Ma connexion est lente. C'est moi, le réseau ou un bridage ?

Regarde sur la page du pays la vitesse mesurée là où tu es. Si tu es loin en dessous du chiffre du dixième centile, c'est l'antenne locale et pas ton forfait, et changer de réseau règle souvent le problème.

Nos forfaits au giga n'ont aucun bridage. Seul le pass illimité à la journée en a un, et il est écrit en chiffres sur le bouton d'achat : pleine vitesse jusqu'à 2 GB par jour, puis 1 Mbps.

How to tell a slow cell from a slow plan

Run a speed test and compare it against the median on your country page. Those medians come from 75,333 samples across 145 destinations, so they are a real distribution rather than a marketing figure. If you are near the median, the network is behaving and the site you are using is the problem. If you are far below it, the cell you are on is congested and switching networks usually clears it.

Where our own product is slower, and by how much

Two places, and both are printed. The unlimited day pass runs at full speed to 2 GB a day and then 1 Mbps for the rest of that day, which is a lower daily allowance than Airalo's, who throttle near 3 GB a day. And in 66 of our 145 destinations there is no 5G at all, so the ceiling is whatever 4G gives you there.

What the rest of the category does

Airalo throttles unlimited plans near 3 GB a day and was measured at 10 to 22 Mbps by third party testing. Holafly publishes no fair use number at all and was measured near 7 Mbps before throttling and under 1 Mbps after. Saily bundles ad blocking audited at 28.6 percent data saved, which makes a slow connection feel faster without making it faster. Ours is 2 GB a day at full speed, printed on the buy button.

What full speed means, by seller

SellerFull speed allowanceAfter that
Orbislo metered plansNo throttle at any pointThere is no after
Orbislo day pass at $5.892 GB a day1 Mbps for the rest of the day
Airalo unlimitedNear 3 GB a day1 Mbps, and measured at 10 to 22 Mbps before that
Holafly unlimitedNo fair use number publishedMeasured under 1 Mbps after throttling, near 7 Mbps before

Competitor figures are from third party testing and from their own published pages, checked 2026-08-31. On the daily allowance we are behind Airalo and we print it anyway.

What we do automatically and what needs you

We never throttle a metered plan, at any point, for any reason. We publish the median and the range for each destination rather than a best case. What needs you is the comparison: without a speed test and a country page open beside it, slow is a feeling and not a fault, and neither of us can act on a feeling.

The numbers, and where they came from

Across 145 destinations the fastest median is 119 Mbps and the slowest is 18 Mbps, from 75,333 traveller samples over 157 upstream networks. 66 destinations are 4G only. Those figures are on the coverage page for each country, including the ones that make us look bad.

What you can change, in order of effect

Four things are within your control and they are not equally worth doing. This is the order that actually helps.

  • Move. A cell two streets away is often a different experience, and indoors on a lower floor is usually the worst place you can stand.
  • Reselect the network manually from the partners named on your country page.
  • Turn off the VPN for one test. It is almost always costing you speed.
  • Airplane mode on, ten seconds, off, which forces a fresh registration and sometimes a better cell.

Questions fréquentes

Are metered plans throttled once I use a lot?
No. There is no throttle on a metered plan at any point. The only speed limit in the product is the day pass one, and it is a printed number rather than a policy you discover on day four.
Why is 5G not showing?
In 66 of our 145 destinations there is no 5G on the networks we hold, and each country page says which. Where 5G exists, the phone still needs the right band, which is on our devices pages per model.
Does switching networks actually help?
When you are far below the country median, usually yes, because you are on a congested cell rather than a slow country. In the 5 destinations where we hold one network there is nothing to switch to.
Is a VPN making it slower?
Almost always yes, by a noticeable amount, because your traffic is taking a longer path. Turn it off for one speed test before you conclude anything about the network.
The speed is far below the median and nothing helps. Then what?
That is a network failure in the destination and it is a full refund plus a credit. An agent settles up to $200 without approval, because arguing about it costs us more than the refund does.
Does the time of day matter?
Yes, and more than most people expect. Cells congest in the evening in dense areas, so the same street can be several times slower at nine at night than at nine in the morning.
Is your median speed figure a best case?
No. It is a median across 75,333 traveller samples, so half of the measurements are below it. The fastest destination medians 119 Mbps and the slowest 18 Mbps, and both are published.

Où aller ensuite

Qui a écrit cette page

Priya Raman, Responsable du support

Répond à la file de questions que ces pages ne résolvent pas et réécrit l'article dès que la même question arrive deux fois.

Responsable de

  • Le centre d'aide
  • Les guides d'installation et d'activation
  • Ce qui bloque vraiment les voyageurs

Relu par Jonah Reyes

Comment nous mesurons

Toujours bloqué ?

Ori répond tout de suite et fonctionne sans aucune connexion, grâce à tes données de secours gratuites. S'il ne peut rien faire, un humain prend le relais avec tout ton diagnostic déjà joint, donc tu n'expliques jamais deux fois.