Intelligence

Automated fault monitoring across your charger fleet: a Charger health overview with Charge success and session fault breakdowns, grouped issues, the raw fault log, per charger and per site analysis pages, fault alert recipients, and the email delivery log.

The Intelligence section watches your network for problems so you find them before drivers report them. It combines two detection sources into one picture:

  • AI analysis of failed charging sessions. When a session ends without delivering energy, the platform sends that session's OCPP logs for analysis and stores what it concludes: what happened, what caused it, and what to do about it.
  • Diagnostic events reported by the chargers themselves. Offline, faulted, unstable connection, temperature and power conditions, detected without any session being involved.

Both feed the same views, so there is one place to watch instead of two.

How the section is laid out

Intelligence has four destinations along the top:

DestinationWhat it is for
Charger healthThe main page: the overview plus the Issues, Fault log, Chargers and Locations tabs
Notification settingsWho gets emailed when faults happen, and for which locations
Email deliveryWhether the emails the platform sent on your behalf actually arrived
Zone monitoringA shortcut that opens the external monitoring tool in a new tab, where your account has it

The Charger health page carries four tabs of its own, each covered by its own guide:

  • Issues: repeated faults grouped into one row per problem, with severity and automatic resolution.
  • Fault log: the flat event list, every fault and charger event in the period.
  • Charger analysis: the Chargers tab and the analysis page behind each charger, including the on demand Run analysis tool.
  • Location analysis: the Locations tab and the analysis page behind each site.

The overview described below stays on screen on all four tabs, because it is the answer to "how is my fleet doing" and that question does not change when you switch tabs.

Needs attention: live, not historical

A yellow strip appears above everything else when chargers need looking at right now, reading as one sentence: "12 chargers offline over 30 min · 3 faulted · 2 with an unstable connection". View chargers jumps to your charger list.

Three things worth knowing about it:

  • It counts chargers, not events, so one charger that keeps dropping is one number.
  • It is live, not scoped to the period you have selected. That is exactly why it sits above the period picker: everything below the picker belongs to the selected period, this strip never does.
  • Offline means offline for more than 30 minutes, so a charger rebooting does not raise an alarm. Unstable connection means an unstable connection was detected in the last 24 hours.

When nothing needs attention the strip is not shown at all.

The period picker

The picker defaults to Last 7 days, with Day view, Week view and Month view as alternatives, and arrows to step back to earlier periods. Everything below it, the insight cards, Charge success and the session fault charts, follows that selection. The Chargers and Locations tabs deliberately do not: they are directories, and their rows would be unreadable if they reshuffled every time you changed a chart.

Top insights

AI summary cards describing the most important things happening in your fleet during the selected period, with an Updated timestamp. When a period had no faulty sessions at all, the cards say so plainly instead of inventing something to report.

Charge success

The share of charging sessions in the period that completed without a failure, shown as a large number with a trend arrow compared to the previous period, plus a Healthy pill when the period ran sessions and none of them faulted. Underneath:

  • Analyzed sessions: how many sessions the period had to judge.
  • Successful sessions.
  • Fault rate, with its own trend arrow.
  • Session faults: the number of faulty sessions.
  • Users affected. Users are counted through their app account or charge key, so sessions paid at the card terminal cannot be attributed and the real number may be higher.

A period with no sessions at all is empty, not healthy, and says "No sessions in this period" rather than claiming a perfect score.

This used to be called the Health index. It measures the same thing it always did, the share of sessions that finished without a failure, and the new name says so: it is a score for the charging, not a grade for the hardware.

Session faults

Four views of the same faults, headed "Session faults · N in this period":

  • What happened: the fault types.
  • What caused it: where the fault points, such as charger hardware, the vehicle, or the driver.
  • Where: the locations.
  • Which chargers.

Each list is a top 5 with a footer telling you how many entries there were in total, and each one adds up to the same session fault number in Charge success. When the period had no faults the four collapse into a single band saying so.

Only session faults are counted here. Charger reported diagnostic events appear in the Needs attention strip, in Issues, and in the Fault log, never in these numbers.

Who sees what

  • The whole section requires the Intelligence permission on your admin account. If you cannot see it in the menu, ask your account owner.
  • Notification settings is limited to account owners and operator level admins.
  • Email delivery has its own permission on top of the Intelligence one.
  • Zone monitoring only appears where your account has the external monitoring tool.
  • If you oversee sub operators, Intelligence shows your own network plus theirs, and the CPO filter narrows it to one of them.

Three things called Intelligence

Easy to mix up, completely separate features:

  • Intelligence (this section): fleet wide fault monitoring.
  • Session Intelligence: the AI analysis card on a single session's detail page. See Session Intelligence.
  • Intelligent charging: home charger scheduling that shifts charging to cheaper hours. Nothing to do with faults.

Common questions

Where do I start when something is wrong? If the Needs attention strip is showing, start there, it is what is broken right now. Otherwise open Issues, which is the grouped view of everything worth acting on.

Why does one charger show thirty faults in the Fault log but one row in Issues? That is the point of the two views. The Fault log is every event; Issues groups repeated faults on the same charger into one problem with a count and a trend.

Charge success looks fine but drivers are complaining. Check the Needs attention strip and the Chargers tab. Charge success only counts sessions that were actually started, so a charger that is offline, and therefore never started a session, cannot lower it.

Why is a period showing no data? Either nothing charged in that period, or the period is before January 2026, which is as far back as Intelligence goes.

Where did Diagnostics go? Charger reported diagnostic events now feed the same Issues and Fault log views as the AI session analysis, so there is one place to watch instead of two.


Did this page help you?