How to Define a Monitoring Goal
Before You Buy Anything

This isn't a complicated process — but it does require slowing down before you spend anything. The questions below will help you arrive at a goal specific enough to drive the right hardware and sensor decisions.

Start with the problem, not the solution

Most people come to building monitoring because something prompted them — an unexpectedly high utility bill, a compliance requirement, a retrofit they need to verify, a system they suspect isn't performing. That prompt is your starting point.

Before thinking about hardware, be specific about the problem. Not "I need a monitoring system" — that's a solution. Something more like: "I don't know why our hot water energy costs jumped 30% last quarter" or "I need to prove to my client that the new HVAC system is delivering the promised savings."

The more specific the problem statement, the more clearly the monitoring requirements follow from it. Vague problems produce vague monitoring systems that don't quite answer anything.

Turn the problem into a question

A well-defined monitoring goal is really just a question your data will answer. Converting your problem statement into a specific question is a useful discipline — because it forces you to think about what a satisfactory answer actually looks like.

For example:

Problem: Hot water energy costs are higher than expected.
Question: Is the recirculation pump running continuously when it should be cycling, and what is the actual supply and return temperature differential across the system?

Problem: Need to verify retrofit savings for a client.
Question: What is the measured energy consumption before and after the retrofit, at 15-minute intervals, over a 12-month period?

Problem: Concerned about Legionella risk in a healthcare facility.
Question: Are all hot water distribution points maintaining temperature above 60°C, and how long does it take for temperature to recover after draw-off events?

Notice how each question implies specific measurements — and therefore specific sensors and data resolution requirements. That's the point. The question is what connects the problem to the hardware.

Ask yourself these questions before you buy

• 1 What specifically do I need to measure? Temperature, flow, power, pressure, equipment status? At how many points? Be as specific as you can — "hot water system" is not specific enough; "supply temperature at the tank outlet, return temperature at the pump inlet, and pump runtime" is.
• 2 How often does the data need to be captured? Monthly billing totals, daily summaries, or second-by-second time series? The answer depends on the nature of the problem — slow drift requires less resolution than a fast-cycling fault.
• 3 Who needs access to the data, and in what form? Is this for your own internal analysis, for a client report, for regulatory compliance, or for a live dashboard someone checks daily? The answer affects how data needs to be stored, formatted, and delivered.
• 4 How long does monitoring need to run? A short-term diagnostic project has different requirements from a permanent installation. Duration affects storage, connectivity, and power supply decisions.
• 5 What does success look like? How will you know the monitoring project has answered the question? Define this before you start — it's surprisingly easy to collect a lot of data and still not have a clear answer if you haven't decided in advance what you're looking for.

One goal or many?

Most buildings have more than one thing worth monitoring. The temptation is to instrument everything at once — and while that's not necessarily wrong, it can lead to a system that's complex to deploy, harder to interpret, and more expensive than necessary.

A better approach for most first-time deployments is to start with the one question you most need answered, get the system working well, and expand from there. BluSENSE hardware is designed to scale — a single BluNODE can handle up to 40 temperature inputs and 36 Modbus sensors simultaneously, so there's no penalty for adding measurement points as your needs grow.

Start focused. Expand with purpose.

When your goal is unclear

Sometimes the honest answer is: "I know something is wrong but I don't know exactly what." That's a legitimate starting point, and monitoring can help — but it needs to be framed differently. Rather than a single targeted question, you're doing exploratory monitoring: instrumenting a system broadly enough to see what patterns emerge.

This is valid, but it requires more sensors, more data, and more time to interpret. If that's your situation, it's worth talking through your system with someone who can help you instrument it sensibly rather than comprehensively. More data isn't always better data — especially when you're trying to find a signal in a lot of noise.

With a clear goal in hand, the next step is translating it into the right sensors for the job. That's covered in How to Choose the Right Sensors for Your Application →