offers, and produces a system that works on the day it is handed over.