Answer guide · English

By Symbioen · Published · Reviewed

How reliable are public EV chargers in Sweden?

In Q2 2026, Swedish public chargers were in a working status 95.57% of the time we observed them. Of the charging sessions we inferred from status changes, 87.4% were successful. The two figures measure different things: a charger can show as available and still fail to start a charge.

Period
30 March–5 July 2026
Source
Public National Access Point status history
Base unit
EVSE (individual charging point)
Method
Symbioen Index v0.5

Key Q2 2026 results for Sweden

Observed inventory
9,693 stations · 55,976 charging points
Public stations listed at the end of the quarter
Stations with status data
8,611 of 9,693 stations
Sent at least one status update during the quarter
Evidence coverage
6,950 of 9,693 stations
Enough status history to be scored in the last full week, 22–28 June
Public EV charger availability
95.57%
Share of observed time in a working status
Successful charging sessions
87.4%
Share of inferred sessions counted as successful
Charging sessions
3,684,528
Inferred from status changes; an undercount because of recording gaps
Symbioen Index
90.00 / 100
Composite reliability score, quarter average

What the two main figures measure

Availability measures time. For each charging point we could see, we count the share of observed time it reported a working status. Successful charging sessions measure outcomes. We infer each session from the charger’s status changes and count it as successful if charging lasted at least two minutes.

The two figures have different denominators, so subtracting one from the other gives a meaningless number. Note also how low the bar for success is. A session that stopped after five minutes still counts, so the figure shows that charging started, not that the driver got the charge they wanted.

How much of the network we could measure

Sweden listed 9,693 public stations at the end of the quarter. Of these, 8,611 sent at least one status update during Q2 and 6,950 had enough history to be scored in the last full week. The other 1,082 sent no status data at all.

A station without a score is not a broken station. We could not see it well enough to judge, so it is left out of every figure on this page. We do not know whether the unscored stations perform better or worse than the ones we scored.

What changed during the quarter

The network grew fast. Sweden had 8,599 stations at the start of the quarter and 9,693 at the end, 12.7% more. Service stayed roughly level: availability was 95.88% in the first full week and 95.36% in the last, and the weekly share of successful sessions stayed close to 87–88% in most weeks.

Recording gaps affected five weeks between 11 May and 21 June. Weekly values for those weeks are estimates, and the session total in the table is lower than the true number. The full report also compares Q2 with Q1. Q1 was calculated with an earlier version of our method, so part of any change between the quarters comes from the method rather than the chargers.

How to use these figures

Use them to compare operators, municipalities and stations on the same definitions. They show where to look, not what went wrong. The station and charging-point history underneath explains the rest.

The most useful signal is a mismatch. A station that shows as available most of the time but has a low share of successful sessions deserves a closer look, especially if it is busy.

What the public data cannot show

Public status data does not show how much energy was delivered, whether payment worked or what each driver experienced. We can check charging outcomes and energy delivered per session from the operator’s session records through Operator Performance Verification. Fleet and vehicle data shared with consent can add what drivers experienced.

Evidence and method

Sources for this guide