What your monitoring data is actually telling you

Monitoring is one of the more useful things about a modern system, and one of the easiest to misread. The most common error is treating a single excellent day as though it described the year.

A clear, mild day in October with the household away will produce a chart that looks like independence. The same system in June, with everyone home and the heating on, tells a completely different story.

Any figure worth quoting states the period it was measured over and the season it covers. If you see a percentage without one, it is a good day rather than a result.

Generation tells you what the array produced. Self-consumption tells you how much of it you actually used. The second is the one connected to your bill.

A system generating well but exporting most of it at a poor rate is not performing as well as the generation figure suggests. That is usually a timing problem rather than a hardware problem, and timing can often be improved without buying anything.

How deeply and how often the battery cycles affects how long it lasts. A battery going to a very low state of charge every single night is working harder than one cycling gently, and it will show that in its service life.

If your usage has changed since installation, the charge thresholds set at commissioning may no longer suit the property.

The genuinely valuable thing monitoring gives you is drift. Compare this March against last March. A gradual decline in generation on comparable days points at soiling, a failing string, shading from something that has grown or a panel that has stopped contributing.

That is a problem worth catching in month three rather than year three, and it is the main reason we configure monitoring at commissioning.