How to Find Good WiFi Anywhere: A Digital Nomad WiFi Checklist

You have booked a cafe corner, opened your laptop, and pulled up a client call in six minutes. The wifi looked fine on the amenity list. Thirty seconds into the call your video freezes, the audio cuts into robot syllables, and you are apologizing while frantically switching to your phone's hotspot. This has happened to almost everyone who has worked remotely for more than a month, and it keeps happening because most nomads are checking the wrong number.

"Fast wifi" on a listing or a cafe's Google review means almost nothing on its own. A connection can post 150 Mbps download and still choke a video call if the upload is 4 Mbps and three other tables are on a Teams meeting at the same time. The number that breaks your call is rarely the one anyone advertises.

This is the checklist we actually use: what to test before you book, what to test in the first five minutes at a cafe, what backup connection to carry, and which numbers matter more than raw speed. No vague "check the wifi first" advice. Real thresholds, real tools, real backup gear.

The Frame: Stability Beats Speed

Every wifi problem a nomad runs into traces back to one of three things: not enough upload, too much jitter, or too many people on the same contended connection at once. Raw download speed is rarely the actual cause, which is why chasing a bigger number on a Speedtest doesn't fix the problem.

A useful way to think about it, borrowed from network engineers who work on remote-access infrastructure: a 25 Mbps connection with 20ms ping and 5ms jitter runs a video call flawlessly. A 50 Mbps connection with 150ms ping and 40ms jitter breaks audio and video constantly. Speed is the headline number. Stability is what your call actually depends on.

That reframes what you should be testing. Instead of asking "is this fast," ask three narrower questions: is there enough upload for what you're doing, is the connection stable enough that jitter won't cause dropouts, and is it going to hold up once the room fills up at 1pm. The rest of this checklist is built around answering those three questions quickly, wherever you are.

The Speed Thresholds That Actually Matter

Zoom and Google both publish minimum bandwidth requirements, and they are lower than most people assume, because "minimum" and "comfortable" are different numbers. Treat the official minimums as the floor you never want to test against, and the practical numbers below as the range that actually holds up.

| Activity | Official minimum | Comfortable real-world target | Why | |---|---|---|---| | 1:1 video call (720p) | 1.2 Mbps up/down (Zoom) | 5-10 Mbps up/down | Leaves headroom for background sync, browser tabs | | Group call with screen share | 1.8-3.0 Mbps (Zoom, 720p) | 3-4 Mbps upload dedicated to the call | Screen share adds a second video stream on top of your face | | Loom or async video recording | 5 Mbps upload (Loom's stated floor) | 10 Mbps upload | Loom streams to the cloud live as you record; a dip mid-recording breaks sync | | Large file / video export upload | Not applicable | 15-25+ Mbps upload | Determines whether an upload takes 5 minutes or 2 hours | | Async work (Slack, browser, git) | Not applicable | 5-10 Mbps download | Bursty, low-bandwidth, rarely the bottleneck |

Two numbers matter more than anything in that table: jitter under 30ms and ping under 150ms, ideally closer to 50ms for HD video. Jitter above 30ms is what causes the choppy, robotic audio that makes people think their microphone is broken when the real problem is the network. If a connection tests fast on download but a call still stutters, check jitter and ping before you blame the microphone or the platform.

Vetting Wifi Before You Book

Airbnb has run a host-side wifi speed test since 2021, and it still shapes what you see on a listing in 2026. If a host has run the test and the result clears 50 Mbps download, the listing badge upgrades from a plain "Wifi" tag to "Fast wifi." That is a fast visual filter when you are scanning ten listings at once.

The catch: that badge only reflects download speed. No upload number, no jitter, no latency. A listing can show "Fast wifi" and still be useless for a client call if the upload is capped at 5 Mbps on a shared building connection, which is common in older apartment blocks across Southern Europe and Latin America.

Airbnb's own messaging system blocks photo and link attachments before a booking is confirmed, so a host cannot send you a Speedtest screenshot directly. Two workarounds that actually work: ask the host to type the download and upload numbers into the message thread directly, or check whether they have already uploaded a Speedtest screenshot as one of the listing photos. Hosts competing for remote-worker bookings increasingly do this on their own, unprompted, because they know it converts.

One more cross-check worth the thirty seconds: compare the host's claimed number against the city-level median. Nomads.com, the rebrand of the old Nomad List, tracks internet-speed data across more than 24,000 cities and blends multiple data points into a city median. If a host claims 300 Mbps in a neighborhood where the city median sits at 80 Mbps, that is a flag to dig further, not a reason to panic.

The 5-Minute Cafe Test

You will not always have the luxury of pre-booking. Sometimes you land in a new neighborhood and need a place to work in the next hour. Run this before you order a second coffee and settle in for a four-hour block.

1. Connect to the actual guest wifi, not your phone's cellular data, and run a Speedtest.net or Fast.com check. Testing on cellular and assuming the cafe wifi is similar is the single most common mistake nomads make here.
2. Compare against a baseline. If you tested your accommodation's wifi earlier that day, you already have a reference point. A cafe testing at half your apartment's speed is not automatically bad, but it tells you what ceiling you are working with.
3. Check the time of day against your plan. A cafe that tests at 40 Mbps at 8am can drop hard by 1pm once the lunch rush fills every table and everyone's phone joins the same access point. If you need a stable 2pm client call, ask if that's typically a busy window before you commit to staying.
4. Ask staff two questions: does the wifi password or network change daily, and have they had recent outages. Informal, but staff usually know if the connection has been flaky that week.
5. Count outlets relative to seating before you pick a table. Wifi that holds up for four hours does you no good if your laptop dies in 90 minutes and the nearest outlet is across the room.
6. Run the test again after 20 minutes if you're staying for a call-heavy session. A network that looked fine on arrival can already be filling up.

This is the same instinct that separates a coworking day pass from a coffee-shop gamble: you are testing the network under the conditions you will actually use it, not the conditions it advertises. We go deeper on this tradeoff, including real upload numbers by neighborhood, in our Lisbon coworking breakdown, where the best-documented space measured 192 Mbps upload against a citywide residential norm closer to 20-40 Mbps.

Backup Connectivity: Pick by Trip Length, Not Habit

Every experienced nomad eventually gets burned by relying on a single connection. The fix isn't buying the most expensive gear, it's matching your backup to how long you're staying and how much a dropped call actually costs you.

| Option | Best for | Rough cost | Tradeoff | |---|---|---|---| | eSIM (Airalo, Nomad, Ubigi) | Trips under 2-3 weeks, occasional backup | $4-30 for 1-10GB regional plans | Cheapest entry, but data-capped and coverage varies by carrier partnership | | Holafly unlimited eSIM | Stays where you'll lean on mobile data daily | ~€25-69 per 7-30 days, or €59.95/month unlimited | Priced by days not GB, easy to budget, pricier per GB if you barely use it | | Local physical SIM | Stays over 2-3 weeks | Varies, usually cheapest per GB | Requires an unlocked phone and a shop visit; best per-GB value at length | | Portable hotspot (Skyroam Solis, GlocalMe, Nighthawk M6) | Groups, multi-device households, frequent movers | $300-450 hardware plus data plans | One device covers everyone, but it's another thing to charge and carry | | Starlink Mini | Remote or infrastructure-poor locations | $199 hardware, $50-165/month, $5/month standby-only | Genuine backup independent of local infrastructure, needs sky view and setup time | | Speedify (bonding software) | Anyone who can't afford a dropped call, ever | Free up to 2GB, $14.99/month standard | Bonds wifi and cellular simultaneously so a drop on one doesn't drop the call, but it's another subscription |

If you are moving through a country every one to two weeks, an eSIM is the right default: cheap, instant activation, no shop visit. If you are settling into one city for a month or more, a local SIM usually beats any eSIM on cost per gigabyte. If you are working from somewhere genuinely remote, like a coliving space outside a major hub, Starlink Mini's $5/month standby mode is a new-in-2026 option specifically built for exactly this: a low-cost always-on fallback that isn't your primary connection, just insurance.

Speedify deserves a specific mention because it solves a different problem than the others. Instead of switching to a backup after your primary connection fails, it runs your wifi and a phone hotspot at the same time and fails over in milliseconds if one drops. For a call you genuinely cannot afford to lose, that's a meaningfully different guarantee than "keep an eSIM in your back pocket."

Tools Worth Keeping on Your Phone

You don't need ten apps. Four cover almost every situation:

• Speedtest.net (Ookla): the default for a quick download/upload/ping check, and the tool most ISPs and coworking staff expect you to reference if you're asking them to fix something.
• WiFi Map: crowdsourced hotspot data across 200+ countries, now the dominant app in this category. Useful for finding a password before you arrive somewhere, but check the "last verified" date on an entry before relying on it. Passwords and networks go stale, especially outside major cities.
• Nomads.com: for planning ahead, not real-time testing. City-level median speeds help you sanity-check a host's claim or decide whether a destination is viable for call-heavy work before you book flights.
• Google Maps, searched for "wifi" within reviews: free, already on your phone, and often the fastest way to confirm whether a specific cafe's connection has been a recurring complaint.

What Breaks Wifi That Isn't the Wifi

A few failure modes get blamed on "bad wifi" when the actual cause is something else entirely, and knowing the difference saves you from switching cafes when the fix was in your own setup.

VPN overhead. A quality paid VPN costs you roughly 10-20% of your speed as encryption overhead, which is normal and usually invisible. A free VPN can cost you 50-80%, which is not invisible at all. If a connection tests fine with the VPN off and falls apart with it on, that's your answer, not the cafe's router.

Contended bandwidth at peak hours. The same network that tested at 60 Mbps at 9am can be unusable by 1pm purely from concurrent users, independent of what the connection is actually capable of. This is the most common complaint we hear about beachfront cafes in popular nomad hubs: great at 8am, unworkable by lunch.

Captive portals on hotel wifi. Some hotels force you to re-authenticate every few hours, and a few intentionally throttle the free tier to upsell a paid "premium" wifi package. Front desk staff usually can't do much about either. If a hotel's free wifi keeps logging you out mid-call, ask directly whether a paid tier exists before assuming the connection itself is the problem.

Old building materials. Stone walls over 60cm thick, common in older European apartment stock, and plaster-and-lath construction with chicken wire behind it, common in Victorian-era buildings, both block signal meaningfully. If the router is one room away and your laptop shows one bar, this is often why. As a guest you usually can't install a mesh extender, so this is a real pre-booking risk to check for in reviews, not just a fix-it-yourself problem.

We ran into a version of this exact issue when comparing extended-stay options for a 90-day Lisbon trip. The cost breakdown, including which accommodation types make wifi verification easiest before you commit, is in our coliving vs hotel vs Airbnb guide.

Building This Into How You Choose a City, Not Just a Cafe

Wifi reliability isn't just a day-to-day annoyance, it's one of the criteria that should shape where you base yourself in the first place, alongside cost, visa runway, and timezone overlap. We built wifi into the 6-criterion scoring framework in our guide to choosing a city as a digital nomad, and it's worth treating your own city research the same way: don't just trust a headline "fast internet" claim, check the median, and check it against the specific neighborhood you're planning to live in. Roma Norte's cafe wifi in Mexico City, for instance, tests reliably around 40-85 Mbps off-peak but drops noticeably during the 11am-2pm rush, a pattern that's easy to miss if you only test once, in the morning, before you sign a lease.

None of this requires new gear or a subscription you don't already need. It requires testing the actual numbers instead of trusting the label, and carrying one backup option that matches how long you're staying. Do that once, before your next call-heavy week, and you stop finding out your wifi is bad in the middle of a client call.

Check wifi speed scores, noise ratings, and outlet counts for 200+ coworking spaces and cafes before you book your next stay at RemoteZone.