Networks end up multi-brand through acquisition, procurement cycles and hedging against supplier risk. It is a reasonable position and it costs more to operate than a single-brand fleet, in ways that are not obvious at procurement.
The costs are concentrated in a few places, and most are addressable by normalising rather than standardising.
Where brands actually differ
Note that all of these are within the protocol. None requires a vendor to be non-compliant, which is why a multi-brand fleet is harder to operate even when everything works as specified.
- Error code mapping, so the same code means different things.
- Configuration keys supported, and how strictly names are matched.
- Status transition behaviour, particularly around suspended states.
- Firmware update mechanisms and what they report.
- Offline behaviour and how stored transactions are delivered.
- Diagnostic depth available remotely.
Normalise rather than standardise
You cannot make the brands behave identically. You can map each one's behaviour onto a common internal vocabulary, so operations sees one taxonomy rather than several.
That mapping is real work and it is the difference between a fleet that can be operated as one thing and several fleets sharing a dashboard.
Keep the raw alongside the normalised
Normalisation loses detail, and the lost detail is frequently what a technician needs. Retaining the original code and vendor-specific information alongside the mapped category preserves both the fleet view and the actionable specifics.
This costs storage and saves investigations.
Spares and skills multiply
Each brand carries its own spares, its own service documentation and its own quirks that engineers learn individually. Three brands is roughly three times the parts holding and three separate bodies of knowledge.
This is the cost most often omitted from procurement comparisons, and at fleet scale it can exceed the unit price difference that motivated the second brand.
Firmware management gets harder
Each brand has its own release cadence, its own update mechanism and its own version scheme. Keeping a multi-brand fleet converged means running several rollout processes rather than one.
Drift accumulates faster as a result, which compounds the diagnostic difficulty that motivated normalising in the first place.
Decide deliberately how many brands
Two brands is a hedge against supplier risk. Five is an operational burden that rarely pays for itself, and it usually arrives through accumulation rather than decision.
Setting a limit, and treating adding a brand as a decision with an operational cost attached, is what prevents a fleet becoming unmanageable one procurement at a time.