How to Write Sequences That Can Actually Be Tested

By Andy Austin | August Bridge Advisory

A sequence that cannot be tested does not define behavior. It invites interpretation.

Many HVAC sequences look complete.

They are long.
They are detailed.
They sound technical.

But they fail in one critical way:

They cannot be tested.

The difference between intent and behavior

Typical sequences describe intent:

maintain comfort
optimize energy
respond appropriately

These statements are directionally correct.

But they are not testable.

They do not tell:

what triggers the system
what the system will do
how to verify that it did it correctly

That gap is where problems begin.

Start with modes, not narratives

The most effective way to improve a sequence is to break it into operating modes:

occupied cooling
heating
economizer
unoccupied
safeties and alarms

For each mode, define:

when it starts
when it ends
what the system does

This structure makes the sequence understandable—and testable.

Anchor behavior to real points

Every important action should tie to a measurable point:

temperature sensors
airflow measurements
static pressure
damper positions
valve commands

Without named points:

Teams are no longer verifying the same system.

They are interpreting it.

Write logic so it can be observed

A strong sequence reads like a test procedure.

It should allow someone to:

force a condition
observe a response
confirm an outcome

That means writing logic clearly:

what happens first
what happens next
what conditions must be met
what defines success

This is what turns a sequence into a verification tool.

Address priorities and conflicts

Real systems operate under competing conditions:

safety vs comfort
protection vs efficiency
manual override vs automation

If priorities are not defined:

The system behavior becomes unpredictable.

The sequence must define:

what takes precedence
how conflicts are resolved
how the system recovers

What this changes

When sequences are written for verification:

controls contractors implement more consistently
commissioning becomes faster and clearer
fewer RFIs occur
performance becomes defensible

When they are not:

each party fills in gaps differently
system behavior varies
testing becomes subjective

Common pattern

Most problematic sequences share the same traits:

long narrative format
vague triggers
no defined modes
no measurable outcomes

They appear complete.

But they do not define the system in a way that can be proven.

Write sequences that can be proven

If a sequence cannot be clearly tested in the field, it will be interpreted—and interpretation is where performance breaks down.

August Bridge helps A/E teams develop sequences that are clear, testable, and aligned with real-world commissioning.